Template talk:Cite journal/Archive 2009 July

Referenced page number(s) -- where?

When using the Cite Journal template within ref-tags, how do I reference the page number(s) for a given citation, as opposed to the article's page range? The template's page(s) field seems intended for the latter, but where does that specific page number(s) go that I want to reference? ARK (talk) 19:51, 20 June 2009 (UTC)

Use {{rp}} after the closing </ref> tag. Martin (Smith609 – Talk) 20:23, 20 June 2009 (UTC)

Great, thanks! Maybe it helped if the {{rp}} could be a bit more explicit about this issue? I'd been to that page before asking my question but didn't find the answer there. ARK (talk) 21:07, 20 June 2009 (UTC)

Translated title

I have added trans_title for including a translated title when the original title is non-English. This also works for {{cite web}} and {{cite news}}. See docs for details. Let me know if you encounter any problems. Thanks, Crum375 (talk) 04:49, 25 June 2009 (UTC)

Full text requires subscription but PubMed ID available

In working to bring the Oxygen toxicity article up to the Wikipedia:Featured article criteria, I have some confusion over links to full text articles that require subscription and PubMed links.

• User:Diberri's PubMed tool will allow us to include any links to full text articles in PubMed regardless of subscription requirement. For example: Moon RE (2005). "Bubbles in the brain: what to do for arterial gas embolism?". Critical Care Medicine. 33 (4): 909–10. PMID 15818131. Retrieved 2009-07-01. {{cite journal}}: Unknown parameter |month= ignored (help)

The example from today is a full text journal article that is not linked from PubMed:

• Yildiz S, Aktas S, Cimsit M, Ay H, Toğrol E (2004). "Seizure incidence in 80,000 patient treatments with hyperbaric oxygen". Aviation, Space, and Environmental Medicine. 75 (11): 992–4. PMID 15559001. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)

Then I added the link to the full text:

• Yildiz S, Aktas S, Cimsit M, Ay H, Toğrol E (2004). "Seizure incidence in 80,000 patient treatments with hyperbaric oxygen". Aviation, Space, and Environmental Medicine. 75 (11): 992–4. PMID 15559001. Retrieved 2009-07-01. {{cite journal}}: Unknown parameter |month= ignored (help)CS1 maint: multiple names: authors list (link)

This was immediately reverted to the older version and explained as "Remove URL to non-free source. It isn't to the full text, just to an abstract, and so is no better than the PMID". The full text is at the bottom of the linked page. At work I have free access to this article and after checking it from home, I still have free access here so the user that reverted my change may have an issue. Either way, what are the official suggestions on how this should be handled for articles that require subscription going forward? My personal preference is to give the readers the link and option to pay or not. Should we continue to include these links to full text along with the PubMed ID's and DOI numbers? Thanks --Gene Hobbs (talk) 21:15, 1 July 2009 (UTC)

My preference is, like yours, to include PMID and DOI whenever it's available but also to include a free full-text url when it's possible to find one and when the PMID and DOI links are non-free. But ingentaconnect links are not usually free; are you sure that one is? (I'm using my campus IP now so I can't tell myself.) —David Eppstein (talk) 21:19, 1 July 2009 (UTC)

That's why I wanted to check from home before I asked. It still shows up for me and I made sure I am not on the VPN now. I am asking a collaborator in the UK to see if maybe that is the issue. Thanks --Gene Hobbs (talk) 21:26, 1 July 2009 (UTC)

I am that collaborator in the UK and I can see the full text without problems - I just downloaded it, even though I have no subscription. Nevertheless the documentation for this template says:

*url: This should point to, in descending order of preference:

1. A free online version of the full text
2. An online version of the full text, for which subscription is required
3. An abstract or information page, if no DOI or PMID record is available

If a DOI or PMID is available, the URL should only be specified if it would point to a different page to that which a DOI or PMID would redirect to.*url: This should point to, in descending order of preference:

1. A free online version of the full text
2. An online version of the full text, for which subscription is required
3. An abstract or information page, if no DOI or PMID record is available

If a DOI or PMID is available, the URL should only be specified if it would point to a different page to that which a DOI or PMID would redirect to.

Doesn't the second (2) suggest that the preference is to link to a subscription version of the full-text (if a free full-text is unavailable), or has consensus changed? --RexxS (talk) 22:54, 1 July 2009 (UTC)

But in this case it seems that free full-text is available, so item (1) prevails. —David Eppstein (talk) 23:14, 1 July 2009 (UTC)

Ah but the discussion is about the fact that some users find that the full-text is subscription-content. If I'm reading it correctly, it imples that we should make a link to full-text content, regardless of subscription status (as long as it's not the page that PMID points to anyway). --RexxS (talk) 23:25, 1 July 2009 (UTC)

Please note you are only partly quoting the current documentation, for after the 3-step 'url' pecking order comes an important qualifier about the 'format' parameter:

*format: Format, e.g. PDF. Don't specify for HTML (implied as default).

** Specify as 'fee required' if free access no longer available.

** Specify as 'Reprint' if a full original version but not hosted by the original publisher.

The default assumption is therefore that if the url is specified without qualification, then it points only to free-for-everyone full articles (as well as being in html)... else the format parameter should be specified to alert a reader that a subscription may be required. This therefore allows both assumptions of previous discussion (that assumption is for free full access, but that as second best can be used for subscription links too). Hence Gene the revert of your linking was technically correct, but a little unfair: for a start the link you gave is to just an abstract and at its bottom of that webpage is shown "The full text is free" and it links so. Hence the url specificed should have been to the full version, but even if that had been available only by subscription, then the markup should have included "|format=fee required" eitherway the url parameter can & should be defined in this case :-) David Ruben ^Talk 02:10, 2 July 2009 (UTC)

Thanks David. Follow up question, I have used several links in the past that go to similar pages with download available because I know these urls are protected and the direct pdf file names tend to change. Is there a mechanism for this case or should we change and monitor all of them (over 1,000 links that I am aware of)? Thanks again. --Gene Hobbs (talk) 02:27, 2 July 2009 (UTC)

Unfortunately, David, you can't directly link to the document. Try clicking your "direct" link and you'll see that it has expired. This website ensures that permanent links to downloads cannot be made. Many websites do not allow you to link directly to downloadable content, and that is their prerogative. I think that we must consider the reader: is the inconvenience of us linking to a page, where they have to spot a download link and then click it, so great that we are going to refuse to offer them the chance to get the full-text free. Personally, I think not. YMMV. --RexxS (talk) 03:03, 2 July 2009 (UTC)

Ah-ha (shows danger in just copying links without retesting) - yes indeed direct link does not work. Ok agree, in this case the url link should be to the perminant abstract page, but I would then specify 'format' parameter as per documentation "|format=fee required". But it might be better if we be clearer with something along lines of "|format=abstract, fee required full article" although that seems a little long winded - better phrasing anyone ? David Ruben ^Talk 03:42, 2 July 2009 (UTC)

I'd be just as happy to not set up bureaucratic and bitey rules about how the format strings must be formatted in cite journal templates with urls and trust in our readers to be able to figure out that the url ends in .pdf and/or that following the link leads to a site that asks them to pay. —David Eppstein (talk) 03:50, 2 July 2009 (UTC)

David Eppstein is right. Its just WP:CREEP. Writers are obliged to cite sources as accurately as possible, they are not obliged to make and keep them accessable. Because "make" and "keep" are subjective, its a spiral with no end. -- Fullstop (talk) 13:10, 2 July 2009 (UTC)

"pp" in {cite journal} and {cite book}

It seems that {{cite book}} adds "pp." before page numbers, and {{cite journal}} does not. This can make references sections look funny (see, for example, Chinese classifier#Bibliography). Might it be better to standardize these (either removing pp from {{cite book}}, or adding it to {{cite journal}})? I mean, I could quick-fix my own article by manually typing "pp" within the |pages= parameter, but that would be stupid, and then if the template were changed I might end up with "pp. pp. 200-300" all over the place. Personally, I think removing the "pp" from {{cite book}} would be better, if we could make sure it would still be evident that these numbers are page numbers. rʨanaɢ ^talk/_contribs 04:30, 3 July 2009 (UTC)

It's only fairly recently that the p. & pp. feature was added to {{Cite book}} and we had to flog around to remove instances of "pp. pp.". However, I'm happy with the template as it now is (ie adding p. and pp. automatically). The nopp= parameter provides the need flexibilty to suppress the p. and pp. if you need to put in something unusual (like an Appendix number or "Front Cover"). I agree that it's confusing to have the two templates different so vote to change {{Cite journal}} instead. Stephen Kirrage ^{talk - contribs} 09:18, 3 July 2009 (UTC)

AWB has recently been updated to remove explicit use of 'pp.' from citation templates due to the automatic adding. If {{Cite journal}} is missing this, then I also think the template should be updated. Rjwilmsi 09:36, 3 July 2009 (UTC)

Ah, ok. I didn't know about |nopp=, but that one will be enough to solve my problems; as for the larger issue of how the templates should be formatted, if there has already been consensus to add p. and pp. to {{cite book}} then it makes more sense to do the same to {{cite journal}}. (Personally I prefer no page numbers, but the important thing is following the consensus and keeping things standard across the project.) rʨanaɢ ^talk/_contribs 14:32, 3 July 2009 (UTC)

As a user, I prefer the pp style. As of now, the cite journal code for some unclear reason uses that style only if no journal or other publication is specified. Otherwise, it reverts to just displaying the 'pages' or 'page' content verbatim. I can easily change it so it will always use the pp style, if there is consensus for it. Crum375 (talk) 14:41, 3 July 2009 (UTC)

My vote is for making pp the default and adding |nopp= like in {{cite book}}. rʨanaɢ ^talk/_contribs 16:00, 3 July 2009 (UTC)

I support that as a user. Anyone else? Crum375 (talk) 16:06, 3 July 2009 (UTC)

• I have added it for now, since it seems we are all in agreement here. If people object, it can be reverted. I notice that in {{cite book}}, if a journal or other periodical publication is supplied, it also reverts to no pp, like the previous {{cite journal}} behavior. This can be eliminated if desired (though it should probably be raised in the cite book talk page). Crum375 (talk) 16:51, 3 July 2009 (UTC)

This is why we should be using a standard citation style, like say Harvard, rather than trying to roll our own sort-of-similar style. If we did, this sort of question would be moot. "pp" is not needed in journal articles because almost all journal article citations have page numbers so we're all expecting to see them there. It is needed in book citations because many book citations don't have page numbers so we need to be told what the numbers mean when they're there. I object to such precipitate changing of basic citation formats. —David Eppstein (talk) 16:57, 3 July 2009 (UTC)

• I reverted per your objection. My reversion can always be unreverted if people so decide. Crum375 (talk) 17:08, 3 July 2009 (UTC)
• I'm not sure why we would think that a reader's intelligence would change between reading one citation and the next. Either form works and stylistic consistency guidelines are essentially unenforceable on WP. The more serious issue is how to segregate citing pages from citing the source that contains those pages, so that we don't have multiple citations of the source work/book/paper. For harvard, we can use {{rp}} but we have no such alternative for other forms of citation.LeadSongDog come howl 17:17, 3 July 2009 (UTC)

Use harvard when multiple cites reference different pages of the same source, and normal inline cites otherwise. Is there a problem with doing that? --RexxS (talk) 18:06, 3 July 2009 (UTC)

By "Harvard" in this context I meant the standards for formatting full references, not the part about how to format the shorter inline citations. By the way, another reason not to do this: currently, {{cite journal}} and {{citation}} are close to identical (there are only some minor variations in punctuation); changing one of them individually rather than changing the core formatting breaks this. —David Eppstein (talk) 19:36, 3 July 2009 (UTC)

I think some of the comments above are mixing up Harv and cite book. A lot of articles, especially bigger ones, use {{cite book}} with |pp= for the general reference section at the bottom, and {{Harv}} with |p= for specific footnotes (again, see Chinese classifier#References and Chinese classifier#Bibliography for an example). The footnotes refer to specific pages; the pp in the bibliography entries refer to the entire length of the work (if I'm citing an entire book I don't use it, but if I'm citing a book chapter in an anthology, or one article in a volume of a journal, I give the pages where it appears...it may not be necessary, but it's the standard I'm accustomed to). My original message was about standardization within the bibliography, not within the reflist; Harv is irrelevant to the problem I was raising (and Harv isn't a problem because you can choose whether or not to display "p" by using either |loc= or |p=). rʨanaɢ ^talk/_contribs 20:49, 3 July 2009 (UTC)

One size doesn't fit all. The multilayer approach in which footnotes point to bibliography entries is suitable for some article types (e.g. World War I) in which we have significant reuse of the sources, though this problem still hasn't been diagnosed (just whining), let alone fixed. For others (e.g. Glucose) it's not such a great fit.There's broad acceptance of Vancouver single-layer inline citation for biomedical subjects. Other "hard" sciences have generally adopted it too, though more recently.LeadSongDog come howl 21:55, 3 July 2009 (UTC)

Reference generator tool for Cite book and Cite journal

The Reference generator should continue to work correctly if the Cite book or Cite journal templates are modified. --DThomsen8 (talk) 00:21, 4 July 2009 (UTC)

Adding a pub-type parameter

The metadata available on Pubmed, Medline (etc.) includes a parameter, denoted "PT" in Medline syntax, "pub-type" in ASN.1 syntax or "PublicationType" in XML. This parameter can take values such as "Review", "Journal Article", or "Short Communication". Adding this to our citations could be of great value in sorting the wheat from the chaff. In guidelines such as WP:MEDRS we explicitly prefer meta-reviews and reviews over journal articles, and those over letters, short communications and such. A facility for indicating which pub-type a source is could expedite reviews considerably, add value for end readers, and reduce the "which source should we use" debates. LeadSongDog come howl 21:23, 16 July 2009 (UTC)