Wikipedia talk:WikiProject Mathematics/Archive/2010/Jan

Error in http://en.wikipedia.org/wiki/Plane_(geometry)?

This is my first attempt to participate in Wikipedia, so please let me know if I get something wrong. I wasn't brave enough to edit the page directly.

I believe that I spotted an error in the Plane(geometry) page.

In the section
Define a plane with a point and two vectors lying on it

the equation
r = r0 + sv + tw

should be
r = r0 + s(v - r0) +t(w - r0) --Rtgjr (talk) 20:52, 26 December 2009 (UTC)

The place for this is the article's talk page. The article is correct. I'll reply to the user's own talk page. ~~ Dr Dec (Talk) ~~ 21:47, 26 December 2009 (UTC)

Do not worry about making mistakes; everyone makes a mistake at some point. Even in the case that one does make a mistake in an article, it is often easily fixed. Your contributions are always appreciated. --PST 12:54, 27 December 2009 (UTC)

Unitary space

I just replaced a (somewhat misleading) redirect from Unitary space to Hilbert space with a (very short) stub. Can someone review the article (err... have a look at the two sentences), and possibly add some more details to it?  Cs32en  05:21, 27 December 2009 (UTC)

A redirect to inner product space is more appropriate I think (if indeed unitary spaces are not complete by definition—someone needs to look into this). Also, a norm is already defined on every inner product space, so "inner product space on which a norm is defined" is a distinction that does not exist anywhere in the literature. This is like saying "a Banach space with a metric" or "a metric space with a topology". Having an inner product is a strictly stronger condition than having a norm. We have already had a similar discussion before at Talk:Hilbert space#Pre-Hilbert space many months ago. Sławomir Biały (talk) 12:55, 27 December 2009 (UTC)

An inner product space is a space on which a scalar product is defined. The existence of a scalar product does not imply the existence of a norm, so having an inner product is not a strictly stronger condition than having a norm.

12.4.1.2 Unitary spaces and their properties
In a pre-Hilbert space, one can, with the help of the scalar product, create a norm [...].
(Bronstein et al., Taschenbuch der Mathematik, 7. ed., 2008, p. 678)

Why didn't you wait for some more input in this discussion before editing the article? I don't think accusing me of original research when I was explicitly referring to a source is appropriate.  Cs32en  19:32, 27 December 2009 (UTC)

An inner product trivially leads to a norm. If the inner product is denoted as <x,y> then define norm(x)=sqrt(<x,x>). But there are normed spaces where no inner product can be defined. I think that is what User:Sławomir Biały meant by a strictly stronger condition. Your statement "The existence of a scalar product does not imply the existence of a norm" is just wrong, assuming you use the phrase "scalar product" to mean "inner product". --Robin (talk) 19:50, 27 December 2009 (UTC)

What if you have a pre-Hilbert space on $\mathbb {Q}$? Then a norm such as ${\sqrt {<x,x>}}$ does not exist in $\mathbb {Q}$.  Cs32en  20:32, 27 December 2009 (UTC)

Well, WP:OR often occurs when the source doesn't support the statement being made. Since we've been over this already, I have taken the rhetoric up a notch. Also, waiting until more discussion before making an edit is not mandatory (refer to WP:BRD), particularly when (unless I am very mistaken) this one is likely to be very uncontroversial here. Otherwise, you are of course free to find a reliable source that unequivocally supports the statement that "The existence of [an inner product] does not imply the existence of a norm." Sławomir Biały (talk) 21:06, 27 December 2009 (UTC)

Is there a reliable source that unequivocally says that an inner product does imply the existence of a norm for all vector spaces (i.e. not only for vector spaces defined, for example, on the real numbers)?  Cs32en  21:24, 27 December 2009 (UTC)

In every linear algebra book I've read, inner product spaces are defined only over R and C. In fact, see the Mathworld article which says "An inner product on a vector space V over a field K (which must be either the field R of real numbers or the field C of complex numbers)" [1]. The article also clearly states that "Every inner product space is a normed vector space". --Robin (talk) 21:33, 27 December 2009 (UTC)

James and James, Mathematics Dictionary, 5. ed., 1992, p. 219, say: "inner-product space. A vector space V on which there is a function (called a the inner product or scalar product) whose domain is in the set of ordered pairs of V, whose range is in the set of scalars (real or complex numbers) [...]. If a norm is defined by $\lVert x\rVert =(x,x)^{1/2}$, then the space becomes a normed vector space."

While not unequivocal, James and James thus imply ("If a norm is defined") that such a norm might not be defined in any inner product space. On the other hand, they state that the inner product is in the range of real or complex numbers, which might be understood as implying that the vector space must be defined on either $\mathbb {R}$ or $\mathbb {C}$ as well. (Note, however, that if the range of a scalar product is in $\mathbb {Q}$, then it would also be in $\mathbb {R}$ and $\mathbb {C}$.)  Cs32en  21:57, 27 December 2009 (UTC)

As far as I can tell, no one else is talking about inner product spaces over fields besides R or C, even the very sources that you have cited. An inner product is a special case of the more general notion of bilinear form, which is defined over more general fields. We already have appropriate articles in agreement with uncontroversial mathematical usage. Also, I read the "if" in the James and James reference as a "let" as is typical in English mathematical writing, so I do not accept this as a counterexample, and certainly not the kind of unequivocal statement that would be needed. Sławomir Biały (talk) 22:20, 27 December 2009 (UTC)

Let us assume that "if" actually means "let" in that source. How do you explain why the authors chose to use the language "becomes a normed vector space"? What was the inner product space before becoming a normed vector space? Bronstein et al., p. 678, say: "A vector space defined on the field $\mathbb {K}$ (in most cases $\mathbb {C}$ is being considered) is being called [...] an inner product space [if an inner product exists]." A field $\mathbb {K}$ does not need to be either $\mathbb {R}$ or $\mathbb {C}$.  Cs32en  23:02, 27 December 2009 (UTC)

Also, if every inner product space would be defined on either $\mathbb {R}$ or $\mathbb {C}$, then every inner product space would be complete, and thus, if we combine this with the assumption that every inner-product space would be a normed vector space with the norm given by ${\sqrt {<x,x>}}$, i.e. a unitary space, every inner product space would be a Hilbert space.  Cs32en  22:08, 27 December 2009 (UTC)

I must disagree. For instance, the set of all complex polynomials on [−1,1] equipped with the inner product $\langle f,g\rangle =\int _{-1}^{1}f(t){\overline {g(t)}}\,dt$ is an inner product space (with complex inner product) that is not complete. The completion is L²[−1,1]. It might be helpful to digest the distinction between a Hilbert space and a pre-Hilbert space before attempting to create new articles on the subject. Sławomir Biały (talk) 22:20, 27 December 2009 (UTC)

Of course, if you define an inner product space on a set $D_{1}\subset \mathbb {R}$ or a set $D_{2}\subset \mathbb {C}$, then such an inner product space does not need to be complete.  Cs32en  23:02, 27 December 2009 (UTC)

Such a "space" would not even be a vector space over R or C, so obviously not an inner product space. Sławomir Biały (talk) 23:08, 27 December 2009 (UTC)

Just to repeat, the difference between a Hilbert space and any old inner product space is the completeness condition. It has nothing to do with norms. Both spaces are normed spaces. --Robin (talk) 23:25, 27 December 2009 (UTC)

Indeed. And as WP:STICK is becoming increasingly relevant in this thread, here is another question for Cs32en: Does Cs believe that every inner product space over the real or complex numbers is complete (with the induced norm), and therefore is a Hilbert space? Presumably this is what the above is intended to mean: "Also, if every inner product space would be defined on either $\mathbb {R}$ or $\mathbb {C}$, then every inner product space would be complete." Sławomir Biały (talk) 00:12, 28 December 2009 (UTC)

$\mathbb {R} ^{n}$ and $\mathbb {C} ^{n}$ are Hilbert spaces.  Cs32en  01:09, 28 December 2009 (UTC)

You didn't answer my question. Sławomir Biały (talk) 01:16, 28 December 2009 (UTC)

You misrepresented my statements and came up with a question that is poorly defined (what kind of norm are we talking about, to begin with). I will not answer this question and thereby open up further possibilities for the continuation of such a type of discussion. I already said that I am not going to discuss this further (see below).  Cs32en  01:27, 28 December 2009 (UTC)

(@ Slawomir): For a vector space defined on $\mathbb {Q} \subset \mathbb {R}$, the vector space axioms are true. Kaye and Wilson, Linear Algebra, 1992, even give the example of a vector space based on $\{0\}$.  Cs32en  23:37, 27 December 2009 (UTC)

Let me respond here with a question: Do you believe that Q, the field of rational numbers, is a vector space over the reals or complex numbers? Sławomir Biały (talk) 23:58, 27 December 2009 (UTC)

I'm not sure how this helps to solve the problem. $\mathbb {Q}$ is not a vector space over $\mathbb {R}$, because there is no vector $kv\in \mathbb {Q} \;\forall \;k\in \mathbb {R} ,v\in \mathbb {Q}$.  Cs32en  00:26, 28 December 2009 (UTC)

(unindent) Just to recap. Cs32en claimed that any inner product space over R or C is complete. When it was pointed out that this was not the case, Cs32en then said that subsets of R and C can be given an "inner product". To this, I then replied that subsets of R and C (excluding of course the zero subspace) are not vector spaces over R or C (and [[a fortiori] are not inner product spaces). To this, Cs responded that Q is a vector space over R or C, and claimed to give a reference in support of this statement. Then, when pressed, Cs agreed that Q is not a vector space over R or C.

This is too much. Please, Cs32en, give a clear statement of what it is that you believe to be true, supported by unambiguous references that do not require us to psychoanalyze the usage of particular words (like "if", "can be given", etc.) Until this time, I shall observe WP:STICK and leave this discussion because it is just as unproductive (and indeed getting to be much longer and more absurd) than even the one at Talk:Hilbert space that covered the same ground. Sławomir Biały (talk) 00:41, 28 December 2009 (UTC)

I have never stated that $\mathbb {Q}$ would be a vector space over $\mathbb {R}$ or $\mathbb {C}$. Your summary of the discussion is highly misleading. I will not go into the details here. I do not appreciate your patronizing language, and I will therefore discontinue this discussion at this point.  Cs32en  01:09, 28 December 2009 (UTC)

Ultrametric norms

Just to confuse things again: the following paper discusses p-adic Hilbert spaces and p-adic inner products: Operator Calculus For p-Adic Valued Symbols And Quantization, S. Albeverio, R. Cianci and A.Yu. Khrennikov, Rend. Sem. Mat. Univ. Pol. Torino, Vol. 67, 2 (2009), 137 – 150, Second Conf. Pseudo-Differential Operators. This doesn't change the fact that real or complex inner product spaces induce a norm, or that non-complete spaces can have an inner product, or that normed spaces need not have an inner product. Charvest (talk) 22:48, 27 December 2009 (UTC)

Strangely we seem to lack a proper article on ultrametric norms. This is a serious omission. Sławomir Biały (talk) 01:42, 28 December 2009 (UTC)

Ultrametric space seems to be all we have. Algebraist 02:03, 28 December 2009 (UTC)

Need additional eyes on Ratio

I've been trying to improve the Ratio article but I'm having a weird argument with another editor who insists on putting incorrect information in it. Anyone care to chime in on the talk page?--RDBury (talk) 00:43, 28 December 2009 (UTC)

I support your view on what a ratio is, and I haven't see any other definition in the literature. I've looked at the talk page, and there are comments from other people as well who interpret a ratio to represent ${\frac {f(x)}{g(y)}}$ or ${\frac {f(x,y)}{g(x,y)}}$, rather than ${\frac {x}{y}}$.  Cs32en  02:36, 28 December 2009 (UTC)

Copy, paste, add one

We now have articles for Seventh dimension, Eighth dimension and Ninth dimension. Some other dimension articles have been created that were formerly redirects. This seems to be part of a general trend of copy, paste, and add one type articles in WPM. In general, I think a very narrow interpretation of the WP:GNG should be used with these. Multiple articles with nearly the same content are not encyclopedic, not useful and generally look bad. A big problem is that people see articles like Fifth dimension and Sixth dimension and, using the WP:Other stuff exists argument, automatically assume that we need to keep adding articles in the series until Wikipedia runs out of server space. The GNG is supposed to prevent this kind of thing but some of the mathematical articles that already exist set a bad example. There are plenty of notable subjects at Wikipedia:Missing science topics that need articles, and we shouldn't be wasting time and effort writing and maintaining articles simply because there is a series of similar articles.--RDBury (talk) 14:44, 26 December 2009 (UTC)

Sorry to intrude by cutting apart discussion like this, but it's only the lead section that is always nearly the same. Nine-dimensional space is the weakest of the lot, certainly, as it isn't used in scientific theories...but all the rest are! 4 = 2 + 2 04:18, 29 December 2009 (UTC)

I noticed this. They are copy and paste jobs from a two day old snapshot of my re-write of sixth dimension, except with random non-notable, often non-scientific additions to them, numerous problems with copying and pasting (e.g. not changing a 6 to another digit in a few places). As for notability I agree: after coming up with a handful of things for sixth dimension I can't think of anything as useful for the others. The only one that I would be happy with the Seven-dimensional cross product, but it already has its own article. --JohnBlackburne (talk) 20:40, 26 December 2009 (UTC)

Another example: Hexatope numbers, C&P of Pentatope number. This just came up in current activity as being PRODed. But it doesn't seem to meet Wikipedia:Criteria for speedy deletion so I doubt the PROD will be accepted, so that means an AfD battle.--RDBury (talk) 11:57, 27 December 2009 (UTC)

Grumble. It's worse than I thought. For most values of m, almost all of mth dimension' should be in m-dimensional space, (with almost all of the rest unsourced nonsense), and almost all of that should be already in n-dimensional space. I've proposed deletion of 7-9, changed 1-3 back to redirects, and commented on 5-6. 4th dimension.

I've also proposed deletion of sextic equation and septic equation. It's likely 4 will remove the prods, but an AfD is likely in order. — Arthur Rubin (talk) 09:03, 28 December 2009 (UTC)

Um...nonsense? I was trying to add some useful content to the articles. 4 = 2 + 2 14:24, 28 December 2009 (UTC)

It appears as though all of your prods have been removed, most of them summarily by 4, and one by Colonel Warden. Since Warden's purported refutation of the prod is the role played by sextic equations in Hilbert's 13th problem, perhaps a redirect is in order instead. Sławomir Biały (talk) 12:40, 28 December 2009 (UTC)

I'm the one responsible for all of this. See fifth dimension and sixth dimension, though: maybe all these lower dimensions should be going to "m-dimensional space". In fact, that's what I'm going to do now. 4 = 2 + 2 12:43, 28 December 2009 (UTC)

But of course, I'm going to wait till people here say "yes" or "no". I have put the 1D, 2D & 3D articles at 1D space, 2D space and 3D space. 4 = 2 + 2 12:46, 28 December 2009 (UTC)

OK, I've made the proposal in various places, but let's do it here:

1. Delete redirects 1st dimension, 2nd dimension, 3rd dimension, First dimension, Second dimension, Third dimension

   Weak, but plausible, as nominated by Prod. It should be noted that redirects may be nominated by prod.

2. Split 4th dimension // fourth dimension // 4-dimensional space into two unrelated articles:
   1. 4th dimension // fourth dimension on the common usage (time, etc.) (base should probably be at fourth dimension, as it's primarily common usage, rather than mathematical usage)
   2. 4-dimensional space on the spacial properties of 4-dimensional spaces.
3. Move, without redirect,
   1. 1. Fifth dimension to 5-dimensional space
      2. Sixth dimension to 6-dimensional space
      3. Seventh dimension to 7-dimensional space
      4. Eighth dimension to 8-dimensional space
      5. Ninth dimension to 9-dimensional space
   2. and then merge all common sections to n-dimensional space or perhaps n-dimensional geometry

— Arthur Rubin (talk) 15:37, 28 December 2009 (UTC)

If there is general consensus for this here, we need to tag the articles to point to a centralised discussion. — Arthur Rubin (talk) 15:41, 28 December 2009 (UTC)

4 is making partially supported changes faster than I can propose correction. — Arthur Rubin (talk) 17:30, 28 December 2009 (UTC)

Aren't you supporting the moves in the first place? 4 = 2 + 2 04:20, 29 December 2009 (UTC)

I agree with the moves; not necessarily with deleting the redirects. At the very least fifth dimension needs to be a bluelink somewhere, if only so that there can be a hatnote, wherever it points, to The Fifth Dimension. When the Moon is in the seventh house.... --Trovatore (talk) 02:41, 29 December 2009 (UTC)

Tensors on Wikipedia

Every so often, someone points out that our treatment of tensors is terrible. We currently have a bunch of badly written stubby articles: Tensor, Tensor (intrinsic definition), Intermediate treatment of tensors, and Classical treatment of tensors, not to mention other satellite articles like covariant transformation and covariance and contravariance of vectors in various states of shabbiness. Some well-meaning editors who otherwise lack any detailed knowledge of the subject seem to be interested in whipping our treatment of tensors into shape (please see my talk page for recent discussions with User:LokiClock). But I think the broader WP:WPM community should get involved (physicists are welcome too). Some of the recent edits have, in unskilled hands, been a step in the wrong direction. I don't have a clear picture of how things should look, but I know that the current status quo is pretty far from ideal, and I think there is at least consensus on this last sentiment.

I would hope that we could be able to get the current four main tensor articles down to two articles: Tensor for the full theory and Introduction to tensors for a gentler approach (on the model of the various "Introduction to X" articles that we have on Wikipedia). Comments and assistance are welcome. Best, Sławomir Biały (talk) 00:21, 29 December 2009 (UTC)

I'm not too keen on 'Introduction to' articles. I think the main article for topics should be written that way anyway for at least the first few sections and extra articles set up where the main article needs expanding. The classic treatment of tensors article doesn't strike me as anything like what I'd have expected there. Dmcq (talk) 10:10, 29 December 2009 (UTC)

Second on 'Introduction to' articles; every article should be an introduction to whatever the subject is. But User:Sławomir Biały has a point in that there are too many articles about basically the same thing. Unfortunately, we have a huge backlog of articles requiring attention, but Tensor seems like a worthwhile and basic subject that should be given a high priority.--RDBury (talk) 12:59, 29 December 2009 (UTC)

IMO, there should be only one tensor article. There is only one concept of tensor, a fact which is often obscured by physicists and mathematicians using completely different languages to discuss them. Since a wikipedia article should not rely on any understanding of any of those two languages this should not hamper writting a good wikipedia article about tensors. (As long as the editor recognized the difference in language and explain both.)

The tensor article will attract readers for diverse backgrounds, and as such should not rely on a detailed mathematical preknowledge. In particular, it should (at least for the first few section) not rely any knowledge of tensor products. Of course, it should contain the precise definition which will include tensor products, but this should be put into context for those that do not know what a tensor product is.

I don't have time right now to spearhead the considerable project of creating a single article. But I will definitely try to help out wherever I can if somebody else takes up the challenge.TimothyRias (talk) 12:37, 29 December 2009 (UTC)

I'm very encouraged by your post. I should clarify that I chiefly want a more unified treatment of tensors on Wikipedia. I would not be opposed to having just one article, but we currently have 4+ articles. A target of two might be easier to hit than a target of one. Or perhaps not. Sławomir Biały (talk) 13:00, 29 December 2009 (UTC)

I think that Tensor (intrinsic definition) looks like a fork of tensor product and that Classical treatment of tensors is almost content free. Merging those two with appropriate pages would be an easy start. The article tensor itself seems remarkably readable as a general introduction, so I am not sure why these other "introductions" are needed. — Carl (CBM · talk) 13:12, 29 December 2009 (UTC)

I've gone ahead and made the merger proposals. See Talk:Tensor. TimothyRias (talk) 15:26, 29 December 2009 (UTC)

That being said, it would be worth contacting the physics project and getting their opinions as well. — Carl (CBM · talk) 13:13, 29 December 2009 (UTC)

I just did a quick search and we have more like 20 articles with a subject beginning with 'Tensor'. We also have Tensor product which probably should have it's own article, and several Tensor product of X. Also Tensor field which I assume is a physics version of the subject. So I think the question is really how big a project to you want to do?--RDBury (talk) 13:21, 29 December 2009 (UTC)

For the moment I'm willing to live with satellites like Tensor field, and even the twins covariant transformation and covariance and contravariance of vectors (at least, since the latter has gotten into somewhat decent shape of late). The plethora of tensor product articles also aren't a huge concern for me at the moment, because these are by and large in better shape and (if you forgive my saying so) of less critical importance. Sławomir Biały (talk) 13:31, 29 December 2009 (UTC)

There a whole bunch of subjects related to tensors that merit their own article. Things like Tensor bundle (and related Tensor field), Tensor algebra, Tensor product, etc. are important of their own right and only loosely related to tensors themselves. We should not be aiming to merge those as well.TimothyRias (talk) 15:30, 29 December 2009 (UTC)

I believe it would be necessary to get the physics projects cooperation on this, otherwise the articles will just end up in the jumble in which they currently are. The first problem is that of the use of the word "tensor". Physicists (and mathematicians) use it to mean "tensor field" (and in fact, physicists mean something even stronger, as they normally consider the the behaviour of the tensor field under the orthogonal group, leading to the "tensor vs. pseudo-tensor" nomenclature). I don't think there's anything wrong with this use, but I believe it should be recognized as a "use", not as the actual name of the object. Thus, the first thing I would suggest would be to obtain a consensus within this project and with the physics project, on the title of articles. For this, it is logical to use the precise names of the different concepts, since they exist; however, these names are mathematical and physicists tend to resist such things as being too pedantic (though, frankly, I believe an encyclopedia should reach levels of pedantry higher than everyday scientific discourse). A "tensor" should be some element of the tensor algebra of a vector space, for example, whereas a varying family of tensors on a manifold should be called a "tensor field". Good luck. RobHar (talk) 16:50, 29 December 2009 (UTC)

There are basically two points here, and they both require respect. First is that Tensor should relate to numerous other articles by proper use of summary style; and doesn't. And secondly POV forks should not be present in the material (this is the point that there is just one concept of tensor). There are other points that help to explain where we are (e.g. that pedagogy in this area is a matter of taste or who you talk to, that Bourbaki is a demon or angel in imposing multilinear algebra as the setting, and the path-dependence of our ability to sort things out). These things are nothing much to do with writing encyclopedias, though, and people who mention them should have that explained, as ever politely. I know why Classical treatment of tensors is there in that form, and something definitely should be done about it, since "treatment" does not, based on what I have been saying, justify a separate topic. Charles Matthews (talk) 16:49, 29 December 2009 (UTC)

Copyright concerns related to your project

This notice is to advise interested editors that a Contributor copyright investigation has been opened which may impact this project. Such investigations are launched when contributors have been found to have placed copyrighted content on Wikipedia on multiple occasions. It may result in the deletion of images or text and possibly articles in accordance with Wikipedia:Copyright violations. The specific investigation which may impact this project is located here.

All contributors with no history of copyright problems are welcome to contribute to CCI clean up. There are instructions for participating on that page. Additional information may be requested from the user who placed this notice, at the process board talkpage, or from an active CCI clerk. Thank you. --Moonriddengirl ^(talk) 17:39, 19 December 2009 (UTC)

Just wanted to add that, even if you aren't interested in looking for copyright violations in these articles, cleaning up the ones identified could be very useful. They are marked accordingly at the CCI page. Articles that are blanked and listed at WP:CP, but not replaced by clean, are likely to be deleted, unfortunately. --Moonriddengirl ^(talk) 19:37, 19 December 2009 (UTC)

As I noted before (#Noncommutative measure and integration), this is an editor who created 20 articles in a short period of time, and these seem to now be target of your investigation. Imo, most of these (at least the math ones) are of dubious value even without the COPYVIO issue. But in hindsight it should have been a red flag for more serious issues. I'll go through and see if there are any math ones that should be saved, but I tend to be a deletionist so someone else should take a look as well.--RDBury (talk) 09:44, 20 December 2009 (UTC)

Here is a list of articles I found that seem meet notability criteria. There is incomplete since Google decided I was a bot trying to crash their server, will finish later after my time-out is over. There are a bunch of articles I didn't include because they aren't math related, and bunch more because notability was questionable. Nearly all the math topics had some sort of mention in research papers, but I was looking for non-trivial mention in a textbook. Even without possible copyvio problems, many of these should be looked at for style, context, questionable references. Most of them aren't subjects I know a lot about so I'm not going to attempt any repairs.

Hadamard variance, Steklov function, Quantum torus, Noncommutative quotients, Quantized calculus, Poincaré–Steklov operator, Fictitious domain method, Multiple-point constraint, Quantum affine algebra, Quantum time (Currently blanked), Quantum invariant (Currently blanked), Quantum double, Quantum algebra, Arithmetic variety, Quantum relativity, Axisymmetric spacetime, Canonical mapping, Concurrent relation, Qualitative calculus, Yang–Mills–Higgs equations, Vorticity tensor, Albert–Brauer–Hasse–Noether theorem, Cyclic algebra, Lambert's trinomial equation, Lambert of Auxerre, Quadratic algebra, Larmor's theorem, Lawson topology, Measure algebra, Open coloring axiom, Paneitz operator, Geodesic ball, Barta's theorem, Gregory M. Fikhtengol'ts, Codistribution, Cotensor, Matthias Flach, Laplace series, Scale property, Judith Q. Longyear

--RDBury (talk) 12:53, 20 December 2009 (UTC)

Here are some more: Model management, Spectral triple, Noncommutative plane, Graded C*-algebra, List of noncommutative topics in mathematics, Noncommutative integral, Noncommutative residue, Noncommutative Jordan algebra, Noncommutative fundamental theorem of algebra (probably should be Fundamental theorem of noncommutative algebra), Connes theorem, Noncommutative mathematics (This didn't actually turn up many hits but it's so broadly defined that maybe it's notable), Noncommutative Shilov boundary, Journal of Noncommutative Geometry (Need to check Wikipedia:Notability (academic journals)), Heinz Bauer, Differential invariant

I've now gone through all the articles he created, there might be more on the major edits list but those can always be reverted to pre-copyvio and probably not subject to deletion. I didn't really check the bio ones, just added them to the list since WP:BIO and WP:PROF are harder to check. Now I'm going to take a noncommutative shower and then go see noncommutative Avatar.--RDBury (talk) 17:44, 20 December 2009 (UTC)

while none of the articles are exactly "thrilling", for most of them I have difficulties to see any apparent signs of copyright violations or a need for deletion.--Kmhkmh (talk) 13:44, 20 December 2009 (UTC)

P.S.: Apparently consulting older versions of the author's discussion page provides further insight. Allegedly he might have copied sentences or small paragraphs from various websites rather than summarizing his cited references (in particular in early versions of the concerned articles). So other than completely rewriting the articles in general, one would need to look at earlier versions and run some bot/software for finding possible originals on the web to figure out whether there was an actual copyright problem needing correction.--Kmhkmh (talk) 14:02, 20 December 2009 (UTC)

I took care of Lawson topology. The referencing and notation there were both somewhat dubious. For example, it included a reference to Topology by T. Lawson, although the Lawson topology is named after J. Lawson and T. Lawson's book does not seem to mention the Lawson topology at all. So be careful in looking at the articles this person created. — Carl (CBM · talk) 13:43, 20 December 2009 (UTC)

I cleaned up a few articles yesterday and noticed two things that are helpful in looking for his copyvios.

1. He likes to cite his sources. Sometimes he tells you exactly where he stole from.
2. He likes to steal from the arXiv.

I've rewritten noncommutative algebraic geometry (the rewrite is at Talk:noncommutative algebraic geometry/Temp), and I checked all the math and physics articles between 221 and 240. Fortunately he doesn't seem to be subtle about his violations. Ozob (talk) 14:15, 20 December 2009 (UTC)

I've replaced the copyvio versions with your clean rewrite. Thank you and others of the project who are responding to this need. :) --Moonriddengirl ^(talk) 14:58, 20 December 2009 (UTC)

I had previously noticed that this editor's mathematical contributions were a bizarre mixture of sensible mathematics and useless nonsense, and I guess this is now explained: the good stuff was just copied. I recommend that all his mathematical edits should be deleted, because anything that is not a copyvio is likely to be wrong. In particular it would save time just to scrub all articles with no significant contributions from anyone else. I checked Connes theorem, which consisted entirely of a blatant copy from one of his sources; anything else like this can be speedy deleted using {{Db-g12}}. He does at least often give the site he has copied from. r.e.b. (talk) 22:08, 20 December 2009 (UTC)

I tend to agree, better to concentrate efforts on fixing the articles that should be kept (there are a few examples in the list above) than worry about the lost causes. It does seem that he put a lot of effort in these articles, so it seems a pity to just delete them, but if the material shouldn't be on Wikipedia then it needs to be removed. Maybe the deleted articles can be added to the list of requested articles so perhaps someone will write another version.--RDBury (talk) 07:48, 21 December 2009 (UTC)

I greatly appreciate the work your project is putting into this. :) r.e.b. at my talk page asks about presumptive deletion where this editor is the only major content contributor. This approach is certainly supported by policy at Wikipedia:Copyright violations. ("If contributors have been shown to have a history of extensive copyright violation, it may be assumed without further evidence that all of their major contributions are copyright violations, and they may be removed indiscriminately.") Some admins may choose to delete immediately. I would recommend blanking the article for listing at WP:CP as a courtesy to give any other interested contributors an opportunity to (a) help evaluate to verify that there is no infringement, or (b) rewrite the article before it is deleted. I have created a new template that can be placed on the talk page in such situations: {{subst:CCId}}. I would recommend placing the following on the article's face — {{subst:copyvio|url=see talk}} — and putting {{subst:CCId|name=Henry Delforn}} on the article's talk. --Moonriddengirl ^(talk) 18:11, 21 December 2009 (UTC)

I have a comment and a question. Comment: I disagree with the proposed remedy of summarily deleting the articles. Firstly, several topics are notable and already linked, and, therefore, having a stub is better than having nothing at all. Further, references and especially comments on talk pages aimed at improving the article will be irrevocably lost. Quantum affine algebra provides a good illustration to these concerns. Question: how is one supposed to go about fixing these articles? It seems to me that the placement of the template which simultaneously blanks the old text (including potential contributions by other editors) and gives a warning not to edit it would prevent one from effectively editing the articles. Does the new text need to be validated or approved by some "copyviolation committee"? Arcfrk (talk) 21:29, 25 December 2009 (UTC)

If you would like to rewrite the article from scratch, you may simply do so. The old copyright-violating revisions will then need to be deleted. You can contact me or any other admin and we can take care of that. — Carl (CBM · talk) 14:26, 30 December 2009 (UTC)

Strange edit summaries by Golumbo

This is a new account making exclusively edits to math articles "reverting edits by a banned user". I looked at some and could not find what was being reverted and who that banned user might be. Can someone look if the edits make sense? Jmath666 (talk) 05:27, 28 December 2009 (UTC)

The banned user in question appears to be Afteread (talk · contribs), who was blocked by John Vandenberg (talk · contribs) as being a banned user (the block log doesn't say which banned user). However, Golumbo's reverts are also indiscriminately taking out other edits by other editors; e.g. his revert of Alexander Grothendieck to its pre-Afteread version also eliminated edits by Mavros-Belas (talk · contribs), 141.160.5.251 (talk · contribs), and Myasuda (talk · contribs). The situation could probably bear more careful inspection by math-literate editors. —David Eppstein (talk) 05:35, 28 December 2009 (UTC)

Sorry for the confusion or dismay. Here's the comment I made on Wikipedia_talk:WikiProject_Physics describing the situation: There appears to be a banned user creating multiple accounts serially, mostly editing BLP and physics articles. Their accounts may include Verbapple (talk · contribs), Afteread (talk · contribs), and Miles1228 (talk · contribs), all currently blocked. The edited articles include Bogdanov affair, Marcus du Sautoy, Edward Witten, Antony Garrett Lisi, An Exceptionally Simple Theory of Everything, Andrew Wiles, Elliot McGucken, Pythagorean triple, Lee Smolin, Leonard Susskind, Steven Weinberg, The Elegant Universe, Frank Adams, Jan Hendrik Schön, John C. Baez, Theory of everything, Peter Woit, Standard Model, Alexander Grothendieck, The Story of Maths, Kent Hovind, Luboš Motl, and several others. The banning policy encourages but does not require editors to enforce the ban:

"Anyone is free to revert any edits made in defiance of a ban. By banning a user, the community has decided that their edits are prima facie unwanted and may be reverted without any further reason. This does not mean that obviously helpful edits (such as fixing typos or undoing vandalism) must be reverted just because they were made by a banned user, but the presumption in ambiguous cases should be to revert. When reverting edits, care should be taken not to reinstate material that may be in violation of such core policies as neutrality, verifiability, and biographies of living persons. Users who reinstate edits made by a banned editor take complete responsibility for the content by so doing."

In the course of reverting some of this banned user's extensive edits, I may have reverted useful edits by them or others. Feel free to edit these back in as appropriate, as per the banning policy. I am not an admin, just another editor who took issue with this banned user. Golumbo (talk) 06:00, 28 December 2009 (UTC)

This is the typical approach to dealing with this particular banned user. Sorry I cant give more details publicly.

Definitely check the edits, and replay any subsequent edits which are good.

John Vandenberg ^(chat) 06:03, 28 December 2009 (UTC)

Thank your for clarification. However, if an editor wants to revert a past edit, whatever the reason, wouldn't it be up to that editor to do properly and actually revert the changes from that edit (which may have to be done manually and may take some work), rather than simply reset the article to just before the edit in question and lose all work since then? Jmath666 (talk) 08:04, 28 December 2009 (UTC)

Yea, it is preferable to carefully remove the edits by the banned user without also removing the edits by other users.

It gets messy when many other people have edited the same passages after the banned user; in cases like this recent sock, it is often simpler to roll right back.

I don't like doing that, but this person is a special case. John Vandenberg ^(chat) 09:42, 28 December 2009 (UTC)

If you're not willing to share information on why this is needed, I think we're going to have to ask you to stop reverting in this fashion, at least on math articles. CRGreathouse (t | c) 19:18, 28 December 2009 (UTC)

Vandenberg has shared some information on the case with me privately, enough to convince me that the user in question should be kept as far from Wikipedia as possible. The number of edits in this case seems small enough that it shouldn't be too much work to figure out whether anything of value was lost and restore it. —David Eppstein (talk) 19:36, 28 December 2009 (UTC)

Fine, I withdraw my suggestion. Hopefully someone will keep track of this so we can restore edits when needed. CRGreathouse (t | c) 19:51, 28 December 2009 (UTC)

John Vandenberg (talk · contribs), could you explain why you believe Afteread (talk · contribs) is a banned user? Inspecting just a few of these articles (John C. Baez, Alexander Grothendieck, Bogdanov affair), all of Afteread's edits seem to me obviously helpful. Noamz (talk) 20:03, 28 December 2009 (UTC)

Wait, no, they don't. Sorry, I didn't notice all the changes Afteread made on the Alexander Grothendieck article -- there are a lot. Some of them look reasonable but many are questionable. Noamz (talk) 20:24, 28 December 2009 (UTC)

I had a look and it looks like someone is writing his less than notable contributions into a lot of places, you notice it more by looking at the stranger entries. Basically it looks like a very energetic ego tripper with some strange ideas who actually does know some maths and physics so its a bit sad they corrupt their contributions with this sort of stuff. Not sure why they don't mention the name I suspect is at the bottom of it all. Dmcq (talk) 23:39, 28 December 2009 (UTC)

Just had a look at more of the stuff and have concluded they are probably starting down the same kind of route as Archimedes Plutonium. Dmcq (talk) 23:53, 28 December 2009 (UTC)

ZOMG WP:CABAL. I have no idea what you guys are talking about, but the edits to Grothendieck were questionable speculation. Pcap ping 00:17, 29 December 2009 (UTC)

We avoid naming the person as a conscious decision to avoid harm, and because we don't want to give them cause to participate.

There are not many people for whom we have a scorched earth policy; this banned user is one of them.(offhand, I cant think of any others)

John Vandenberg ^(chat) 00:44, 29 December 2009 (UTC)

That's an awfully vague explanation for something that looks as non-transparent as this. I think you owe us better. --Trovatore (talk) 01:49, 29 December 2009 (UTC)

Just a cowardly remark: I would point out that the edits of this person precede User:Afteread, and appear to be traceable to the real life identity. I'm comfortable that Checkuser confirmed this, that blanket reversal is a required and appropriate approach, and also agree that a number of good edits are temporarily moved down the "history" list (until it's cleaned-up, which it will). For supporting the ban from an admin side, I would suggest not to point to some "inner circle"-style authority for making "the right" decisions, but simply state the abuse pattern: A large number of Wikipedia user names with a huge number of "minor" marked edits (that are mostly of high quality) attempt to obscure individual edits here and there, that range from sneakily "slanting" (primarily in the physics community) to outright suppression attempts (primarily in Christian young-Earth creationism related articles). -- Signed by another coward. BeforeAfteread (talk) 02:09, 29 December 2009 (UTC)

I don't see why it's our problem if his edits are traceable to real-life identity. Let them be traceable. Frankly there is far too much bending over backwards to prevent people's identities from coming out, when those people give clues all over the place as to who they are. --Trovatore (talk) 02:33, 29 December 2009 (UTC)

The issue isn't the real-life identity of the user being reverted. Paul August ☎ 03:31, 29 December 2009 (UTC)

• John Vandenberg is correct. Please assist him by fixing any collateral damage due to reverts. Thank you. Jehochman ^Brrr 02:10, 29 December 2009 (UTC)

The situation here is truly a special circumstance. There are very sound and considered reasons behind these reverts and the circumspection with which the reverts are being carried out. Paul August ☎ 03:27, 29 December 2009 (UTC)

I think I've figured it out.

This user is a suspected sock puppet of Cthulhu. Revert all edits of this user immediately. Account information: block log – current autoblocks – edits – logs – abuse log

CRGreathouse (t | c) 05:02, 29 December 2009 (UTC)

LOL BeforeAfteread (talk) 14:27, 29 December 2009 (UTC)

I am here to say that I agree with the actions that John and Paul and others have taken in this matter and to ask for forbearance if less explanation than usual is offered. That's not entirely satisfactory but the less discussion of this, the better. ++Lar: t/c 05:25, 29 December 2009 (UTC)

You guys are doing good work. User:Golumbo's list above (and his/her contributions) are a good reference point for the large amount of edits affected. The abuse of Wikipedia is systematic, and the revert is justified from the nature of these edits alone. Nothing else needs to be said. While it is regrettable to revert many edits of high quality, from a knowledgeable editor, the attempt to slant articles in a way that is labor intensive for others to follow-up and dispute, cannot be accepted. If someone believes that an article is incorrectly representing its content, then he/she can make a point with far fewer edits, and discuss on the talk page. The burden of justifying a change in tone on a Wikipedia article should lay with the editor who brings forward that change, not with all others who'll have to do the clean-up. BeforeAfteread (talk) 14:27, 29 December 2009 (UTC)

I have been sent to this discussion from John Vandenberg following the tersely-justified deletion of a page I had in my userspace, which has something to do with this case. Can someone explain clearly what is going on? I mean, we don't have a scorched earth policy even for Daniel Brandt. What is going on here? No matter how much AGF I can put in my tea, it is unacceptable such a full secrecy between admins and the community. --Cyclopia^talk 14:32, 29 December 2009 (UTC)

Some of us may have overreacted, our bad; we're human, too. Nobody is "banned", there is no "scortched earth", and every editor's contribution is a gift to Wikipedia, whatever the takeaway. -- I stand behind my support for the blanket revert, due to the vast scope of the edits, impact on the clean-up crew, and intent. BeforeAfteread (talk) 14:43, 29 December 2009 (UTC)

I don't know anything of Golumbo, the reverts, etc. and I have no opinion on that. What I have opinion on, instead, is the secrecy that is going on in this case. Unless we're protecting someone's identity for outing reasons (but above is mentioned that it is not the reason), there should be no secrets on WP. I even was refused any explanation by email. This is not acceptable. At least, a reason for all this secrecy has to be given. "We know better" is not a good reason: it only shows contempt for the community. --Cyclopia^talk 15:05, 29 December 2009 (UTC)

Support that. If there is somewhere this is discussed I'd have thought that would be a reasonable place to direct people to even if the banned user might unfortunately follow the links back to here. Dmcq (talk) 15:12, 29 December 2009 (UTC)

The banned user only ever needs to click "What links here" in the toolbox on the left on their banned accounts user pages to see where they're being discussed. And the secrecy only seems to be prolonging the discussion. Ben (talk) 15:19, 29 December 2009 (UTC)

Just to chime in as a relatively junior WP:Math contributor but also as a wiki functionary who has a little more info about these edits that this is a rare case that requires both quick and decisive action with a minimum of "on-wiki noise". If there is ever a time that we should trust that ArbCom is acting in the best interests of the project and each and every one of its members, this is it. -- Avi (talk) 15:25, 29 December 2009 (UTC)

I don't think the issue is about trust so much as it is about respect. I trust that the ArbCom has made the right decision in banning the editor, but as Cyclopia noted, there is a feeling of contempt for the community in the way this is playing out. That userfied pages are now being deleted only amplifies that feeling. Ben (talk) 15:40, 29 December 2009 (UTC)

People who ask reasonably, and privately, once... get answered to the best of the ability of the person they ask, keeping in mind what can and can't be said. Please remember, that those of us handling it have the trust of the community to handle sensitive matters (everyone involved holds at least CU here, if not more permissions, and most are ArbCom, ex ArbCom, stewards, oversighters and the like). That means trusting us when we say that not everything can be discussed, and not worrying at the matter repeatedly. That defeats the purpose of not discussing it in the first place. Please review WP:RBI and think about what that page means/implies if what you are dealing with is one of the worst editors ever. Which we are.

People who keep worrying at this, over and over, and who get increasingly shrill about their "rights" being trampled, or demanding that they be given text back, or saying we're all exactly equal and they are entitled to know every detail of every private matter... these people get increasingly short shrift. If they keep up to the point of being disruptive, they may be sanctioned. It has happened before. I would not like to see it happen again but it could. ++Lar: t/c 15:53, 29 December 2009 (UTC)

Threats. Nice. --Cyclopia^talk 15:57, 29 December 2009 (UTC)

Knowing what Lar knows, I think he just lost patience with the discussion here. I don't think sanctions or the threat of sanctions are necessary or helpful here. And I'm sure that Lar agrees. This is really a very unfortunate situation. Maybe we can all count to ten? Paul August ☎ 16:24, 29 December 2009 (UTC)

Nod. @Cyclopia: If you're reading what I wrote as a threat, I'm really sorry. it's not intended as a threat at all. Just a warning. This is a very serious matter. People have been sanctioned before and I'd rather not see it happen again. As I said... asking privately is fine but when those given extra permissions and extra trust say "please trust us"... the thing to do is trust them. All of ArbCom has awareness of this matter (in the larger context) and what we do to handle it is being scrutinized, closely, so trusting us is the appropriate and polite thing to do, instead of worrying at it. Will you do us that courtesy? ++Lar: t/c 16:29, 29 December 2009 (UTC)

Lar explained me the bare minimum necessary to understand what is at stake. As such, yes, issue solved, and sorry for the hassle. It would have been simpler if the seriousness of situation was clear for the start, but I'd say issue resolved. --Cyclopia^talk 16:44, 29 December 2009 (UTC)

Thanks. Paul August ☎ 17:44, 29 December 2009 (UTC)

Afteread (talk · contribs) and his odd edits had already been reported 2 weeks ago on WP:FTN here. Mathsci (talk) 17:05, 29 December 2009 (UTC)

@Avi. As much as I respect John Vandenberg he's no longer part of ArbCom, and did not claim to act under their mandate either. Pcap ping 02:28, 30 December 2009 (UTC)

He is acting on behalf of ArbCom, and he could have noted as such, although he didn't :) -- Avi (talk) 05:33, 30 December 2009 (UTC)

I believe that there are members of Arbcom who responded here, and if not, please trust me that from the discussions I have been privy to, they are pretty much all in agreement. -- Avi (talk) 05:37, 30 December 2009 (UTC)

People, the average case and the worst case are not the same, or anything like, in these matters. One feature of the "worst case" is that even defining why it is not just a "bad case" can be very problematic. I have no idea at all about the details of this matter, but if the ArbCom is in the loop, that's good enough for me. Charles Matthews (talk) 17:03, 29 December 2009 (UTC)

It is not good enough for me. Nobody here has pointed to any discussion or decision on the part of ArbCom. I am not in the loop, not connected to all the right people, not in the know; I cannot tell whether this editor is being condemned because he is one of the worst editors ever or because a cabal has a personal vendetta against him. I trust many of the people involved in this discussion, so I would like to believe that this editor is one of the worst ever. But nobody has provided evidence; we are asked to believe not in the truth but in interpersonal politics. I am sorely tempted to call these reversions blatant vandalism and to re-revert myself. I would, however, switch my position immediately if only someone could prove that Afteread is subject to legitimate sanction. Ozob (talk) 17:59, 29 December 2009 (UTC)

Unfortunately providing public "proof" may do more harm than good. Paul August ☎ 18:21, 29 December 2009 (UTC)

Ozob, I understand you completely, per my comments above. But it turns out it is right, even if maybe how admins are handling the matter with users is suboptimal. Ask some admin to email you in private, like I did and (not without some difficulty) obtained. Let's say that there are very serious real life concerns. --Cyclopia^talk 18:23, 29 December 2009 (UTC)

@Charles: I suspected as much from the edit history, but I'm really clueless why writing "sockpuppet of someone banned in BA case" is beyond the pale. Pcap ping 02:35, 30 December 2009 (UTC)

By the way, I also clueless why an WP:AE report (now archived) that seriously concerns NPOV in an article sees little commentary and practically no action, whereas this is scorched earth territory. Pcap ping 02:41, 30 December 2009 (UTC)

Someone more familiar with the details emailed me privately. Apparently this user is unusual, and there is a good need for discretion, yes, even to the point of making this "scorched earth territory". But I still object to the secrecy around this case. I still do not see why it is unacceptable to say, "He is a sock of the banned user So-and-so" or at least to make a public statement as to why identifying him is inadvisable. As I wrote in an email,

Maybe I can present my concern from a different perspective. In Britain at one time they had a court called the Star Chamber; you can look it up on WP, it has a decent article. The proceedings of the Star Chamber were entirely secret. As you might expect, this eventually lead to abuses and to the abolition of the Star Chamber. One of the reactions to the Star Chamber was the English and American legal systems' principle that evidence and proceedings should generally be public. Exceptions are very rare and contentious (think of Guantanamo and all of the scrutiny which military tribunals there underwent). I think that principle is extremely valuable, and I am very reluctant to endorse any exceptions to it. I think it is a valuable principle for WP also. A system such as WP's which relies so heavily on consensus cannot function without wide distribution of information.

Of course, as I went on to note, I don't like to get involved in user conduct disputes. Furthermore, I suspect that in this case, the details are sensitive enough that the general public, including me, should not know all of them. So I'm left uneasy and unsatisfied. I have learned, again by email, that there are several of us who feel this way.

I do have one suggestion for those who have more detailed knowledge of the situation. I would not be surprised if some day there is another thread similar to this one. It might be good if you could write a one sentence description along the lines of, "We cannot discuss this situation because of ..." I think that would help relax those of us who don't know the details. Ozob (talk) 05:19, 30 December 2009 (UTC)

I have been sent the name of this editor via email (I just had to ask for it). I have found some of the details in the internets. Turns out that I had previously read about this editor in the internets, and I was adequately impressed by his... uh... varied accomplishments and his .... uh.... actions in.... uh....... you know what, forget it, this just can't be summarized without revealing private stuff. I understand why the details are not being released, and I fully endorse the nuking-from-orbit approach to this matter. --Enric Naval (talk) 07:05, 30 December 2009 (UTC)

If we strip out all the stuff here that we maybe shouldn't have been told about from starters, all those details that cannot be revealed (and honestly that I don't care to know, either), and only look at the edits to hundreds of pages that John and others are cleaning up now, then one can recognize the sophisticated and deliberate abuse pattern. The only "trust" (if any) that one may want to bring forward IMHO is towards chechusers for validating the identity behind the dozens of sock puppets and IP addresses. And even then, even if you are suspicious of checkuser privileges (which is a good thing! question authority and power!), John is marking his edit summary clearly by stating "banned user", which makes everything that's going on to the content in Wikipedia fully transparent. And that's all I'm concerned about, for Wikipedia, and by looking at the content alone I am satisfied with the actions. There is no secrecy or hiding of content. As for the rest, it's good drama - wow. BeforeAfteread (talk) 16:25, 30 December 2009 (UTC)

My actions can be reviewed by looking at my contribs from 2009-12-28 09:10 (UTC) until today (and probably inc. tomorrow).

Here is a link of the first 500 edits that are relevant[2].

If you see some good content being lost in this process(and only if you know the topic well), please don't re-instate the reverted content; instead, rewrite it in your own words.

John Vandenberg ^(chat) 00:03, 31 December 2009 (UTC)

David Madore

Do you guys think he qualifies for a bio article here? Based on the current contents of his bio, I'm in doubt. There was a discussion in 2005, way before we had any standards. The argument that he invented Unlambda is exceedingly weak; I could barely find secondary references to add to that article— it's not a well-known language by any stretch. Pcap ping 08:31, 30 December 2009 (UTC)

Based on what's in the article now no. The best thing might be to put it up at AfD: it may get deleted but that also often leads to interested editors taking the time to find references to support the subject's notability, improving the article in the process. --JohnBlackburne (talk) 10:05, 30 December 2009 (UTC)

Done so. Thanks for the advice. Pcap ping 12:17, 30 December 2009 (UTC)

In general, it's impolite to use AFD as a way to force other people to look up references for you. But in this case, the AFD seems to be appropriate, based on Madore's CV. The spirit of WP:PROF is that we want to cover only the people who are more notable than the average successful academic. It appears that Madore is doing well, but has not yet become more notable than average. In general, if someone who has only had their degree for a few years passes WP:PROF, they will be obviously notable (e.g. Ben J. Green). — Carl (CBM · talk) 13:56, 30 December 2009 (UTC)

Adding an example to Weak convergence (Hilbert space)

I'd like to add an example to this page to help bright the concept to life a bit, but I want to collaborate with someone who know more than me about this before making any major edits. futurebird (talk) 02:49, 31 December 2009 (UTC)

Tool for maths ratings

Apropos to previous discussions about articles whose {{maths rating}} data is incomplete, I have two tools to make it easier to manually fill in the missing information.

1. http://toolserver.org/~cbm/cgi-bin/mathsrating.fcgi – this wraps the Wikipedia interface inside a frame, and randomly loads an article that needs to have its maths rating template fixed. You can edit as usual within the frame. Click 'reload' at the top to either skip the present article or move to the next one. The count at the top is for all articles in any of these categories:
   • Category:Unassessed_mathematics_articles
   • Category:Unassessed_quality_mathematics_articles
   • Category:Unassessed-Priority_mathematics_articles
   • Category:Unassessed_field_mathematics_articles
2. User:CBM/ratemath.js – this adds a link to the user interface at the top of an article, allowing you to set the maths rating template without having to load the talk page or type any code by hand. Occasionally, it may balk at removing the old template, in which case you have to do it by hand in the usual way. To use this tool, you need to put importScript('User:CBM/ratemath.js'); in your monobook.js or vector.js (depending on which skin you use) and then clear your cache. When using the tool, I recommend always waiting for the talk page to be loaded, to double-check the output. This is not intended for automated use in any way.

If you have time and are interested, doing just a few articles a day would be a great help to the project. There are about 1500 articles that need attention. — Carl (CBM · talk) 18:00, 29 December 2009 (UTC)

Very useful tools. Thanks, Carl! I have a question about each of them.

1. For the first one, is it possible to have the category searched be determined by the user? For instance, I would like to assess only those math articles that I understand, so for instance I might only want articles in Category:Graph theory.
2. For the second tool, can it be programmed so that when you click on "rate math", it picks up the current rating and displays it by default? For example if someone has already assessed the field, but not the priority, can it show the pre-assessed field, and load in some default values for the rest. I guess this might be hard to program, but it would be neat if that were possible.

Robin (talk) 20:10, 29 December 2009 (UTC)

Those would not be as easy as it seems. The problem with categories is that our category tree is very deep, so the system would have to do a recursive scan of the category tree somehow. The javascript tool is something I hacked together a while back despite not knowing any javascript. I don't think I will be able to improve them much, although I agree that the improvements would be nice. — Carl (CBM · talk) 21:00, 29 December 2009 (UTC)

I understand. I have a makeshift solution for both problems, so I'm satisfied. For the first, User_talk:GregU/randomlink.js randomly loads a page linked to on a given page. It also implements a link depth of 4 links, so one could start at a Category like Graph theory, and the random link could take you to a page like Graph theory>Graph coloring>Perfect graphs>Chordal graph. That's good enough for me. For the second problem, I installed the project banners script -- User:Pyrospirit/metadata/projectbanners.js. This gives a quick overview of the project banners on the article page itself, saving me a click on the "talk" button and back. --Robin (talk) 21:56, 29 December 2009 (UTC)

Not sure if this has been covered before, but I think even putting the maths rating tag in without parameters has some value. Specifically the bot that updates the article alert page only recognizes articles which have the tag.--RDBury (talk) 12:08, 30 December 2009 (UTC)

We don't really use that system; we use Wikipedia:WikiProject_Mathematics/Current_activity which does not rely in any way on talk page tags and thus covers a more complete set of articles than article alerts. This is one reason why we have always rejected letting a bot go through and add a template to math-related articles, because the value in the tags comes entirely from the assessment information. — Carl (CBM · talk) 13:30, 30 December 2009 (UTC)

There's also a counter-argument, that if we let a bot go through all the math articles and tag them with "maths rating", then interested editors who wish to assess them can do so very easily, by just going through the list of unassessed articles (in their field of expertise). Right now it's really hard for an editor to find math articles in field X which aren't tagged by "maths rating". --Robin (talk) 16:04, 30 December 2009 (UTC)

The current activity page doesn't include everything (yes, I monitor that as well). For example the articles with copyright problems doesn't show on the current activity and we've had quite a bit of activiity there recently.--RDBury (talk) 11:52, 31 December 2009 (UTC)

Re RobinK: there are about 16,000 articles that are on the list of mathematics articles but not tagged with {{maths rating}}. Experience says that a backlog that big will simply sit forever. That is, if a bot tagged 16,000 articles, we do not have the volunteer time to go behind it and fill in the missing information. We already have a backlog of 1,500 articles that I am hoping to work through over a period of several months.

However, I can definitely make a tool to list mathematics articles in a particular field that are not tagged. — Carl (CBM · talk) 12:59, 31 December 2009 (UTC)

Re RobinK: I whipped up two tools for you this morning:

• nomathsratingcat - browse the category tree looking for articles without the maths ratings template
• mathsratingcat - browse the categorytree looking for articles with incomplete maths ratings

All the numbers reported are from the live toolserver database. There is often a small amount of lag between the toolserver and the production database used for the actual site, but changes should be reflected very quickly overall.

None of the numbers shown is recursive. That means that the number only refers to articles that are directly in the corresponding category.

I realize these tools are not yet ideal, but I wanted to get something working right away. I'm happy to take feedback and improve the tools if possible. These are really just a proof of concept. — Carl (CBM · talk) 14:07, 31 December 2009 (UTC)

P.S. I do think these two tools can be combined into one tool, once I am sure there are no performance issues. — Carl (CBM · talk) 14:40, 31 December 2009 (UTC)

They're great tools, thanks! I used them for a while, and I see no problems. I've taken care of most of the incompletely rated articles in Discrete mathematics and Theoretical computer science. Can't think of any improvements, except adding recursive scanning, but seems fine even without recursion. Any way for me to know how many articles are left? Like where did you get the number 16,000 from? It's not necessary to know this, but it adds to the feel-good factor if you see that you've brought down the number from 16,000 to 15,900 or something. --Robin (talk) 16:01, 31 December 2009 (UTC)

On an unrelated note, how does the tool "Find articles without maths ratings by category" work? Does it intersect some categories? (For example, when I search in category complexity classes [3], I get exponential time but not pseudo-polynomial time, even though both are in exactly the same categories and have no WikiProject templates on their talk page.) --Robin (talk) 19:14, 31 December 2009 (UTC)

The tool only considers articles that are listed on (subpages of) List of mathematics articles. I don't know why exponential time is listed there but not pseudo-polynomial time. EXP was added back in 2005, probably because it had Category:Exponentials at the time. I don't know the exact algorithm that the bot uses to update the list of math articles (in particular, I don't know the criteria for removal). Oleg Alexandrov would know more details.

To get 16,000, I used the count of 23,000 from [[Wikipedia:WikiProject Mathematics/Count] and the count of 7,000 at Wikipedia:WikiProject_Mathematics/Wikipedia_1.0. I could certainly add an overall count of articles with no maths rating tag, although it will be very large. I think it should be efficient to generate.

Recursive scanning is somewhat more work to implement. But, because there can be a lot of semantic drift as you get deeper down the category tree, I think it would make more sense to make a better interface for navigating the tree (maybe by adding a history so you can see how you got to the location you are currently at). A limiting factor is my development time; my first WP priority right now is finishing the new WP 1.0 bot, so I can't spend too long developing a new assessment tool until that is done. The tools I whipped up yesterday could be written quickly because they didn't involve too much complexity. If anyone else is interested in working on code for an assessment tool, I can help walk them through the process of getting a toolserver account. — Carl (CBM · talk) 13:04, 1 January 2010 (UTC)

OK, thanks. I think the tool is pretty usable as it is. I managed to assess quite a few articles yesterday. --Robin (talk) 15:33, 1 January 2010 (UTC)

Pierre Deligne

Tiresome POV pushing by someone concerned to denigrate the achievement of Deligne in completing the proof of the Weil conjectures. Please help revert. Charles Matthews (talk) 10:25, 31 December 2009 (UTC)

I have reverted his/her most recent revert. However, if he/she continues to be disruptive, WP:3RR applies, and he/she could be blocked. This should be considered should he/she revert once more. --PST 10:40, 31 December 2009 (UTC)

There's a discussion going on here. (He edited Charle's user page, rather than his talk page, so I moved the reply from there). Pcap ping 10:44, 31 December 2009 (UTC)

He already broke 3RR, if he continues, just report him with these diffs: [4] [5] [6] [7] [8] . I need to go offline soon. Pcap ping 10:51, 31 December 2009 (UTC)

I have reported him. The user has reverted 5 times, and has only responded once to our requests (and not waited for a reply before reverting again). This exceeds the limit for WP:3RR by at least one edit. Thanks for the diffs; I have used them. --PST 11:57, 31 December 2009 (UTC)

He replied on your user page as well, and then he reverted himself. I think he's trying to communicate in good faith, he's just confused about the multitude of pages here. I did my best to consolidate the discussion on his talk page. Pcap ping 13:02, 31 December 2009 (UTC)

I think that you may be right, and hopefully the block does not discourage him from editing Wikipedia. However, it did not seem that pure communication could prevent him from reverting; I think that he was unfamiliar with WP:3RR. As a consequence, I have apologized to him on his userpage; he may not have deserved the block, but Wiki rules do apply. Perhaps my apology may remove heat from the matter. --PST 13:23, 31 December 2009 (UTC)

Comment requested at Talk:Legendre polynomials and Talk:Associated Legendre function

Could someone please comment on the suitability (or unsuitability as the case may be) of including detailed proofs of the Rodrigues formula in the articles Legendre polynomial and Associated Legendre polynomial, as was recently done in these edits, respectively: [9], [10]. I am sorely tempted to remove both proofs as essentially trivial calculations from the Leibniz rule for repeated differentiation. This seems to be the kind of proof that our project has decided against including. I am not opposed to adding something like this which delivers the main idea behind the proof without giving step-by-step details. I have tried to solicit some input on Talk:Legendre polynomials, but it has been a few days and so for it has not elicited any discussion. On the contrary, the new proof has just now been added to Associated Legendre function. I would like an outside opinion on what to do before proceeding with my own preference: remove the added sections, and replace them with a short summary of the proof. I would do this myself, but I have had other rather abrasive interactions with User:Stamcose, and an outside opinion might help to defuse the situation. Sławomir Biały (talk) 17:24, 31 December 2009 (UTC)

Yes, a dull proof to write out in detail, OK as a thumbnail "applying the Leibniz formula to ... ". Charles Matthews (talk) 14:42, 1 January 2010 (UTC)

What about replacing the proof with a short summary, and then creating new pages (Legendre function/Proofs) which gives the proof in full detail? Ulner (talk) 15:00, 1 January 2010 (UTC)

The '/Proofs' idea is against WP policy; the articles namespace does not allow subpages. There are no references given so it might be better to check to see if this is OR before deciding other issues. OR proofs can be valid but there might be simpler ones available in the literature.--RDBury (talk) 15:17, 1 January 2010 (UTC)

It's pretty obviously not OR, but suffers from the opposite problem: it is painfully routine. Sławomir Biały (talk) 15:34, 1 January 2010 (UTC)

Why can't something be OR and routine? If it's just plugging in the definitions and turning the crank then all the more likely that it is OR. If it's proved as a special case of a general theorem or using some clever trick then the more likely that it's from a published source. That's been my experience anyway.--RDBury (talk) 15:54, 1 January 2010 (UTC)

I agree that subpages are currently disallowed, and propose insread to use 'Legendre function (Proofs)' as a separate page instead. The following pages should be moved accordingly: Christoffel symbols/Proofs, Approximation theory/Proofs, Vector Laplacian/Proofs, Boy's surface/Proofs. Ulner (talk) 16:28, 1 January 2010 (UTC)

You're basically proposing subpages with parentheses instead of a slash. The point of doing the subpages was to get around needing to have proof pages satisfy WP:GNG and changing the name won't solve that. This is a complex issue and the previous consensus was that these articles need to be handled on a case by case basis. I'm currently working on this (anyone who wants to help is more than welcome) and you can find the current status at User:RDBury/Proofs articles for cleanup--RDBury (talk) 18:40, 1 January 2010 (UTC)

If there is nothing more to Legendre function (Proofs) than such proofs that were routine by about 1850, I would propose it for deletion. Wikipedia doesn't include pages on miscellaneous calculus exercises, even though you could undoubtedly reference such pages to multiple standard texts. This all falls under "Wikipedia is not a collection of miscellaneous information", really. The topic of a page has to have independent status as something adding to the encyclopedia's coverage. Charles Matthews (talk) 18:55, 1 January 2010 (UTC)

I agree with the arguments given here; new subpages should in this case not be created. Ulner (talk) 18:58, 1 January 2010 (UTC)

After rethinking, I suggest to follow Sławomir Biały's suggestion: rewrite the proof to only provide a short summary of the argument (with a reference to (preferably) some textbook where the complete proof can be found or (if not possible to find textbook/research article) to some wikibook). Ulner (talk) 19:02, 1 January 2010 (UTC)

FYI, links in Proofs talk page

I added links to threads from this page to the Wikipedia talk:WikiProject Mathematics/Proofs since it seems that much of the relevant discussion takes place here (previous thread included). There are probably more than I have listed, but it's a start. Look in the archives box.--RDBury (talk) 22:19, 1 January 2010 (UTC)

Recognized Content subpage

There is a new recognized content bot that the physics project has signed up for, see Wikipedia:WikiProject Physics/Recognized content. Any thoughts on signing WPM up as well?--RDBury (talk) 12:42, 31 December 2009 (UTC)

We have a list directly on WP:WPM (and not nearly as many articles). Still, if the bot just updates a subpage of WPM and you're interested in it, I can't see any reason not to sign up for the bot. — Carl (CBM · talk) 12:53, 31 December 2009 (UTC)

I can't imagine anyone objecting to having that subpage, so I've gone ahead and created it at Wikipedia:WikiProject_Mathematics/Recognized_content. Of course it can be moved/renamed if people prefer a different name/location. I've used the same parameters as the physics page, but someone can double check to make sure I haven't missed anything. I haven't added "features pictures" and "featured portals"; I'm not sure if we have any of those on WPM. --Robin (talk) 18:10, 31 December 2009 (UTC)

It looks like it populated ok and I added a link from the project page. I think the main advantage is that it's populated automatically. On the other hand some of the articles under FYI are a bit questionable so maybe it's better to have a human to filter them.--RDBury (talk) 13:56, 1 January 2010 (UTC)

It might even be a good idea to transclude the recognized content page on the main page, like they have on WikiProject Physics, for instance. I think the bot formats the page quite well, so it would look pretty neat on the main project page. The only problem right now is that the bot makes all the heading level 2 headings, which wouldn't work on the main page, since they should be level 3 headings, but I think the bot's creator is working on that problem. Assuming that problem does get fixed, do people want to have this page transcluded on the main page? --Robin (talk) 15:47, 1 January 2010 (UTC)

Wow that is really cool. If I recall correctly, we do indeed have a few featured pictures. Is it possible to add them to the list? Cheers, Ben (talk) 14:25, 1 January 2010 (UTC)

Doesn't seem to be one of the parameters but I'll request it.--RDBury (talk) 15:21, 1 January 2010 (UTC)

I think the bot can handle featured pictures too. I'll add it to the page so that they'll get added on the next round of updates. --Robin (talk) 15:35, 1 January 2010 (UTC)

I added the features picture option, but the bot didn't add any when it updated the page. I think the reason is that featured pictures are not tagged with the "maths rating" template, whereas the bot's algorithm looks for the this template. --Robin (talk) 15:54, 1 January 2010 (UTC)

The parameter is there, I missed it earlier, so most likely you're right in that the maths rating tags need to be added. See User:JL-Bot/Project content/documentation for the documentation.--RDBury (talk) 18:51, 1 January 2010 (UTC)

(outdent) I added the tag for ten featured images so hopefully they will show up in the next refresh.--RDBury (talk) 19:09, 1 January 2010 (UTC)

The pictures are there in a nice gallery.--RDBury (talk) 12:48, 2 January 2010 (UTC)

Looks good here. Thanks for that RDBury. Ben (talk) 13:58, 2 January 2010 (UTC)

User:Indefiniteintegral

This editor has created three math articles, Leibniz function, Leibniz differential, and Fermat differentiation. Two of these are currently PRODed and the third is dubious at best. All three use the same calculus textbooks as references. My question is, is it a coincidence that these three subjects appear in the same two calculus books and nowhere else or is this person creating hoax articles? The non-math articles this editor created are also dubious.--RDBury (talk) 15:26, 2 January 2010 (UTC)

These three subjects do not even seem to appear in the cited "Advanced Calculus Demystified" source. I am unable to check the Abbott and Neill source. I support the prods, as well as closer scrutiny of the editor's other contributions. Sławomir Biały (talk) 16:10, 2 January 2010 (UTC)

Thanks, just wanted to make sure it wasn't just me. This user turns out to be a sockpuppet of a banned user; I did a quick check on articles created by other aliases but not other articles created turned up. Some review of major edits might be worth while.--RDBury (talk) 18:44, 2 January 2010 (UTC)

De Moivre's formula

An editor keeps trying to say the De Moivre's formula‎‎ is true for non-integer powers and stick in their own proof. The problem is they have a copy of a Schaum's outline backing this error though it says the proof is outside the scope of this book. Any idea how to explain things to them better?, I put in a bit into the article showing why it was wrong for non-integer powers but it obviously doesn't convince them. Dmcq (talk) 19:04, 2 January 2010 (UTC)

The problem may be that there is always a value of the non-integer power for which de Moivre's theorem is true. Septentrionalis PMAnderson 04:28, 3 January 2010 (UTC)

I'll see if I can't put a link to the generalization section in the bit only being true for integers or somewhere else near the top. Dmcq (talk) 07:09, 3 January 2010 (UTC)

Discussion on Trott curve

There's a slow but ongoing discussion on Talk:Trott curve regarding whether the subject is sufficiently notable for a standalone article, should be moved to a more general article concerning the (much older and more famous) result that quartics have 28 bitangents rather than this specific quadric, should be merged with bitangent, etc. Please participate. —David Eppstein (talk) 19:59, 3 January 2010 (UTC)

Saturated model

There was a minor kerfuffle, more or less resolved now, at saturated model, where an editor put {{technical}} and {{context}} tags on the article. It is certainly true that the article could stand better contextualization (Hans, maybe you'd like to take a crack at it?).

But that's not the main point — as it turns out the editor thought the article was about something completely different, a concept from something called structural equation modeling, about which I know nothing. Also when I look at "what links here" from saturated model, one of the links is from friction, which is certainly not referring to the model-theory concept. Whether it's referring to the SEM concept, or to still a third one, I don't know.

Anyway the point is that this/these concept(s) could probably use a writeup. Does anyone know anything about it/them? --Trovatore (talk) 23:59, 3 January 2010 (UTC)

The one from friction appears to be a third concept. I could probably write a stub on the statistical concept. Michael Hardy (talk) 02:47, 4 January 2010 (UTC)

Well, this kind of kerfuffles keep us sharp. Anyway in the meantime I have found (I think) a basic definition of what a saturated path model is: A path model having an equal number of (non-observed) parameters as it has observed variables [needs some extra tweaking]. In this case the model has zero degrees of freedom (df=0). Every covariance matrix (of the right size for the model) can then be reproduced using the model and all standardized residual covariances equal zero. I hope I got this right, my teacher didn't make it all too clear, I had to reconstruct it from disjointed fragments. It'll need better phrasing and stuff, but still.

I think it would also be a good idea to make at least a stub describing a path model. --JorisvS (talk) 11:35, 4 January 2010 (UTC)

All-ones vector

There is a new article All-ones vector with no references or evidence of notability; at first glance it appears to be OR. The term is used in the article Boolean algebras canonically defined without definition, but perhaps it was assumed to be self-explanatory. It seems to me if the term is self explanatory then the new article should be deleted, if not then the definition should be in the Boolean algebras article and the new article should be a redirect. In either case, a new article is not justified. I'm not an expert in Boolean algebras however, so I don't think I'm qualified to make the call, but it would be helpful if someone who has some expertise in the area could either add the definition to the Boolean algebras articles or indicate that it's not needed. Also, 'Boolean algebras canonically defined' is a very strange name for an article and probably violates naming criteria, any chance that could be resolved as well?--RDBury (talk) 13:01, 30 December 2009 (UTC)

I prodded it as (mathematical) WP:DICTDEF. Pcap ping 13:13, 30 December 2009 (UTC)

It's related to this WP:RD/MA#The all-ones vector, and how to notate it. The editor who asked the question seems to have taken the unusual step of turning that unresolved query with little interest in it into an article. So yes, un-notable OR by any reasonable measure. --JohnBlackburne (talk) 13:28, 30 December 2009 (UTC)

Not sure if the query was turned into an article or the query was to help write the article. Either way, it's probably not a good idea to be creating an article if you're dependent on the ref. desk for material.--RDBury (talk) 12:02, 31 December 2009 (UTC)

The prod on this was declined by an uninvolved admin, so Wikipedia:Articles for deletion/All-ones vector. Pcap ping 20:50, 6 January 2010 (UTC)

Any idea why this isn't working?

Wikipedia:WikiProject Mathematics/Wikipedia 1.0/Analysis and Wikipedia:WikiProject Mathematics/Wikipedia 1.0/Discrete mathematics don't seem to be working, but the other fields like Wikipedia:WikiProject Mathematics/Wikipedia 1.0/Geometry work fine for me. --Robin (talk) 05:52, 5 January 2010 (UTC)

It's because the pages grew too large. The hidden category on the broken pages was a symptom of this. I added a parameter to the templates on those two pages that makes them work again, by not putting so much data on one page.

The new WP 1.0 bot is going to include the ability to generate the per-field tables, and I was working on that just yesterday. I added talk page categories for the fields a couple weeks ago, so you can already use the lists feature of the new WP 1.0 bot to make a list of articles in a particular field (e.g. Analysis). This should also make it easier to use other tools that are based on category intersection. The list of categories is at {{maths rating}}. — Carl (CBM · talk) 11:04, 5 January 2010 (UTC)

Ok, thanks. --Robin (talk) 14:15, 5 January 2010 (UTC)

RfC on recent changes to Bivector

I've attached a RfC to Talk:Bivector#Problems with second paragraph, but as it's primarily a mathematics issue I thought I'd mention it here too. It's just me and User:Brews ohare working on the article who seem never to see eye-to-eye on anything, so we could do with other viewpoints. --John Blackburne (words ‡ deeds) 17:01, 6 January 2010 (UTC)

Computational complexity theory as part of "mathematics"

I noticed that the list of mathematical articles had some complexity theory articles, like P versus NP problem, Cook–Levin theorem, Karp's 21 NP-complete problems and PCP theorem, but does not have articles P (complexity), BPP or Karp–Lipton theorem. I realized that the reason for this is that Category:computational complexity theory is not part of list of mathematics categories. I would like to propose that Category:computational complexity theory be considered as "mathematics" on Wikipedia and this category be added to list of mathematics categories. I personally believe that complexity theory is mathematics. The biggest open question of complexity theory , the P versus NP problem is considered to be one of the biggest open problems in mathematics, not just computer science. Also, for example, The Princeton Companion to Mathematics lists computational complexity theory as one of the branches of mathematics. (See table of contents which lists 26 branches, and complexity theory is the 3rd one on the list.) What do people feel about this? --Robin (talk) 03:57, 4 January 2010 (UTC)

If we put Category:computational complexity theory in list of mathematics categories, then mathbot will put all the articles in that complexity category into the list of mathematics articles. There is a lot of non-mathematics there, such as Electronic Colloquium on Computational Complexity, Molecular models of DNA, etc. The whole category smacks of hard computer science to me. Perhaps a few articles from there can be added by hand to the list of mathematics articles, rather than having the bot automatically add any articles there to the list. Oleg Alexandrov (talk) 04:03, 4 January 2010 (UTC)

True, but such stuff always creeps in when we add any sufficiently large category to list of math categories. For instance, Category:Graph theory brings with it articles like Small world routing and Liquid schedule, which seem quite non-mathematical. Moreover, the current list of mathematical categories does include categories that I would think are way more computer science than complexity theory. E.g., Category:Arbitrary precision algorithms, Category:FFT algorithms, Category:Computational geometry, Category:Optimization algorithms and Category:Dynamic programming, which leads to articles like Artificial Bee Colony Algorithm and Swarm intelligence. I'm not saying WP:Other stuff exists, but some primarily math categories will inevitably bring along some non-math, and this is not specific to only Category:computational complexity theory. --Robin (talk) 04:29, 4 January 2010 (UTC)

Molecular models of DNA was a bad example — it was heavily overcategorized and should not have been in the category in the first place. I removed it. —David Eppstein (talk) 05:02, 4 January 2010 (UTC)

I support adding that category to List of mathematics categories. CRGreathouse (t | c) 14:37, 4 January 2010 (UTC)

There's a gazillion of non-math in there, such as AI-complete, DNA Dynamics, Randomness extractor, Circuit complexity, Unary language, etc. Besides, the article Computational complexity theory itself clearly says that this is part of computer science.

Adding this category to the list of math categories is not a big deal, but who will regularly spend time inspecting what articles the bot adds and removing the non-mathematical ones? That is hard enough with mathematics categories, but now we are talking about adding a huge category which is only marginally related to mathematics. Oleg Alexandrov (talk) 19:09, 4 January 2010 (UTC)

DNA Dynamics was another badly overcategorized article that should have not been in that category at all. Unary language should already be in the math hierarchy by virtue of Category:Formal languages. Randomness extractor and Circuit complexity are no more nor less mathematical than the rest of the category — there is no reason to include the rest of the category and manually exclude them. AI-complete is a bit of a border case (and a questionable inclusion in the category) but we're never going to have a system that avoids border cases and I think the number of them is quite small. —David Eppstein (talk) 01:57, 5 January 2010 (UTC)

Indeed, Randomness extractor and Circuit complexity are standard complexity theory articles, and are examples of exactly the sort of articles I'm proposing we include in the list of mathematical articles. --Robin (talk) 02:41, 5 January 2010 (UTC)

I guess this is dvelving into the ideological and pragmatic divide between mathematics and TCS. There will obviously be different opinions on dealing with this. I had a long discussion with User:Arthur Rubin in the past on dealing with Category:type theory, which contains mostly topics of interest to computer scientists nowadays. I think the pragmatic approach is to ask: which topics of TCS (computational complexity in this case) are generally of interest to mathematicians? And add just those articles to the math lists, possibly creating a helper category. Pcap ping 14:08, 5 January 2010 (UTC)

I think it's more of a philosophical question: does the math hierarchy include only mathematics of interest to mathematicians in research mathematics departments (in which case for the math department in my university much of combinatorics would be excluded) or does it include as well subjects such as statistics, computational complexity theory, etc. that are treated by their practitioners as rigorous mathematical subjects but that are largely sorted into other departments in our research universities? The Clay math prizes obviously took an inclusionary approach, including problems from complexity theory (P vs NP) and fluid dynamics in their million dollar prize problems. But we don't have to follow their lead if we want to be more exclusionary and focussed. —David Eppstein (talk) 15:57, 5 January 2010 (UTC)

I have no problem adding this category. I'll ask more folks to comment though. Oleg Alexandrov (talk) 17:09, 5 January 2010 (UTC)

As far as I am concerned, anything that can be formalized as a theorem of ZFC is mathematics (not necessarily vice versa). The bulk of computational complexity is mathematics, hence inclusion of this category is quite appropriate. — Emil J. 17:45, 5 January 2010 (UTC)

In addition to what David Eppstein said, besides the Clay math institute, many institutes and people share a similar belief. For instance, Avi Wigderson, who is a complexity theorist, is a faculty member of the school of mathematics [11] at the Institute for Advanced Study, Princeton. And as I said before, the fields medalist and editor of The Princeton Companion to Mathematics, Timothy Gowers, lists computational complexity theory as one of the branches of mathematics. --Robin (talk) 18:40, 5 January 2010 (UTC)

I'm comfortable calling most of these articles mathematics.

However it might be good to ask at WikiProject CS and see what they think. It can come across as arrogant to start categorizing things as part of one's own field, if the workers focused on that field don't see it that way. There's a very irritating contributor who every now and then changes the opening sentence of set theory to make the bizarre claim that set theory is a branch of computer science. A few cybernetics editors have a similar overreach. I wouldn't want us to be seen as engaging in the same sort of imperialism. --Trovatore (talk) 18:48, 5 January 2010 (UTC)

Categories are devices for finding articles you do want? False negatives should be more of a worry than false positives, really. Go back six years and there was plenty of physics listed as mathematics. It was a good idea to weed out those articles, but it is much less clear that computer science connections should be treated in the same way. Save energies for who gets tenure :-( Charles Matthews (talk) 20:53, 5 January 2010 (UTC)

Just a tangent on that: Certainly that's part of what categories are (although I personally rarely use them to find articles — would be interested in hearing others' practices on this).

But categories also, in practice if not by design, take a position on how human knowledge is structured. That can make them targets for POV pushing, especially given that there's no such thing as a citation for a category.

The various pseudo categories, such as category:pseudoscience, are especially problematic in this regard. --Trovatore (talk) 21:30, 6 January 2010 (UTC)

These topics are not math. If wrong categories are applied to an article, they should be removed. Mis-categorization deteriorates the information value to a reader. There is some overlap between CS and math, such as numerical analysis (linear and non-linear algebra) and lambda calculus. Oddly, neither of those two topics is listed in any math category. It looks like we have some work to do. Jehochman ^Brrr 21:13, 5 January 2010 (UTC) (I supposedly know something about theory of computation.[12])

Lambda calculus is in category:recursion theory, which is a mathematics category.

But I'm not sure which topics exactly you mean by these. Some of them are definitely mathematics by my lights (pretty much everything about the polynomial-time hierarchy, for example). --Trovatore (talk) 21:54, 5 January 2010 (UTC)

Considerations of what else is in what categories aside, I would argue that computational complexity theory belongs as much as, say, algeabraic topology does. don't be decieved by the "computational" part of computational complexity theory. computational complexity theory makes rigorous mathematical statements about, well, mathematical statements. Part of the beauty is, like algeabraic topology, those statements apply regardless of what language one chooses to express their mathematical statements in; regardless of the formal system. It can tell you how many times a symbol is traversed in relation to the number of, say, vertices of a graph. Much like things in set theory or geometry do. It's a mathematically rigourous way of analyzing a formal structure - or one might say a set of formal structure spaces. point is, it is pure mathematics at its best. Kevin Baas^talk 21:45, 6 January 2010 (UTC)

Computational complexity theory added, help requested

I added Category:computational complexity theory to the list of mathematics categories, per the comments above. At the next bot run it will add the articles from that cat to the list of mathematics articles.

It would be nice if folks knowledgeable in this area could watch User:Mathbot/Changes to mathlists or Wikipedia:WikiProject Mathematics/Current activity to see the articles added by the bot due to this change, and fix the category of the non-mathematical articles coming in (as well as removing those articles from the list of math articles).

Also, which subcategories of Category:computational complexity theory should be adopted by our project as well? Oleg Alexandrov (talk) 02:03, 6 January 2010 (UTC)

What do you mean by a "non-mathematical article" — do articles about journals in specific areas of mathematics, or about mathematicians, get counted as mathematical articles, for instance? I am asking because to me the answer is obviously yes but there was a comment above about Electronic Colloquium on Computational Complexity not counting and I'm unsure why not. Also, what is the actual process for removing an article from the list of math articles?

As to what subcategories to also include: probably Category:Circuit complexity, Category:Complexity classes, Category:Computational resources, Category:Descriptive complexity, Category:Probabilistic complexity theory, Category:Probabilistic complexity classes, Category:Quantum complexity theory, and Category:Structural complexity theory. Everything in those categories looks like it could fit equally well within the parent category. But I'm not at all sure about Category:Computational problems and Category:Analysis of algorithms — the former is big and branches off in a different direction than the rest of the complexity theory hierarchy, and the latter really does start to look like CS and not math to me — it's less about formal proofs and more about how one defines mathematical models of the behavior of real computers. —David Eppstein (talk) 17:08, 6 January 2010 (UTC)

I agree. Category:Circuit complexity, Category:Complexity classes, Category:Computational resources, Category:Descriptive complexity, Category:Probabilistic complexity theory, Category:Probabilistic complexity classes, Category:Quantum complexity theory, and Category:Structural complexity theory are standard complexity theory categories, and would fit equally well in the parent category. Category:Computational problems and Category:Analysis of algorithms are a tough call, since the articles in those don't really fit in the parent category. Probably best to not include those two. --Robin (talk) 18:56, 6 January 2010 (UTC)

I guess an article about a mathematical journal is considered mathematical. The mathematicians have their own list. If the bot adds an article which is obviously non-mathematical to the list of mathematics articles, one needs to remove it from any mathematical category, as defined at the list of mathematics categories. Then, one needs to remove the article by hand from the list of mathematics articles. Oleg Alexandrov (talk) 18:52, 6 January 2010 (UTC)

I added the subcategories suggested above. It would be nice if people monitor to see what articles the bot will add from these categories at the next run. Oleg Alexandrov (talk) 22:36, 7 January 2010 (UTC)

Information theory

A less extensive discussion of this kind took place maybe three or four years ago, concerning Category:Information theory. Many articles in that category are mathematics because they're information theory; others apparently are not mathematics. Michael Hardy (talk) 17:45, 6 January 2010 (UTC)

Category:Mathematical software and subcats not watched by the mathbot?

I see that Logiweb, which is indirectly part of that category (via Category:Theorem proving software systems) is not being listed in the recent activity for deletion. Is this by design? Pcap ping 23:38, 4 January 2010 (UTC)

It looks like the way it works is there are mathematics categories and mathematics-related categories. Articles in categories on the first list get put into the list of mathematics articles by mathbot and thence to the activity page by a different bot if there is activity. Mathbot ignores categories in the second list. Logiweb is in two categories, Category:Theorem proving software systems and Category:Mathematical software which seem like they might go in the first list, but small changes in the categories tend to made large and sometimes unexpected changes in the lists of articles so care should be used in making any changes. I could be wrong about how it works though; it seems like it should be documented somewhere but I didn't see anything about it except in the code itself. It's a bit disconcerting that we've got WPM specific bots for this, it seems like there should be more universal tools available like the new recognized content bot. It's also a bit disconcerting that Logiweb doesn't appear on any current activity lists. It seems like there shouldn't be any categories that fall through the cracks but it looks like Logiweb is in a couple.--RDBury (talk) 19:04, 7 January 2010 (UTC)

Actually I did find a universal bot you can subscribe by adding the template ArticleAlertbotSubscription to the project page, but articles need to have the WPM banner in the talk page for it to work.--RDBury (talk) 19:55, 7 January 2010 (UTC)

I don't find it disconcerting: we have specific needs, and so it makes sense to use specific bots for them. You're correct that there is one bot that updates the list of mathematics articles, and another that uses that list to check for daily activity. But in general the list of math articles is much more complete than any manually-generated list would be, because we catch new articles automatically as soon as they are put into a category that is on the list.

It seems to me like we should do an audit of mathematics-related categories to see which of them should be added to the list that is watched by the bot. This was done by hand for a while by the bot operators, I believe, but we can't expect them to commit to doing it indefinitely. — Carl (CBM · talk) 20:23, 7 January 2010 (UTC)

I am adding categories to the list of mathematics categories on a regular basis. See User:Mathbot/New math categories for suggestions. Any help here would be very much appreciated. Oleg Alexandrov (talk) 22:32, 7 January 2010 (UTC)

I'm not sure how specific needs is applicable here. My understanding is that the bot maintaining the List of mathematical articles basically creates a union of a manually updated list of categories. It seems to me that many projects could use a tool like that. It's true we get new article notification quickly, but we also get a lot of false positives and every once in a while there will be days where 20 or 30 new articles show up because a category was moved. I do occasionally try to attack the backlog of Wikipedia:Pages needing attention/Mathematics, but it's suprizing how many of these aren't really mathematical. Do we really want to be responsible for maintaining Squoval for example? Documentation seem to be the most pressing need though. I've been looking at this for a while and am just beginning to understand how it works and unsure about whether by understanding is correct. As to how to respond if someone says article X obviously is/isn't a math article but it isn't/is on the list, I'd still have no idea.--RDBury (talk) 16:58, 8 January 2010 (UTC)

If you see a non-math article that is on the list of math articles, you need to remove it. To make sure that the bot does not add it back, you need to remove that article from a math category. Looking at mathbot's page can clarify a few things as well. Oleg Alexandrov (talk) 17:18, 8 January 2010 (UTC)

That's the kind of thing that needs to be documented so people don't have to spend hours digging through code to figure it out. I think the reason Squoval is there is because it was at some point in Category:Curves, which is legitimate because it is a curve. But that category gets added to the math articles and that's as expected because most curve topics are about mathematical curves.--RDBury (talk) 13:17, 9 January 2010 (UTC)

(Actually, it was in Category:Geometric shapes: [13].) — Emil J. 15:51, 9 January 2010 (UTC)

Quick note: the subcat Category:Spreadsheet software should definitely not be added. Pcap ping 17:30, 8 January 2010 (UTC)

Book:Derivatives

Please give feedback concerning the title of this book on the book's talk page. Headbomb {^ταλκ_{κοντριβς} – WP Physics} 04:53, 8 January 2010 (UTC)

Misleading title :( Robin (talk) 05:01, 8 January 2010 (UTC)

Which is why there needs to be feedback! Headbomb {^ταλκ_{κοντριβς} – WP Physics} 05:35, 8 January 2010 (UTC)

BackFed :) -- Avi (talk) 05:43, 8 January 2010 (UTC)

Definitely a misleading title. Put brain in gear. Then you won't need further comments beyond that. Michael Hardy (talk) 06:09, 8 January 2010 (UTC)

To what extent are original research math discussions tolerated in user talk?

Since I don't have the patience to check if most of the stuff discussed there is indeed WP:OR, please see Wikipedia:Miscellany for deletion/User:Linas/Lattice models. Pcap ping 10:30, 9 January 2010 (UTC)

A lot is tolerated in user space. As long as the original research isn't being injected into articles or article talk pages, we ought to leave them be. Jehochman ^Brrr 10:43, 9 January 2010 (UTC)

I think technically the rule is there shouldn't be OR anywhere on Wikipedia, but I doubt anyone bothers to enforce it in user talk. There are different kinds of OR though, e.g. one user telling another about an new proof of quadratic reciprocity is probably ok, but making libelous statements about someone is bad no matter where you do it. Basically, user talk <> e-mail and anything written there can be read by the general public, so anything potentially harmful to Wikipedia or a user or anyone else should removed.--RDBury (talk) 12:44, 9 January 2010 (UTC)

I believe you are quite mistaken here; WP:OR only applies to mainspace pages and was never intended for projectspace, such as article talk pages, user talk pages etc. In some sense much if not most of what goes on in projectspace involves OR, as well it should. Certainly all the policy and guideline pages themselves are more or less pure OR and there is nothing wrong with that. The problematic conduct in projectspace of the type you mention (such as libelous statements) is ruled out by other policies (not by WP:OR), such as WP:BLP, WP:NOT, WP:CIVIL and the like. Nsk92 (talk) 01:20, 11 January 2010 (UTC)

I don't see how the policy against OR could apply to user talk pages. Those discussions generally should be about editing Wikipedia, but that can cover a lot of ground. Michael Hardy (talk) 19:02, 9 January 2010 (UTC)

Even on an article talk page rather than a user talk page I think some amount of OR should be tolerated, in the form of "I can see that such and such a fact is true — does someone know whether this is OR or whether a source can be found so that we can add it to the article". On the other hand, there is a general rule that material on all pages here (User or whatever) should in some way be about advancing the encyclopedia; this is not so much in WP:OR but WP:NOT. —David Eppstein (talk) 19:24, 9 January 2010 (UTC)

There are some fine lines here. For example when you see editors arguing over something you thought the outside world was in some consensus about, you could point harder to sources for that consensus. But sometimes you might want to bring some unity or clarity to those sources by tying them together in the talk page. At that point the question arises as to how effectively you're allowed to do that. If the answer is "not very" then it makes it harder for Wikipedia to present a coherent picture of a scattered consensus.

As a case in point I felt that the mainstream consensus in the Sokal affair was running against Sokal, and made an effort on its talk page to tie together in one coherent story the many scattered thoughts on it that I'd seen. Although only one person objected to my doing so (I also had positive feedback about it) I could see the logic behind his objection. However I could also see a meta-argument for my argument and gave it there. --Vaughan Pratt (talk) 21:00, 9 January 2010 (UTC)

The OR policy applies to articles, portals, templates, etc., that are supposed to be part of the encyclopedia. It seems very strange to apply it to any kind of talk pages. Talk pages are for discussion of how to improve Wikipedia. Outlawing particular topics a priori in such discussions is not a good idea. Michael Hardy (talk) 01:04, 11 January 2010 (UTC)

I agree; there's no good warrant to apply a blanket ban to such discussion in userspace or even on article talk pages.

On the other hand there's a limit somewhere. I don't know exactly where. I argued keep at Wikipedia:Miscellany for deletion/User:Neptunerover/Theory About Everything, and I still think that would have been the right choice, but it was a reasonably close call, and unfortunately the user in question got everyone (including me) annoyed at him during the discussion. (To be honest, I'm not sure that wasn't his explicit intention.) --Trovatore (talk) 01:28, 11 January 2010 (UTC)

Bertrand's ballot theorem

An article I worked on extensively last year, Bertrand's ballot theorem, has acquired a 'Too technical' tag. I'm probably biased but I think it has just the right amount of technicality. I'm asking for someone who hasn't worked on the article too much to take a look at it and either remove the tag or offer suggestions as to how to make the language simpler without removing content. I'm thinking the article should be accessible to a bright high school student, though the proofs may require some additional mathematical sophistication. In the mean time I'll try tweak it a bit.--RDBury (talk) 14:46, 13 January 2010 (UTC)

I've also replied on the talk page there. Paul August ☎ 16:56, 13 January 2010 (UTC)

It is less technical than most articles in probability theory. Boris Tsirelson (talk) 18:03, 13 January 2010 (UTC)

I added an example section per Paul August. This is really a combinatorics article and the subject has been around since the 19th century when combinatorics was still in its infancy. The article isn't ready for FA status yet but I'm going to at least remove the tag unless there is an objection. 1 too technical tag down, 126 more to go.--RDBury (talk) 19:03, 13 January 2010 (UTC)

The version that was tagged as "too technical" was quite literally the sort of math that secondary school pupils work on. The person who tagged it was offended by the word "clearly" in a sentence that said that if one of the two candidates gets no votes at all, then clearly the other candidate will be ahead throughout the whole counting process. Michael Hardy (talk) 21:15, 13 January 2010 (UTC)

To be fair, that version had a proof in the lead section which is probably not a good idea. I also did a bit of rearranging which seems obvious to me now but probably not a year ago. Thanks to everyone who looked at this.--RDBury (talk) 21:36, 13 January 2010 (UTC)

Unsigned infinity

The article Limit of a function uses ∞ to mean +∞, in other words a neighborhood of ∞ is a set of the form {x: x > c} where c is a fixed number. This seems to be common but by no means universal terminology. The textbook I learned analysis from, A First Course in Analysis by Protter and Morrey, uses three distinct symbols, ∞, +∞ and −∞, where neighborhoods of ∞ are sets of the form {x: |x| > c}. The unsigned infinity and the signed infinities correspond to two compactifications of the real line, namely the Real projective line and the Extended real number line. The sources I looked at seemed pretty evenly divided between the different notations, those at freshman calculus level or less seem to follow the +∞=∞ school while those intended for people specializing in math follow the other (see [14] for example). The article on the extended real number line allows ∞ to be used for +∞ when there is no danger of confusion but there are articles which confuse the two even when there is such danger. This issue would not be a problem if the two notations were consistent; I'm trying to work on the Asymptote article where it naturally comes up and it's turning into something of a roadblock, at least for me. For example the statement

$\lim _{x\rightarrow 0}{1 \over x}=\infty$

would seem to make sense but it is false in the +∞=∞ notation. On the other hand the statement

$\lim _{x\rightarrow 0^{+}}\log x=\infty$

is true in the other system but looks wrong. So my first question is, do people have a sense of whether one of these notations is predominant? If one is standard then it should be adopted as the standard for Wikipedia and the other should be relegated to a statement somewhere that it's used by some authors. My sense is that if you just looked at calculus books it would look like +∞=∞ is standard notation but it you looked at higher level texts the other would seem to be standard. Certainly having ∞ distinct from +∞ is very useful in complex analysis. So my second question is, if there is no generally accepted standard can we come up with a convention to follow to minimize conflicting usage. This issue has come up before at Wikipedia talk:WikiProject Mathematics/Conventions#Infinity notation but there doesn't seem to have been a consensus. Even if there were a convention it seems like it would be necessary to explain it in every article where it's used since you can't assume that a reader will be familiar with it. But at least readers won't have to figure out from context what the meaning is supposed to be.--RDBury (talk) 18:14, 7 January 2010 (UTC)

Well, it isn't a notational question at all; it's a question of which structure you're working in. There isn't any "standard" answer to that. I guess I would say that tendentially the "unsigned infinity" is more likely to come up in a complex setting than a real-number setting; if you're in a context where you know you're dealing only with real numbers, and not with complex numbers even by implication, then infinities are expected to be signed, and therefore ∞ means +∞. --Trovatore (talk) 21:45, 7 January 2010 (UTC)

As you point out, even if you had a convention, you'd have to explain it in every article. I think this is unavoidable. Also, note that if you did want to write that the limit of 1/x as x → 0 was ∞, then all the time you'd have people misinterpreting ∞ as +∞ and "fixing" it. I think this is probably better said in words, e.g., "As x tends to zero, 1/x tends to get far from zero." If you need to make a precise statement, I think readers are more likely to understand you if you talk about 1/|x|. Ozob (talk) 23:47, 7 January 2010 (UTC)

(ec)Protter and Morrey barely mention complex numbers at all, but they use distinct symbols. So it's not a matter of real vs. complex. Also, if it's a matter of readers going in and fixing things, if half the people reading the article read the notation one way, and half read it the other way, there's potential for it to be "fixed" in either direction, as I was about to do with the Limit of a function article based on what I learned in college. Then I decided to check some other sources which led to me bringing this issue up again. To muddy the waters a bit more, Protter and Morrey somewhat inconsistently use the notation (a,∞) instead of (a,+∞) for {x: x > a}. The conclusions I'm drawing at this point are 1) ∞ appearing in a limit has two widely accepted interpretations, 2) neither meaning should be assumed implicitly in an article, 3) most readers will be used to the ∞=+∞ interpretation. For now, I'll try to update the Limit of a function article to represent both interpretations.--RDBury (talk) 14:16, 8 January 2010 (UTC)

Exactly; rather than assuming a convention, one could just say that as x goes to 0, the limit of |1/x| is infinity. That statement is much less likely to be misunderstood, because it is correct under both conventions.

It takes a little experience to get used to the sorts of misinterpretations that people make. My favorite example: there is some article about reasoning that said something like this:

The following statement is false:

3 is an even number.

Multiple editors changed that to "3 is an odd number" without bothering to read the context. There's some benefit in planning for this an writing in a way that makes it less common. — Carl (CBM · talk) 13:10, 8 January 2010 (UTC)

Well now I see two issues here, which I resolve thusly: 1/x as x->0, doesn't much matter which side you're coming from (0+ or 0-) When you actually reach 0, the line is discontinuous and so are all of it's derivatives. So to say that if you move a differential to the right away from the origin, the sign is positive, is disingenuous: the function doesn't have a differential at the origin! Now when you get things like f(x)/g(x) as x->0 where f(x)->0 and g(x)->0, (i.e. you have 0/0) you're kinda in a different situation, because L'hopital's rule states that you just keep taking derivatives of the numerator and denominator until you get something defined - and there's your sign. So whether the inverse of the top part (1/f(x) = 1/0) or the bottom part by itself (1/g(x) = 1/0) was approached from the "right" or "left" side; i.e. whether it was "negative" or "positive" infinity, doesn't matter at all because as soon as you put them together: f(x)/g(x), you get a definite sign no matter which side you come from on either 1/f(x) or 1/g(x). And since a negative times a negative equals a positive and so on, if you work it backwards you see that you can't really put a sign on the infinity when you have just 1/f(x) as x->0 by itself on account of which side you "approach" 0 from, because then it doesn't work out; then composition doesn't hold. so if you want the rules of mathematics to hold and be consistent, and f(x) - > 0 as x -> 0, then 1/f(x) as x -> 0 can't be said to have a sign regardless of which way you "approach" 0 from. In which case I always found the whole positive/negative infinity thing to be quite moot, if not just plain absurd.

Okay, now onto complex infinity: no such thing. you can break down a complex number two ways: real and imaginary part (in which case you have two real infinities), or magnitude and angle. now if you come to an equivalent situation to what i just described for complex numbers, well i would be tempted to say that your magnitude is "positive" infinity and your angle is undefined. but then again i would be just as well to say that your magnitude is negative infinity because well, your angle is undefined, anyways. point being your angle is now what encodes your sign so you might as well just screw the whole "sign of infinity" thing - your magnitude has no sign. and you don't have "complex infinity", you have normal infinity with an undefined phase angle (rho=infinity, theta=?). (Is suppose this is what one might mean by "complex infinity", in which cause ya i guess there "is" but saying so really wouldn't have grabbed your attention as well.)

Having said all that. I don't know what other interpretations there can be of signless infinity than, well, signless infinity. And the context should make it quite clear whether your infinity is signless or "positive". Though I would say that by the point that you start using negative infinity it's just good form to be signing your positive infinities anyways, for consistency's sake, if nothing else. So I say, do it: sign those infinities. But keep in mind that infinity or not, a negative times a negative should always equal a positive. A negative is just a half rotation in the complex plane and two half rotations equals a whole one. Infinity or not. if you can't guarantee that, then there's something wrong with your equation, IMO. I'm not an expert mathematician or anything like that but that's what I have to say about the subject. Kevin Baas^talk 21:58, 13 January 2010 (UTC)

Actually there's no problem with complex numbers. The extended complex numbers have only 1 infinity. There is no sign. The problem is with the reals due to different ways to extend the real numbers to include infinity (Real projective line and the Extended real number line). I don't have any useful suggestions to offer for the real number case. Just try to make it obvious from context, I guess. If not, perhaps try using notation that would be least confusing, or offer a short explanation? --Robin (talk) 22:49, 13 January 2010 (UTC)

Okay, so I see: the real projective line arises because of the whole wavelength = 1/frequency thing, as exemplified by things like 1/f(x) as x->0 and cosh(x). And the extended real number line is because, well, not all functions are like that. e.g. f(x)=x^2 (and maybe fourier transforming a function or something like that can move it back and forth between the real projective and real extended lines) Then I'd say sign any infinities on the extended real number line, and don't sign any on the real projective line. i.e. i reaffirm my "sign it" position. That would be "least confusing". (I suppose that's a simpler way to say my whole l'hopital's rule thing, because when signing an infinity would break composition, rest assured you're on the real projective line.) Kevin Baas^talk 15:29, 14 January 2010 (UTC)

looking at the asymptote page, I see them using +/- infinity for functions on the projective real line. I think that's clear. Maybe even better than unsigned infinity, because it distinguishes between real projective line infinity and complex infinity. I'd say go with that convention: sign on the extended line, +/- on the projective line, and just plain infinity on the extended complex plane. Kevin Baas^talk 15:51, 14 January 2010 (UTC)

Chaosbrot - OR ?

The author of chaosbrot, User:Doomed Rasher, is so far unable to provide sources for this generalisation of the Mandelbrot set, but claims the article is not OR because it is a "simple calculation" - see discussion at my talk page. Thoughts ? Gandalf61 (talk) 16:39, 9 January 2010 (UTC)

An article can't consist only of "simple calculation". Looks like OR to me, and probably needs to be deleted. Paul August ☎ 21:27, 9 January 2010 (UTC)

The chaosbrot author (who I agree seems also to be Doomed Rasher) writes the following: "I was messing around with Project Euler one day and dug up an old applet that I had written. It turned out that I could define a whole family of fractals from it, so I decided to patch the applet with a slider and text box so a user could define arbitrary values to see different fractals. My blog post has the download link and more details on how it actually works, or you can consult the relevant wikipedia article."

DR argues at User talk:Gandalf61 that the chaosbrot article deserves the same treatment as Gravity set. What this ignores is that the gravity set (a nice idea) was invented long ago by Chris Green at Atari, even before there were any browsers let alone Wikipedia, and that others such as Green's coworker Fred Mitchell at Atari and Jason Rampe in Australia have since taken an active interest in it, with Mitchell now promoting himself as its creator and with Rampe adding gravity-set images at http://softology.com.au/gallery/gallerygravity.htm to his wide range of Mandelbrot images, which he makes a business selling. If anybody but DR had found even that much redeeming value in chaosbrot a case might be made for it. As it stands, what is there about "chaosbrot" to distinguish it from promotion of a program by its author?

What DR could argue much more convincingly is that Fred Mitchell (mathematics) deserves the same treatment as chaosbrot, since Mitchell's main claim to fame seems to be as the programmer who translated Green's program from Forth to C, added a GUI, and (according to Gravity set) "discovered the incredible richness in the gravity set by manipulating the many parameters involved." The last paragraph of the Mitchell article reads "Currently, Fred Mitchell is working on a new mathematical system to describe networks of arbitrary complexity and structure. Nobody cares. He has also taken a keen interest in molecular biology in his hopes to understand the mechanism of life and death in deeper detail." User talk:Flajann comments in an edit summary, "Someone keeps vandalizing my page. I am putting it back to what it should be." The word "my" here suggests two problems: (a) WP:OWN and (b) Mitchell wrote his own biography under the name Flajann. --Vaughan Pratt (talk) 23:18, 9 January 2010 (UTC)

Please note that assertions about "conflict of interest" are problematic on two fronts: one, issues of "attempted outing"; two, the merits of material and our view of its inclusion should have nothing to do with who typed it in (among users in good standing). Charles Matthews (talk) 14:25, 10 January 2010 (UTC)

It's not outing if the editor links to his real name facebook profile on his user page. Pcap ping 16:47, 10 January 2010 (UTC)

Fred Mitchell (mathematics) fails our standards for inclusion. I've prodded the article, but I suspect an AfD will be necessary. Pcap ping 23:48, 9 January 2010 (UTC)

Notability is established by references, which the article has none of. A quick google finds only references that would do little to establish notability and suggest this is just OR and self promotion. As for simple calculations, that excuses you from doing or providing references for the intermediate steps of a calculation if an average reader for the topic could do the calculation themselves. It does nothing to establish the notability of the topic. If there wasn't already a discussion here I'd be tempted to start one at AfD.

And what other articles exist is irrelevant, as per WP:OTHERSTUFF. If other articles are on non-notable topics then it's something that should be raised on those articles talk pages, or via one of the various merge and deletion nomination processes. Unless it's a proposal to merge them all together into a Generalisations of the Mandelbrot set article they should all be addressed separately. --JohnBlackburne^words_deeds 23:33, 9 January 2010 (UTC)

So, is there a consensus that chaosbrot should be raised at AfD for wider community discussion ? Gandalf61 (talk) 09:38, 10 January 2010 (UTC)

Looks like Afd material to me.--RDBury (talk) 12:42, 10 January 2010 (UTC)

I AfD'ed the article since I do agree that the arbitrary name of "Chaosbrot" is by itself unnotable and not supported by references, but it remains questionable whether or not the same generalization has been published before under a different name. Suggest merging to Mandelbrot_set#Generalizations and removing references to "chaosbrot".

This pretty much establishes that Flajann = Mitchell Green and that he censors WP talk pages. Per WP:ENN, a commercial website/image gallery does not qualify as a reliable source for Gravity set, consider adding reliable references to the gravity set article other than the creator's website and a gallery of images. Doomed Rasher (talk) 16:20, 10 January 2010 (UTC)

'pretty much establshes'? I looked this over, and the article probably does, at best, deserve movement to the generalizations section with no name attached; but and also a 7-year-ago single small example to me says more about the person citing it than the person who did it.Julzes (talk) 13:32, 12 January 2010 (UTC)

I agree that that was a flub by a newbie - Flajann responded to a comment by overwriting it; we might all have done that in our first days of editing. This, however, is his most recent edit - and the most extensive and explicit legal threat I have seen in some time. Septentrionalis PMAnderson 01:46, 15 January 2010 (UTC)

To be fair, he was reacting to this (the "Friends report..." part). Flagged revisions might have prevented it, but having poorly sourced biographies of people that few watch-list seems just as bad. Pcap ping 02:07, 15 January 2010 (UTC)

I blocked Flajann (talk · contribs) per WP:NLT, but see also Wikipedia:Don't overlook legal threats. —David Eppstein (talk) 04:00, 15 January 2010 (UTC)

Shouldn't you be blocking the editor that added the presumably libelous material too, User:Genicity? It seems to be a WP:SPA who only edited that bio for apparently not so noble purposes. Doesn't seem to be here to build an encyclopedia: [15] [16] [17]a. Pcap ping 05:01, 15 January 2010 (UTC)

Ok, I left a standardized warning message on Genicity's user talk page. I don't think it rises to the level of an immediately-blockable offense the way the legal threat does. —David Eppstein (talk) 07:44, 15 January 2010 (UTC)

Gravity Set

Does anyone here think that fractal article should be kept? I can't find secondary references about it, and given the history of WP:COI edits of the article's creator, I'm tempted to AfD it too. The only 3rd party program known to implement it is Visions Of Chaos [18] (looks cool though). Pcap ping 08:18, 15 January 2010 (UTC)

Looking at Gravity set#Mathematics, I'm not sure that calling it a fractal is justified. The graphics seems derived from a sort of simulation of an n-body problem. Pcap ping 08:51, 15 January 2010 (UTC)

I also think that the article does not establish notability. Please AfD it to obtain consensus. Geometry guy 11:09, 15 January 2010 (UTC)

Wikipedia:Articles for deletion/Gravity set. Pcap ping 19:44, 15 January 2010 (UTC)

Progress on /Proofs articles.

See Wikipedia talk:WikiProject Mathematics/Proofs if you're interested. Crossposting here in case people aren't monitoring the other page.--RDBury (talk) 16:26, 16 January 2010 (UTC)

Polynomial function

Someone created a new article titled Polynomial Functions, plural and with a capital F. I moved it to polynomial function, which formerly redirected to a section of polynomial. The article should either get a lot of work or get redirected as before, in which latter case if something is worth keeping it should get accordinginly merged. Michael Hardy (talk) 03:31, 18 January 2010 (UTC)

I haven't looked at the article in detail but I'd be surprised and a bit disappointed if a standard pre-calc subject like that wasn't already covered completely.--RDBury (talk) 07:53, 18 January 2010 (UTC)

I've looked at the article more carefully and it appears that much of the material is elsewhere. The article is being expanded but it looks like the editor is ignoring the advisory about not duplicating material.--RDBury (talk) 15:16, 18 January 2010 (UTC)

I identified other articles with the same material, see the update history for details, and turned the article back into a redirect. There is an argument that there should be a summary article that links all this material but that would be polynomial.--RDBury (talk) 18:06, 18 January 2010 (UTC)

Certain widows

The only even prime number is 2.

This equation is to be solved for x.

Suppose there exists an odd perfect number n.

It would be unseemly to read the first of the sentences above in a book with the word "is" at the end of the line at the bottom of the page and then turn the page to find only the "2" and an otherwise blank page, and similar comments apply to the other two examples. In typography, a lonely last word at the top of the page is called a widow. A less extreme situation would be the "2", the "x", or the "n" in the examples above being the first character on the next line after the "is", the "for", or the "number" appears as the last item on the previous line. Hence I've set them with "nbsp" between the word and the final character. I've also been doing that in articles when things like this appear:

One seeks the solution x of the second equation above.

so that it won't appear as

One seeks the solution

x of the second equation above.

Is this worth prescribing in WP:MOSMATH? Michael Hardy (talk) 16:29, 16 January 2010 (UTC)

This seems pretty hostile to editors not used to writing html. I'd rather have a very rare ugly paragraph than confuse some would-be contributors and muddy the wiki syntax. Staecker (talk) 16:37, 16 January 2010 (UTC)

I think if the article is being cleaned up for FA status it should be something to add to the to do list, but there are other priorities in general. For example there is a backlog of over 1000 articles with an unreferenced tag. I personally have a hard enough time remembering to put nbsp inside ordered pairs.--RDBury (talk) 18:52, 16 January 2010 (UTC)

That sounds about right to me. It is always an improvement to prevent bad widows like this, but in most cases it does not have as big a cost-benefit ratio as other improvements (where the cost is the editor's time that could be spent constructively doing something else). It should certainly not be something that rises to the level of warning other editors that their edits are subpar because they neglected to do this. —David Eppstein (talk) 21:35, 16 January 2010 (UTC)

To me this is not a problem just for maths but one that could be seen in other contexts: dates and times, numbers, words, names like "Toys 'R' Us", etc.. As such it would be best addressed by WP as a whole, not just by mathematicians, if it needs addressing. I don't know if there's anything in HTML, or in a future version of HTML, that would handle this: a paragraph style or style hint. It's something I think word processors have been able to do for years so I'd be surprised if it's not been looked at. But it's also in part addressed by the nature of the web. If a user finds something difficult to read they can always increase or decrease the font size, adjust the window size, change their browser settings and often as here change per-site settings to make things more or less readable.--JohnBlackburne^words_deeds 21:57, 16 January 2010 (UTC)

That is something that imho needs to be solved on a software level, if it is perceived as serious problem at all. Generally it seems to me like a rather bad idea to let display issues dictate your writing style or structure of your sentences. WP:MOSMATH is already big enough as it is and as pointed out above there are plenty of more pressing issues than this.--Kmhkmh (talk) 20:13, 19 January 2010 (UTC)

Goedel picture

The photo of Kurt Goedel has been deleted on the grounds that it didn't have a fair-use rationale. Surely this is a clear-cut one; free-licensed pictures of the man are going to be hard to come by. --Trovatore (talk) 08:41, 19 January 2010 (UTC)

This is iffy because you can't argue it's needed for critical commentary—he isn't famous for his looks. On the other hand, I'd think some picture from his youth would be out of copyright by now. Pcap ping 08:46, 19 January 2010 (UTC)

I replaced it with the (rather poor quality one) from MFO found on commons. Pcap ping 08:51, 19 January 2010 (UTC)

Ah, that's a pity. I hate to see quality go down because of the unreasonably restrictive rules on fair use. We need to defend fair use by using it; we need to make sure the lawmakers understand that it has a constituency.

But with the rules as they are, unfortunately, the existence of that photo probably means there's no hope of bringing back the better one. --Trovatore (talk) 08:54, 19 January 2010 (UTC)

Edit war?

Something of a edit.....disagreement...seems to have broken out at Negative binomial distribution. I expressed some views a few weeks ago on the talk page. I haven't yet looked at this latest skirmish very closely, but maybe I will later today. Michael Hardy (talk) 19:09, 19 January 2010 (UTC)

Indentation format

How many other people see this? Quite often for me a displayed formula starting with :<math> is not indented in the standard fashion. I use Firefox. I haven't found any systematic reason why this happens. Charles Matthews (talk) 14:58, 20 January 2010 (UTC)

I don't recall ever seeing this (I use Safari). Can you give examples? Paul August ☎ 15:24, 20 January 2010 (UTC)

I also can't think of a time I've seen this, and I use Firefox. CRGreathouse (t | c) 15:26, 20 January 2010 (UTC)

Is it consistent, or does it come and go for the same formula without editing? An example that other people could look at would be a great help diagnosing it. — Carl (CBM · talk) 15:29, 20 January 2010 (UTC)

I'm seeing this now for Hypergeometric series#Notation, first displayed formula only (with the betas), and the rendering is not the standard font. Just that one formula in the whole article. Charles Matthews (talk) 15:44, 20 January 2010 (UTC)

It seems to use HTML rendering. Try adding \, to force PNG rendering? — Miym (talk) 15:50, 20 January 2010 (UTC)

Sigma-algebra too technical?

The article Sigma-algebra's talk page has had the "Too technical" template added to it. Since I saw no discussion on the talk page about the reasons for the template's addition, and since, in my opinion, the article is at about the right technical level for the likely or intended audience, I removed the template. But my edit was reverted, so I'm attempting to start a discussion here. In particular I'd like to know other editor's opinions on this article, the reasons why this article might be thought too technical, and for suggestions for how this article might be made usefully less technical. Paul August ☎ 15:31, 20 January 2010 (UTC)

The editor who originally tagged the article, User:Stpasha, has been quite active over the lasts couple days, so you could maybe post a query on their talk page ? Their edit comment when they tagged was "the lead section must at least make an attempt to explain the concept in layman terms". Maybe they are looking for a sentence in the lead such as "Informally, a sigma-algebra is like an xxx with a bit of yyy and some zzz" - although I must confess I wouldn't know how to give an informal description in this case. Gandalf61 (talk) 15:48, 20 January 2010 (UTC)

Thanks I've notified Stpasha on their talk page. Paul August ☎ 15:53, 20 January 2010 (UTC)

What I find much more puzzling on the Sigma-algebra talk page is the

{{Reqequation}}

template. Does anyone have an idea what's this request supposed to mean? — Emil J. 16:11, 20 January 2010 (UTC)

As I understand it, that template is asking for an image (in particular an image of an equation) be added to the article (see this edit). I can't think at the moment of what an appropriate image might be — I certainly don't think an image of an equation is needed. Paul August ☎ 16:27, 20 January 2010 (UTC)

A crazy template. I mean "Reqequation"; requesting an image is OK with me. What about Template for Delete? Boris Tsirelson (talk) 16:38, 20 January 2010 (UTC)

I agree. Sigma-algebra seems to be the only article where it is currently used, and it's very hard to think of a legitimate use. Hans Adler 17:47, 20 January 2010 (UTC)

It is used at RAMDAC also. Algebraist 17:50, 20 January 2010 (UTC)

Thanks, I missed that. And of course it's equally silly there. Hans Adler 18:57, 20 January 2010 (UTC)

After reading the article and the comments on the talk page, I realized that I had been misunderstanding what "too technical" meant, at least in this case. Since others commenting on this page seems to have the same misunderstand, I will try to explain. "Too technical" does not mean "too advanced" or "too difficult". Indeed, the definition given in the article was quite clear: a sigma algebra on a set X is a collection of subsets of X closed under complements and countable unions. But this definition is too technical: it tells you nothing about what a sigma algebra is for, why it is defined as it is, or why it is considered interesting. The technical details are correct, but the sense is missing.

It seems to me that this misunderstanding of "too technical" is a recurring minor problem in this WikiProject: someone will post that an article is too technical, and the project members will respond as if the complaint were instead that the article was too advanced. I believe I have done this in the past, and I hope I and others will be able to understand these complaints better in the future. —Dominus (talk) 18:01, 20 January 2010 (UTC)

If they have that misunderstanding, I think the text of the template is to blame. The {{context}} template is friendly and easy to understand. I know how to address it. But the {{technical}} template comes across as "I didn't understand this article" or "this needs to be dumbed down". And I have no idea how to address it based on the text of the template. — Carl (CBM · talk) 20:09, 20 January 2010 (UTC)

There's also {{cleanup-jargon}} — often appropriate on our math articles, but the part about either simplifying or explaining could again be misinterpreted as "this needs to be dumbed down". —David Eppstein (talk) 21:22, 20 January 2010 (UTC)

Being the previous person to ask for help in resolving a "too technical" tag, I'd say it's partly the tag and partly that many editors feel free to add vague tags without explaining what they are referring to. For the article in question here it seems pretty clear that there's no reason to expect someone who hasn't taken a couple of real analysis courses to be able to follow it. Wikipedia is supposed to serve a range of audiences, from grade schoolers to post-graduates. When people stumble across an article meant for a more advanced audience they are bound to get confused and sometimes they complain. On the other hand, sometimes the complaint is justified when an otherwise elementary topic is written as if the only people who might want to read it have Ph.D.s. In an ideal world every article would have a prerequisite tag somewhere so people who come across it will know what to expect and editors will be warned not to have an article meant for high school students be primarily about algebraic number theory. But I expect that idea is impractical.--RDBury (talk) 21:47, 20 January 2010 (UTC)

I guess, for "sigma-algebra" there exists a more specific cause: undergraduates study probability and see probability space (triple) with sigma-algebra; however, they see only countable probability spaces; thus they are puzzled, why the fuss. See probability space. Boris Tsirelson (talk) 06:34, 21 January 2010 (UTC)

Proof in Bernstein polynomial

There was an RfC for the proof in Bernstein polynomial. I looked at it and it seems use a probabilistic proof of uniform convergence of a series of polynomials to a given function f. It is, at best, hard to follow and there is some question as to whether it's valid. It could use more eyes on it.--RDBury (talk) 13:32, 21 January 2010 (UTC)

I did; see the talk there. Boris Tsirelson (talk) 20:50, 21 January 2010 (UTC)

Addition of natural numbers

I have proposed merging Addition of natural numbers into Natural number#algebraic properties. Please feel free to comment at Talk:Natural number#merger. — Carl (CBM · talk) 16:47, 21 January 2010 (UTC)

WP 1.0 bot announcement

This message is being sent to each WikiProject that participates in the WP 1.0 assessment system. On Saturday, January 23, 2010, the WP 1.0 bot will be upgraded. Your project does not need to take any action, but the appearance of your project's summary table will change. The upgrade will make many new, optional features available to all WikiProjects. Additional information is available at the WP 1.0 project homepage. — Carl (CBM · talk) 03:35, 22 January 2010 (UTC)

In case you wonder: this is separate from VeblenBot, although eventually the new WP 1.0 system will replace VeblenBot as well. — Carl (CBM · talk) 03:36, 22 January 2010 (UTC)

Prototile

Prototile is a fairly weak effort that could use some work. Michael Hardy (talk) 18:57, 21 January 2010 (UTC)

I'm more concerned that it's used in the first sentence of Penrose tiling with no definition. So an a reader comes to an article with a fair amount of coverage in the popular press and is immediately confronted with a word that not even professional mathematicians are familiar with, and is sent to a poorly written and unreferenced article to get the meaning. This is unacceptable writing style in a highly visible article.--RDBury (talk) 14:38, 22 January 2010 (UTC)

New page header

I took the liberty of adding a page header with links to related talk pages. If it's going to break something then delete it but otherwise I think it will be useful.--RDBury (talk) 05:29, 22 January 2010 (UTC)

Looks good to me. Paul August ☎ 19:12, 22 January 2010 (UTC)

Coxeter graphs are also a family of graphs.

Hello,

right now there is an article Coxeter graph. This is one very particular graph, named after Coxeter. However, as I know Coxeter graphs, there are plenty (in fact: infinitely many) of them, and they are constructed by use of maximal parabolic subgroups of a Coxeter group. (To add to the confusion, THE Coxeter graph is not one of them!).

This is examined in detail in Chapter 10 of Distance-regular Graphs by Brouwer, Cohen and Neumaier, but I don't find many other authors using this term.

Coxeter graphs are pretty important, because they include the Johnson graph (when using A_n), the binary Hamming graph (using B_n), a relation of the 24-cell (using F_4), the Schlafli graph (using E_6), the Clebsch graph (using D_5) and the Gosset graph (using E_7).

Does anyone have any suggestions for this problem?

Evilbu

Coxeter was a busy guy so I'm not surprised there are naming conflicts. These things are usually easy to deal with but there isn't a single recipe, see for example Cauchy theorem. In this case maybe you could add a Coxeter graph section (with references of course) to the Coxeter group article and add hat notes with links to straighten out people who land on the wrong article.--RDBury (talk) 13:57, 22 January 2010 (UTC)

Speed of light FA

FYI if you any interest in physics, Speed of light is being groomed to regain FA status. See Wikipedia:Featured article candidates/Speed of light/archive2 for the current discussion.--RDBury (talk) 19:04, 22 January 2010 (UTC)

Edittools

Could you perhaps have a look at the proposal MediaWiki talk:Edittools#Math and logic, and in particular the suggested list of symbols in MediaWiki talk:Edittools#Sandbox? Feel free to edit the list if you think something should be added, removed, or organised differently. Once the list stabilises, we can make a request to add those symbols to the toolbar. — Miym (talk) 18:03, 22 January 2010 (UTC)

They've been added - I only just noticed it - so have a look and play and see if you've any thoughts on improving it. I've already suggested a rearrangement, but so far that's been the only feedback (not counting a correction to an error of mine). --JohnBlackburne^words_deeds 22:28, 23 January 2010 (UTC)

set function

I was surprised to find that set function was a red link. Four articles linked to it. I've created a stub. Work on it. Michael Hardy (talk) 01:23, 23 January 2010 (UTC)

It was listed in Wikipedia:Missing science topics but lost amongst a thousand other red links. I think it a problem in general with the WP:MST list that valid subjects are overlooked because there are so many dubious titles to sort through. Imo we should have the Springer EoM & St Andrew's titles by now, at least as redirects. MathWorld titles tend to be somewhat idiosyncratic so I don't think these should be a high priority. I don't use the other sources for WP:MST much so I can't judge.--RDBury (talk) 16:07, 23 January 2010 (UTC)

Complex icosidodecahedrons

Created pages on Wikipedia:Missing science topics. Could someone with more knowledge about this stuff than me do the donkey work on the articles thanks?

These are the articles, btw:

• Small complex icosidodecahedron
• Great complex icosidodecahedron

Thanks again, 4 T C 03:04, 24 January 2010 (UTC)

sweeping revisions of Cantor's first uncountability proof

User:RJGray has extensively revised the article titled Cantor's first uncountability proof (remember here that no matter what "they" say, this is not Cantor's diagonal argument, which came two or three years later). He worked on this for some months in the Talk space and it's finally been moved to the Article space, and I've restored the revisions deleted by the move in order to merge the edit histories. (Apparently the merger takes a while to appear, so if you click on "history" within immediate future minutes, you might not see the merged history.)

I don't know who does the "ratings" that appear in the template atop the talk pages. Could those who attend to ratings look this one over to see if a revision of the rating is in order? Michael Hardy (talk) 03:39, 23 January 2010 (UTC)

No small group is in charge of ratings. Anyone can change article ratings at any time. The idea (hopefully) is that if enough people change them then the ratings will settle into a correct value over time. If you see a rating that seems wrong, just change it and don't bother to leave a comment. If someone else wants to change it back, they can. — Carl (CBM · talk) 03:55, 23 January 2010 (UTC)

I have to say I don't see that as really a recipe for convergence. But that's OK I guess; as far as I can see neither the priority nor the quality rating really matters much. I doubt very many editors are really looking for high-priority, low-quality articles per se to work on; they work on whatever seems interesting at the moment, and what they personally know about. --Trovatore (talk) 04:03, 23 January 2010 (UTC)

I suspect that most people will always work on what interests them the most, and that is good for a volunteer project. The rating scale is coarse enough that there aren't too many levels to switch between, which is the only reason to hope for convergence. And I agree that the ratings are not of crucial importance, which is why it is OK if they are not perfectly accurate. However, the ratings are used to select articles for release versions of Wikipedia, so there is some benefit in marking which articles are truly Top-priority versus which are truly Low-priority, etc. But nobody is forced to do that, and ignoring it is perfectly acceptable. — Carl (CBM · talk) 04:08, 23 January 2010 (UTC)

I understand very well CBM's point, but just as anyone can edit a math article whether they know any math or not, but nonetheless there are some who know how to do it and some who don't, so also with ratings. Michael Hardy (talk) 06:13, 23 January 2010 (UTC)

Maybe I'm in the minority but actually do look for high priority low quality articles to work on. But of those I tend to work on the ones I have an interest in and tend not to work on ones where I don't know much about the subject. I also check article ratings to make sure they're reasonable. To me it's a matter of maximizing benefit for the time invested, benefit being defined as the relative improvement to the article times the number of people who are likely to read it. But I also work on things just because I'm interested in them because I need a bit of fun as well.--RDBury (talk) 15:05, 23 January 2010 (UTC)

Based on Wikipedia:WikiProject Mathematics/Wikipedia 1.0/Assessment, I've changed the rating from "B" to "A". Michael Hardy (talk) 20:10, 23 January 2010 (UTC)

I have to say that, like mid-size car versus large family car, the ratings business is confused by cultural assumptions. It would not be a bad idea for this project to spell out its assumptions, based precisely on the idea that the transition B → A represents the upgrading of an article from being indicative of the content of a piece of mathematics, to being clearly useful to a student. Or whatever we mean. It is clear enough that things that trickle down from "featured article"/stretch limo are not particularly helpful, and so we should try to peg a more reasonable reference point from which to calibrate. Charles Matthews (talk) 12:02, 24 January 2010 (UTC)

There is some description at Wikipedia:WikiProject_Mathematics/Wikipedia_1.0/Grading_scheme; is that what you were looking for? — Carl (CBM · talk) 13:10, 24 January 2010 (UTC)

A template debate

Wikipedia:Templates for discussion/Log/2010 January 23 has a debate on a template allowing referencing of Éléments de géométrie algébrique. Not a mathematical discussion, but still. Charles Matthews (talk) 14:35, 24 January 2010 (UTC)

1-th 1-st

Hello. In the mathematics must I writing “1-th”? Or must I writing “1-st”? Or is both correct? --Diwas (talk) 12:51, 21 January 2010 (UTC)

You mean as an abbreviation for "first"? If so then "1st" would be correct (followed by 2nd, 3rd, 4th, 5th ... nth etc). Paul August ☎ 13:17, 21 January 2010 (UTC)

Sorry my question was not very precisely. I think about this in Ordinal number So we can freely speak of the γ-th element in the class (with the convention that the “0-th” is the smallest, the “1-th” is the next smallest, and so on). I think n+1th is correct but 1th is not. (in the first minute I saw many 1th but mostly that is (n+1)th, (n-1)th, ... ) --Diwas (talk) 14:54, 21 January 2010 (UTC)

Ah. Fixed. — Emil J. 15:26, 21 January 2010 (UTC)

Actually, in the given context you can find examples of "1-th", see for example [19]. I have no idea what is standard here. Paul August ☎ 15:35, 21 January 2010 (UTC)

Regardless of whether its 1th or 1st, this is a bad convention. 0th and 1th are unused in ordinary English but 4th and 247th already have a definite meaning, and this convention is changing that (unless one is attaching great meaning to the dash). Why not say "element number γ", that's hardly longer, and unambiguously means the element that carries number γ, which avoids confusion. Even in ordinary language "object number n" does not have to be synonymous to the n-th object. Personally I am all in favour of starting to count at 0 (I wish we learned out children to count that way...) but the i-th usage is too grained-in to want to give a new meaning, better just avoid it if you don't like it. The best way to make a notion/notation unusable is to give two conflicting meanings to it. Marc van Leeuwen (talk) 08:43, 26 January 2010 (UTC)

Hmm? Not at all. We have two (or more) conflicting meanings for lots of stuff in mathematics. We get along alright just the same. Causes occasional minor annoyances; it's no warrant for any heavy-handed attempts to enforce uniformity.

In set theory in particular, sequences are ordinarily zero-based; if you want to say "element number γ" you need to make it clear somehow what you mean. Sometimes that'll take care of itself automatically, on the grounds that one meaning will be nonsensical (say, if you talk about a "set of rank ω").

On the linguistic question, I always say "en plus first", never ever ever "*en plus oneth". --Trovatore (talk) 08:49, 26 January 2010 (UTC)

Formula and Equation

I think at the very least the two articles Formula and Equation need to refer to each other explicitly, but I'm not aware of any formal distinction of the two terms so I don't know how to go about that. In the extreme case perhaps these two articles could be merged. Does anyone have any suggestions? Cheers, Ben (talk) 12:07, 22 January 2010 (UTC)

The thing they have in common is they involve mathematical expressions. They look similar as they're both expressions with equals signs, but they are quite different. Formulae are expressions that express a relationship, equations are expressions that need to be solved. They both link to expression and are linked from there, so in that way the relationship is already established. --JohnBlackburne^words_deeds 12:30, 22 January 2010 (UTC)

Hmm, so would you say that x² + y² = 1 is a formula or an equation? Ben (talk) 12:53, 22 January 2010 (UTC)

It's the equation of a circle, in particular the unit circle centre the origin. Alternately it's an equation with the solution being the set of all points a distance 1 from the origin in the x-y plane. --JohnBlackburne^words_deeds 13:08, 22 January 2010 (UTC)

A better example might be r² = x² + y². It's an equation for a circle and a formula for computing distance. The distinction is one of viewpoint, a mathematician wants to understand the implications of an equation, a carpenter wants to know how long to cut a board to make a diagonal brace. I'd also throw the article Expression (mathematics) into the mix (it even has a note saying it contradicts Formula.) They all need to at least refer to each other. The present formula article is dreadful. A merge with equation could work as long as the practical viewpoint is preserved. (I just found Formulation and Formulations, lots of mess to clean up.)--agr (talk) 13:39, 22 January 2010 (UTC)

I would say that the "formula" there is x² + y², and the "equation" r² = x² + y² says that the square of the distance is given by that formula. That is, a formula is something that returns a value when you substitute inputs, and an equation joins two formulas with an = sign. However, it is very common ignore this distance and call the equation itself a formula, particularly if one of the sides is a single variable, and particularly in elementary classes. But I would say that the formula for the area of a circle is "πr²", not "A = πr²". — Carl (CBM · talk) 13:45, 22 January 2010 (UTC)

I think that equation means "a statement that some things are equal" (they "equate"), and that formula means "a generic statement involving mathematical symbols". Using that definition 2 > x is a formula, but not an equation. See [20] and [21] for example. Gabbe (talk) 13:43, 22 January 2010 (UTC)

This is certainly the meaning when people "display a formula" in TeX. — Carl (CBM · talk) 13:45, 22 January 2010 (UTC)

I tend to think of a formula as "a formula for ...", so the above example would become a fomula for r if written

$r={\sqrt {x^{2}+y^{2}}}$

I would call the example 2 > x an inequality: another use for an expression or expressions, not an equation or formula. --JohnBlackburne^words_deeds 13:56, 22 January 2010 (UTC)

(edit conflict) Exactly. By the terminology I used, all equations are formulas, all inequalities are formulas, but inequalities are not equations. Does anybody have any references other than the one I've provided (MathWorld)? Gabbe (talk) 14:05, 22 January 2010 (UTC)

To throw a bit more mud in the water, neither of these terms is strictly mathematical. For example every compound has a chemical formula and there are plenty of equations floating around in physics. I'll have to look at the articles but it seems like there are few WP:NOTs that they might be violating.--RDBury (talk) 14:04, 22 January 2010 (UTC)

From the math point of view, I'd support the idea that a formula expresses how to calculate a quantity, while an equation indicates two quantities are equal. So in the context of finding area, A = π r² is a formula, while in a calculation one might inquire when a square and a circle have the same area using the equation a² = π r². Solving this equation leads to a formula for the side of square with area equal to the circle: a = √π r. On this basis, equations are a subset of conditions, and formulas are expressions for evaluation of quantities. Brews ohare (talk) 17:15, 22 January 2010 (UTC)

I notice that Excel uses the vocabulary "function" for what I'd call the right-hand side of a formula. Here are some example usages of these terms. Brews ohare (talk) 17:39, 22 January 2010 (UTC)

We solve equations, but never solve formulas. Here's] an example. That agrees with JohnBlackburne's observation. Brews ohare (talk) 18:20, 22 January 2010 (UTC)

Perhaps one definition might be that a formula is an equation with a simple variable on one side and an expression on the other.--agr (talk) 12:43, 24 January 2010 (UTC)

The formula $0\leq p\leq {\frac {1}{4}}$ is not an equation at all, but it is a formula. JRSpriggs (talk) 22:16, 24 January 2010 (UTC)

For me a formula does not even have to include an equals sign; that is just a manner of presentation. I could accept saying that

$-b\pm {\sqrt {b^{2}-4c}} \over 2$

is a formula for the roots of the polynomial X²+bX+c. According to the Abel-Ruffini theorem there is no formula involving only arithmetic operations and radicals for the roots of a general polynomial of degree 5, but there trivially is an equation for them (which of course also involves the equality relation). Note that nothing in the discussion of that theorem ever mentions an equals sign. As for inequalities, I would not classify them as formulas at all. Certainly one can give inequalities for the roots of a general quintic (assuming they are real). Marc van Leeuwen (talk) 09:28, 26 January 2010 (UTC)

Some eyes on WP:V would be good

Someone tried to make autobiographical refs unacceptable, which would be a problem for us since a number of academic BLPs rely on institutional home pages for a number of details. Pcap ping 13:56, 23 January 2010 (UTC)

Not a lot of publications are doing features on math professors so in the majority of cases I would think that personal/academic websites are going to the only place you can find personal information. If you've ever tried to do a bio of someone who died b.i. (before internet) you know how big a problem it can be to have to rely just on published resources. You can have lots of material on the person's work but nothing on the date and place of birth.--RDBury (talk) 15:29, 23 January 2010 (UTC)

MacTutor is very good on the people it covers; unfortunately that's mostly a subset of our bios. Ray^Talk 16:58, 23 January 2010 (UTC)

How would one write about the life of St. Patrick without using autobiographical sources? Michael Hardy (talk) 19:36, 23 January 2010 (UTC)

By using the innumerable secondary sources who have themselves used the autobiographical sources - and in so doing have clarified the problems with them. I doubt any of our mathematicians have their dating disputable by a century, but the principle is the same. Septentrionalis PMAnderson 01:34, 26 January 2010 (UTC)

There is a more general problem, for which I don't have any suggestions, but it's worth thinking about. On the one hand, WP:PROF applies to a large number of living mathematicians. On the other hand, many of these mathematicians have little-to-no biographical coverage in published sources. — Carl (CBM · talk) 19:50, 23 January 2010 (UTC)

SlimVirgin was just restoring long-standing wording, which does not prohibit autobiographical sources: the key words in "Self-published sources should never be used as third-party sources about living persons" are "third-party". It just says we can't use things like a personal home page or website (institutional pages are arguably published by the institution) as a source about somebody else, who is not the author.John Z (talk) 20:53, 23 January 2010 (UTC)

"institutional pages are arguably published by the institution": yes and no. Anyone working or that has worked in academia can tell you they can write almost whatever they want on their institution hosted home page. Surely, making some outright false claims would cause an investigation by some ethics committee, but there's no editorial oversight in the traditional sense of the word. On the other hand, staff catalog entries are somewhat independently written, but there's usually little contents on those. Pcap ping 11:48, 24 January 2010 (UTC)

Obviously autobiographical can be used as source (as can interviews and other personal statements). They need however to be treated like primary sources, i.e. be used in a limited and careful fashion and not as neutral third party source. If some editors seriously consider to ban the use of any biographical material in principle, then this needs to blocked/corrected.--Kmhkmh (talk) 12:09, 24 January 2010 (UTC)

Interval finite element

Hi Maths experts. The main author of Interval finite element has created a huge technical article. You've got to admire the extent to which content has been added; perhaps it is being copied from somewhere else? Could someone please review it for {{jargon}}, {{howto}} and {{original research}}? Tayste (talk / contrib) 19:07, 24 January 2010 (UTC)

Looking at the history it's pretty obvious that it's the product of many hours of work over a period of several months. My concern would be more about WP:NOTTEXTBOOK.--RDBury (talk) 19:52, 24 January 2010 (UTC)

Theoretical engineering, if that's not an oxymoron. First of all it needs general cleanup, starting with the title. I would probably thin it out, trying to preserve the survey nature; but also bring the major points closer to the top (concentric style_. Charles Matthews (talk) 16:27, 26 January 2010 (UTC)

List of limits

The page list of limits is a basic calculus page and apparently has not been developed very much (many famous limits are missing)... this seems strange, maybe there are pages which cover the same topic?--Pokipsy76 (talk) 18:00, 26 January 2010 (UTC)

These list of formulas type articles are in kind of a gray area in terms of being encyclopedic. Lists of integrals and related pages are of the same type. You can make a case that the articles should exist, but long indiscriminate lists are against guidelines and these types of articles seem to be natural collection points for just that kind of thing. So I'd say add references and try to establish reasonable criteria for inclusion, but don't start adding tons of new formulas. Abramowitz and Stegun is already online so there isn't a huge need for pages like this.--RDBury (talk) 20:44, 26 January 2010 (UTC)

Boy's surface/Proofs AfD

FYI, the article Boy's surface/Proofs is being considered for deletion here. For some reason it's not showing up on the current activity page.--RDBury (talk) 04:57, 25 January 2010 (UTC)

The correct wikilink for the AfD is Wikipedia:Articles for deletion/Boy's surface/Proofs (2nd nomination) (the one you mention above is for the 2007 AfD). Nsk92 (talk) 05:00, 25 January 2010 (UTC)

The article is now in current activity; it probably wasn't there before because I missed a step in the deletion process. Thanks for your patience.--RDBury (talk) 18:31, 27 January 2010 (UTC)

A big mess of many non-notable statistics articles

Please see Wikipedia:Articles for deletion/Ligong Chen. (Crossposted to Wikipedia talk:WikiProject Statistics.) —David Eppstein (talk) 01:32, 29 January 2010 (UTC)

Trapazoid/Vandalism

The formulas in the Trapazoid article have recently been tweaked by anon users with no comments. This seems to be a particularly pernicious form of vandalism and general problem, people can go in and make a small change to a complex formula, just enough to make it wrong, and it takes significant time and effort to recheck the formula to make sure the original version was correct. I know Wikipedia:Assume good faith and all that but it seems like it would be more practical to just revert difficult to verify changes to formulas when they are made by anon users, especially if there are no comments.--RDBury (talk) 20:27, 27 January 2010 (UTC)

Reverting such unexplained edits as "dubious" is reasonable. Also having a citation that allows easy verification is extremely beneficial in such cases. I always try to add one when I come across edits like this. Paul August ☎ 21:42, 27 January 2010 (UTC)

I just checked the coefficient 4 inserted by the anon to the formula

$A={\frac {(a+b)}{4(b-a)}}{\sqrt {(s-b)(s-a)(s-b-c)(s-b-d)}}$

for a simple-to-check case of b=a+epsilon and d=c. The result: after the suspicious change the formula is definitely wrong; before the change it was hopefully right. Boris Tsirelson (talk) 21:23, 27 January 2010 (UTC)

Oops... According to that talk page, notations in the article are inconsistent: a,b,c,d sometimes becomes a,c,b,d or so. I used notations shown on the picture, but I got unsure. Maybe the whole fuss with the formulas happens for this reason? Boris Tsirelson (talk) 21:34, 27 January 2010 (UTC)

The formulas are coming from MathWorld, so they are referenced (though MathWorld has been known to have occasional typos). But even when you don't have to derive them from scratch there are still notation issues, different variables used, etc. If it was just this article I wouldn't have brought it up here but I see this kind of thing pop up a lot in my watch list and it struck me that it takes much longer deal with this kind of thing, even if it's not intentional vandalism, than the usual random gibberish that people insert into articles.--RDBury (talk) 05:05, 28 January 2010 (UTC)

MathWorld is not really a good source for us, for both procedural and substantive reasons. The procedural reason is that MathWorld, like Wikipedia, is a tertiary source, and what we want are secondary sources with possibly a light mix of primary sources when absolutely necessary. The substantive reason is that MathWorld is idiosyncratic, especially with regard to terminology, and has been known to propagate outright neologisms. --Trovatore (talk) 22:42, 29 January 2010 (UTC)

I somewhat disagree with that, though secondary sources are preferable in general, many reputable/reliable tertiary sources are perfectly alright to use, even more so for standard formulas. Note that many math textbooks and special subject encyclopedias (such as Springer's Encyclopaedia of Mathematics) are tertiary sources. Dismissing such tertiary sources is not a good idea from practical perspective alone and you might even argue that they provide an additional layer of error checking and notability for the concerned subject. As for this concrete case I'd agree that (unneeded) formula modifications without proper sourcing or explanation should simply be reverted without furher ado. The same holds for formula edits that might be sourced but are inconsistent with regard to notation and names used throughout the article.--Kmhkmh (talk) 06:12, 30 January 2010 (UTC)

Thanks for the help, that should save time in the future. Regarding MathWorld, if I saw an article that used only that as a reference I wouldn't hesitate to add a refimprove tag, but I also wouldn't hesitate to add MathWorld as a reference if it had some relevant information, especially if the article didn't list any other references.--RDBury (talk) 15:47, 30 January 2010 (UTC)

I suppose that's reasonable. I have a slight bug up my butt about MathWorld, because of a couple of WP articles that were based on MathWorld neologisms. --Trovatore (talk) 05:16, 31 January 2010 (UTC)

Edwards curve

Edwards curve and twisted Edwards curve need work. Michael Hardy (talk) 04:49, 29 January 2010 (UTC)

These are two of a set of articles created by a single editor about current research in cryptography and whose notability is marginal at best.--RDBury (talk) 16:27, 29 January 2010 (UTC)

They seem to be referenced. Maybe the level of nitty-gritty detail is excessive. Charles Matthews (talk) 22:56, 29 January 2010 (UTC)

I looked on Google last night and found plenty of references. I added a textbook to the article, and I added Edwards's original paper. This seems to be notable in the cryptography community, though I agree that the article as it stands isn't very good. Ozob (talk) 19:12, 30 January 2010 (UTC)

Displaytitle help

Resolved

I cannot seem to make Π⁰₁ class format its title correctly. Does anyone else have experience with getting subscripts and superscripts into page titles? — Carl (CBM · talk) 14:39, 30 January 2010 (UTC)

DISPLAYTITLE is limited in scope. It can't change the "Pi" in the title into a "Π". It would (I think) work if the page was moved to Π01 class, but that triggers the title blacklist for being a mix of latin and greek characters, so I can't make the move. Algebraist 14:46, 30 January 2010 (UTC)

Brilliant, thanks. — Carl (CBM · talk) 14:50, 30 January 2010 (UTC)

Harold Edwards (mathematician)

Harold Edwards (mathematician) is a new article. Get busy. Michael Hardy (talk) 04:26, 31 January 2010 (UTC)

PS: One of the things to work on is this: find all articles that ought to link to this new article and put the links there. Michael Hardy (talk) 04:27, 31 January 2010 (UTC)

Ok, I think I've found most of them. The article itself could still stand improvement (as could most of our mathematical biographies). —David Eppstein (talk) 05:18, 31 January 2010 (UTC)