Talk:vi (text editor)/Archive 1

Archive 1

Unicode

I am wondering if Unicode support should not be added into the main article. I believe people would be interested if vi can support Unicode or not. I am not a vi user, so I have no way to check this myself. --Saoshyant 15:11, 1 October 2005 (UTC)

You'd have to say which program you're interested in. Vim can support Unicode. 207.245.124.66 15:16, 14 December 2005 (UTC)

vi's date of birth

I'm going to try to put a date on vi's creation. The commonly-quoted date is 1976.

In this interview, Bill Joy says "it was probably '76". He apparently started it in 1975, but lost the ex source code in 1976. He started the screen editor late in '76, while listening to election results on the radio.

O'Reilly contradicts slightly, saying the ex Pascal source code was published with BSD early in 1977. That is consistent with ex/vi being written in late 1976. But I have not found any information about when the name vi came into use. --Nate Silva 01:01, 9 Jan 2004 (UTC)

I was there (UCB computer center staff & student) and I'm fairly sure it was '76. Visual mode in the ex editor was preceded by something called "open mode", which was the equivalent of visual mode, but only operated on one line at a time. This was due to ADM-3 terminals not having cursor addressing; only when ADM-3a terminals (fully cursor-addressable) became common in the Evans Hall terminal rooms was it possible to use visual mode. The debut of vi as /usr/pascal/vi probably happened in late '76, at least sometime after April. It was definitely before any release of BSD. I probably have this on an old 9-track tape but I don't know if it can stil be read (I doubt it). I also have printouts and notebooks (somewhere!) I know none of this is verifiable or a reliable source for Wikipedia purposes, but I do know quite a bit about the origins of vi and especially of csh (to which I made some trivial contributions). MCB 05:29, 1 May 2006 (UTC)

Capitalisation

Am I the only one who feels uncomfortable with capitalizing it as "Vi" at the beginning of a sentence? Or is there a Wikipedia policy about mangling case-sensitive stuff? -- Kimiko 13:12, 3 Mar 2004 (UTC)

vi is one of those odd things caused by Unix: a proper name that must start with a lower-case letter. A good typographical way to get around it would be to consistently render it as <tt>vi</tt> (rendered as vi) throughout the article. - David Gerard 13:29, Mar 3, 2004 (UTC)

The classic thing to do here is to re-write the sentence so that it doesn't begin with 'vi' or whatever lowercase doo-dad (program, function call, etc.) you're talking about. --Ansible 00:36, 2004 May 26 (UTC)

vi looks weird when the monospace font doesn't look at all like the variable width font used for the rest of the text (most mono fonts don't look so good in between varwidth text). Leaving it as vi everywhere looks best imho. -- Kimiko 13:44, 3 Mar 2004 (UTC)

Is the screenshot vi or nvi?

Is the screenshot in this article "vi" or "nvi"? I don't think the original "vi" shows the status line at the bottom of the screen (Solaris (:version = Version SVR4.0, Solaris 2.5.0) and HP-UX (:version = HP Version $ B.11.23 Jul 15 2003 02:09:47 $) do not), whereas "nvi" (:version = Version 1.79 (10/23/96) The CSRG, University of California, Berkeley.) does. — Lady Lysine Ikinsile 09:01, 2004 Jun 20 (UTC)

I'll check at work tomorrow (on Solaris 8) and get a new shot if needed - David Gerard 19:45, 20 Jun 2004 (UTC)

Enormous discussion about whether vi is modal

Except, vi is not actually modal, or at least, that's how it was explained to me. "Don't think of vi as being modal!". It's actually a stateful filter :-). It's a very very fine distinction, but vi is based on ex, and you can actually still use it as a Filter_program! Kim Bruning 09:35, 21 Jun 2004 (UTC)

What, exactly, is a "modal editor", then? — Lady Lysine Ikinsile 09:36, 2004 Jun 21 (UTC)

What most people actually use vi as. vi is actually more capable than merely a modal editor would be, but most people don't see the distinction, and consequently don't make optimal use of vi. (Not that I'm so great at it :-P ) Kim Bruning 10:04, 21 Jun 2004 (UTC)

Are you referring to ex mode? Or ex commands in vi mode? — Lady Lysine Ikinsile 10:06, 2004 Jun 21 (UTC)

It's kind of funny to be asking me that , right at the moment where I was just claiming vi isn't modal. Having noted this, the following are all valid answers:

• yes.
• it doesn't matter.
• mu.

Ah well, you left your funny-bone at home today you say? Very well, heres a slightly more serious answer.

In vi, you can do things like:

20iI am not naughty^M^[

(where ^M is return, and ^[ is escape)

I think my school teachers knew about this before I did, because for some reason they always insisted that I use handwriting for this kind of thing.

The above works because it is interpreted as a single command by vi, and not as a sequence of mode switches like some folks might expect.

I hope that gives you a bit of an idea at least, I'm a bit too sleepy to go figuring out something more decent. I hope this gives you a first inkling. If it doesn't, please forgive me, and maybe I'll type something more enlightening tomorrow. :)

Have a nice evening! Kim Bruning 20:33, 21 Jun 2004 (UTC)

Well, actually the above would work anyway, because of the way Unix's keyboard/terminal buffering works: extraneous input is held in the buffer, and vi reads it as needed. So if you type "ifoobar^[", vi reads "i", enters insert mode, then reads "foobar" from the terminal buffer, then reads "^[". All this is handled outside of vi—it doesn't require any special code for this to work. Even if this wasn't the case (which I'd find unlikely, but who knows) the effect as observed by the user would be identical. So I don't believe there is any functional difference between a "modal editor" and what you're saying vi is, and non-functional differences are details of implementation. — Lady Lysine Ikinsile 14:41, 2004 Jun 22 (UTC)

There's 2 answers to this. I'll just do one for now: what does :

20ifoobar^[ 

do?

Try it. There's something a bit wrong there wrt assuming vi is a modal editor :-) I'm terribly sorry about not giving this my all, but for some reason I keep being either tired or have a headache when logging on to wikipedia recently. Kim Bruning 17:44, 22 Jun 2004 (UTC)

User-visible effect: vi types "foobar" 20 times.

What vi sees: User types "20i". Vi enters insert mode with an argument of 20. User types "foobar" then ^[. Vi types "foobar" 20 types, then leaves insert mode and enters command mode. What is not modal about this? — Lady Lysine Ikinsile 17:53, 2004 Jun 22 (UTC)

So which mode does vi enter when you typed 20? Kim Bruning 18:13, 22 Jun 2004 (UTC)

Why would it change mode for that? "i" is a command, "20" is the argument; it's "i" that changes the mode, although the full "command" can be considered "20i", as it's one unit. — Lady Lysine Ikinsile 18:16, 2004 Jun 22 (UTC)

Okay, well vi has a repeat last command. It's '.' (a period). So does vi now repeat 20i (and leave you in 'insert mode' like you claim) , or does vi actually repeat 20ifoobar^[ . Try it and see which? Kim Bruning 18:21, 22 Jun 2004 (UTC)

I'd like to ask you something. If vi isn't a modal editor, what is a modal editor? - David Gerard 18:23, 22 Jun 2004 (UTC)

Please be more careful when attributing words to other. I never "claimed" that the "." command does anything, since it wasn't discussed. Now, as to what "." actually does; according to my vi manual (vi(1)), the "." command is used to "repeat [the] last change" [1]. It doesn't mention last command, although I would think describing it as "repeating the last command" is a reasonable way to describe it without leading to claims that vi "is not a modal editor". Now, rather than continuing to provide vague examples, would you like to actually describe why, in your opinion, vi is not a modal editor? — Lady Lysine Ikinsile 18:29, 2004 Jun 22 (UTC)

It's a test to show what vi actually sees as a command. We're arguing about split hairs anyway, but well, this is an encyclopedia, so I guess that's ok just this once. You postulated that 20i is a full command. The idea was to show that in fact vi sees 20ifoobar^[ as the full command.

That's interesting, because it means that vi might actually see it as a single unit. If so, we can skip asking about mode changes, since they never actually come into the picture.

Summing up, vi is an ed descendant through ex. I don't think many folks seriously see ed (UNIX) (or ex) as being a true modal editor at all. Then people look at visual mode and just assume I guess.

Kim Bruning 19:14, 22 Jun 2004 (UTC)

So what you're saying is that you do understand that everyone else considers it a modal editor, you just have a private definition of "modal" that says it isn't. Settled then - David Gerard 19:28, 22 Jun 2004 (UTC)

Actually I hadn't addressed your question yet. :-) See the current article. Also see: mode under computer software. A modal editor is an editor in which the keys change their meanings depending on what mode the editor is in. Sounds a lot like vi, right? Well, not quite. Apparently vi isn't really a great example of a modal editor, based on that definition. I've heard people quite seriously explain why vi is not really modal. Somehow I've managed to maneuver myself into a position where I'm defending that side of the argument for once. :-P Kim Bruning 20:08, 22 Jun 2004 (UTC)

Is your entire platform for calling it not modal based upon the existence of the "." command? And if so, how does that make it not modal? You can't "skip around" the mode changes in the "." command because there aren't any. "." repeats the last change—that is, the result of the last command(s). There is, obviously, no mode change involved in that, even though "20ifoo^[" involves two mode changes. — Lady Lysine Ikinsile 19:37, 2004 Jun 22 (UTC)

Hmmm, I didn't directly mention '.' again in my last argument at all. But aren't you reaching a bit? What's easier to implement, just execute a single parsed command again, or try to figure out exactly what changes happened last (when is last?) combined whith which spaghetti tangle of mode changes?

I'm not really happy with me being the one to bring forward this dissent, it's just that I have known folks who did put it forward like this. (Specifically, the fellow who taught me vi :-) ). If I can just convince you it's not quite as cut and dried as you think, and that some people have a different opinion, well, that's all I'm asking for. Kim Bruning 20:08, 22 Jun 2004 (UTC)

Some people are also of the opinion that the earth is flat. I have yet to see any actual suggestion of why "modal" does not apply to vi, please provide one (not more "examples", please, an actual reason).

Apparently vi isn't really a great example of a modal editor, based on that definition -> why not? In what way do keys not change meaning based on the mode? (FWIW, mode doesn't say that 'keys change meaning', only that vi has 2 (3) modes.) — Lady Lysine Ikinsile 20:34, 2004 Jun 22 (UTC)

It is argued by some that vi doesn't really have modes. For some reason I'm defending that argument ;-) To try and do that, I've shown that vi has some behaviours that are not easily explained by modes. You tried to explain the example I gave, and got a bit of a spaghetti explanation, I tried it and explained in one sentance. I've also told you that vim is derived from ex, which is not a modal editor at all.

The current vi article is the article that talks about 'keys change meaning'. Also, the article points out that vim is the only vi which actually 'displays your editing mode', while all the others apparently don't. I've been told that vims' behaviour is actually considered incorrect in this respect.

Once again, I'm not the best person to defend this argument, but I'll try to do my best until/unless a better expert on this pov comes along. Kim Bruning 20:59, 22 Jun 2004 (UTC)

Okay, well I'm really spending way too much time on this, but here goes.

Consider these 3 lines:

#1
echo -e '$_="just another crazy hacker\\n";\ns/crazy/perl/;\nprint;\n '| perl

#2
echo -e 'a\njust another perl hacker\n.\ns/perl/ex/\nw! testme'| ex && cat testme

#3
echo -e 'ajust another perl hacker\E:s/perl/vi\n:w! testme\n:q!'| vi && cat testme

They all have similar input, and equivalent output. The thing is, there's probably not many folks who would say the top 2 lines are some form of modal input.

Based on what I think is being said here, you might claim that line 3 *is* modal in some way.

Okay, that's interesting.

So did I predict right, and will you be calling the 3rd line modal? If so, why? I'm interested in hearing your arguments. :-)

Quickly answering your question "why does modal not apply to vi", well, the answer to that was right at the start of this debate: vi is actually a stateful filter.

See if you can look at it this way: vi is a special mode of ex, which takes input in a specific way, and parses it as a special case of ex commands.

Kim Bruning 08:49, 23 Jun 2004 (UTC)

From the source code point of view. I have been working on 2 of the major vi clones, elvis and nvi, and they do not have anything that is called "mode", as in "insert mode" or "append mode". "i" and "a" are commands, not modes. As someone said in the above, "i" or "a" can be repeated by numbers (10ia<ESC> inserts 10 "a"s). I will have to check the real vi (now ex-vi) but i believe it goes the same,

Jun-ihiro itojun Hagino 12:42, 9 Sep 2007 (UTC)

In terms of readability for a Wikipedia article I understood "modal" instantly. While it may be a stateful filter I don't believe the initial description (About vi and Modal Editing) would be as easy to read for the average person were the editor to be described as a stateful filter. Marc Kupper (talk) (contribs) 19:41, 11 September 2007 (UTC)

Feature article?

Although I'm not a vi user (I'm a BBEdit guy), I'm wondering if there are plans to promote this article as a feature article, considering that Emacs recently got its 5 minutes of fame. Surely we can tap into the rivalry in a good way, as opposed to the typical flame war. :-)

The vi article is nowhere near ready for being a feature article. So anyone supporting such an idea had better a cunning plan for improving the article first. Nothing to stop you from putting that plan into practice. Rl 07:54, 6 Oct 2004 (UTC)

I actually like the brevity of the vi article in comparison to the Emacs one. Call me a sucker for the minimalism of vi. Plus this one is much more factual, to the point, and useful than that other editor.  :^)

Pronunciation

Some people actually do call it /vi/

Moved from main page:

But it fairly frequently is pronounced /vi/. Shouldn't Wikipedia be descriptive rather than prescriptive?

It's like the argument about what a particular city should be called; as the people who founded it / live there, or as outsiders might call it. In this case, the originators of the program always said /V-I/, so that's the official pronounciation in my book. As an aside, if you say /vi/ or /siks/, then you'll sound like a doofus to more experience Unix users, and you wouldn't want that, now would you? -- ansible

I take it it's meant to be pronounced "V. I", like the two letters. We need to fix this:

we-aye" = /vi aI/

as the two things don't agree. See SAMPA. "V.I." is [vi: aI]. The often-used but incorrect pronunciation is (I'm guessing) like the word "vye" as in "vye for his affections": [vaI]. COuls someone with knowledge of phonetics AND this program check? -- Tarquin

---

I'm generally in favor of descriptive-rather-than-prescriptive, but as a 20-year Unix veteran, I will verify that indeed, anyone who says "vye" or "six" will be instantly tagged as a clueless newbie. It's "vee eye". In fact, it's "vee eye" even when it's "vim" :-) I'm not that familiar with SAMPA, but I'll look it up and try to fix that. --LDC

If it's "V. I" like the two letters it's [vi: aI]. I'll fix that on the subject page. -- Tarquin

/vi/ again

The "vai" pronunciation appears to be standard usage at Sun Education in Melbourne, Australia. So I changed the entry to say 'generally' not pronounced that way. In any case, it's evident there's enough controversy around it to mention it - David Gerard 16:06, Jan 9, 2004 (UTC)

/vi/ isn't "vai". That'd be /vaI/. I have actually heard it being pronounced /vi:/ ("vee", like the letter V in English, or like "wee" with a V ionstead of a W) and vim pronounced as /vIm/ (like "whim" with a V instead of a W) in Germany, but I would wager that this pronunciation is uncommon in English speaking countries because /vi:/ is practically indistinguishable from the letter V, which would probably be too ambiguous.— Ashmodai ^{(talk · contribs)} 16:01, 3 December 2006 (UTC)

SAMPA

I've converted the SAMPA to IPA (without prejudice to the debate as to hte question how it should be pronounced - by the way from my UNIX days I always remember it being called "vye" as a single word - and removed the convertIPA template. rossb 14:22, 24 Feb 2005 (UTC)

I always remember it being called "vye" as a single word

I think that must be a regional dialect. :-) In the shops I've worked in, it's always been "vee-eye". But there is some support for your position to be found in the editors "vile" and "vim", which are definitely pronounced as words.

Atlant 17:06, 24 Feb 2005 (UTC)

For goodness sakes - it appears to be pronounced in one of two ways, can't we just accept that, get over it and have the article show the two options (thereby avoiding this whole silly schoolboy argument on the front page)? For the record, I call it vi. And always will do. —The preceding unsigned comment was added by 147.114.226.173 (talk • contribs) .

Just so I'm clear on this (perhaps my humour detector is broken), 147.114.226.173, which pronunciation were you trying to indicate when you said "I call it vi. And always will do."

Atlant 13:20, 7 July 2006 (UTC)

Sourcing for pronunciation

I have cited the pronunciation to a reliable source in a footnote. While I am sure that some people pronounce it differently, Wikipedia sourcing policy would require that the alternate pronunciation be cited to a reliable source as well, and I have yet to see one cited. Plenty of people pronounce "nuclear" as "nucular", but we don't put than in a Wikipedia article either. Anecdotal assertions like "The 'vai' pronunciation appears to be standard usage at Sun Education in Melbourne, Australia" are not encyclopedic. Wikipedia is not a dictionary of regional or subcultural English, so we don't need to record various people's random mispronunciations of things. So, if you find a reliable source that gives an alternate pronunciation, feel free to add it. Otherwise, it no more belongs in the article than "nucular" does. --MCB 22:36, 6 December 2006 (UTC)

Ironically, your "reliable" source also probably supports the alternative pronunciation; I don't have the printed version, but the Jargon File claims to be an online version of the said dictionary, and it gives "vi: /V I/, not, /vi:/, never, /siks/", which indicates by the variation of not and never that the second pronunciation is considered "incorrect", but since it is not "never" used, it clearly must be used.

Now, note that WP:RS is not a policy, and indeed states in bold text in the very first paragraph that it is not binding. The actual policy is WP:V, which requires only that it be possible to verify claims, not that every single claim be cited (otherwise it would be necessary to provide a reliable source for the claim that water is wet!)

In this case, it is very easy to verify the claim that some people use the alternative pronunciation: a simple Google search will return many, many instances of people arguing over the question of which is the proper pronunciation. It is consequently trivially verifiable, and a demonstrably controversial issue that it is reasonable to document on Wikipedia.

Thank you for adding a source for the claim that actually benefits from one, however (the claim that one pronunciation is considered correct, which, as a controversial issue, requires that we cite an authoritative opinion rather than stating a POV ourselves).

Incidentally, you may be interested to know that the "nucular" pronunciation actually is documented in Nuclear - as of course it should be. — Haeleth Talk 20:09, 9 December 2006 (UTC)

Please don't take this the wrong way, but I confess that I'm immediately suspicious of any argument that starts with "note that WP:RS is not policy...". Even if it were not strictly applied here, can you supply any reasonably plausible source that supports the "alternative" pronunciation? I mean, something other than hearsay, folklore, personal opinion, etc.? Your reading of the Jargon File/Hacker's Dictionary (the latter is the same as the online version, yes) seems very strained; using someone's authoritative statement "A, not B, and never C" as support for the legitimacy of "B" seems very odd to me.

The so-called "alternative" pronunciation /vi:/ is simply an error, committed by people who have never heard it pronounced correctly and but see it written. It's just a trivial mistake, not worthy of encyclopedic preservation. My example of "nucular" was perhaps not a good one; "nucular" is a highly notable mispronunciation (due to having been uttered by some prominent persons), and while it does have its own article (Nucular), it would be completely out of place as an alternative pronunciation in the nuclear physics article. /vi:/, on the other hand, is a non-notable mispronunciation, and any actual use of it is limited to people who don't know better, and I strongly disagree that there is any actual controversy as to the pronunciation, nor have I seen any sources, strictly reliable or not, verifying that. Descriptive or not, I don't think Wikipedia should be in the business of preserving trivial errors that arise from ignorance. --MCB 06:50, 10 December 2006 (UTC)

It appears you have failed to realise from my edits that I use the /vaɪ/ pronunciation myself. Now that I have made this clear, I would appreciate it if you would stop characterising me as an "ignorant" person who "doesn't know better" and makes "trivial mistakes".

Now, the reason it is relevant that WP:RS is not policy is that there is no consensus as to what actually constitutes a reliable source. I suspect, for example, that you would reject any of the numerous Usenet and blog discussions on the subject as "unreliable", though I personally believe they would be perfectly adequate to demonstrate that multiple pronunciations are in common use. Meanwhile, I am not convinced that the source you have provided is authoritative: the Jargon File is not a work of scholarship based on impartial research, but rather a list of ESR's personal opinions on how "real" hackers talk. It cannot be a reliable source for the unqualified statement that "vi is pronounced /vi:.aɪ/" - at best, it is a reliable source for the statement that "Eric S. Raymond states that vi should be pronounced /vi:.aɪ/".

Still, given how strongly you feel on this point, I have done as you asked and looked for published sources that document the fact that the pronunciation /vaɪ/ is common. This was surprisingly easy. The book I have cited in the article actually claims that /vaɪ/ is the correct pronunciation, with no mention of any alternative; you may think the author "ignorant", but the question here is not whether ESR would approve, but whether the pronunciation is in common use or not. — Haeleth Talk 17:18, 10 December 2006 (UTC)

A word of advice: If you come into my shop claiming to "know" Unix, and it isn't a case of English being a foreign language to you, and you pronounce the editor's name as "vie", I'll be very suspicious of your claims of "knowing" Unix. I have never heard an experienced Unix practitioner call the editor anything other than "V I". But don't let me change your pronunciation ;-).

Maybe someone can ask Bill Joy how he pronounces it? He's a bit more-authoritative on this topic than ESR ;-).

Atlant 19:23, 10 December 2006 (UTC)

Haeleth, be assured I had no idea you were among those who used what you would like to be considered a legitimate alternative pronunciation, and I certainly intended no personal attack; I assumed you were defending an abstract point. That said, this discussion is both painful and funny to me, since I was "in the room" (both literally and figuratively) when Bill Joy wrote vi in 1975-76, right there on the 2nd floor and 5th floor of Evans Hall. (As with csh and some other tools of that era, Joy had a number of informal collaborators, including students. I don't think I wrote any part of ex/vi, but I did have some code in csh dealing with directory handling). There is absolutely no question as to the correct pronunciation; vi and ex were named and pronounced in accordance with all the other two-letter fundamental UNIX utilities/apps in V6 and V7, like ls, cp, rm, cc, as, ln, m4, and so forth. All were, and are, pronounced as two-letter initialisms. Three- and four-letter commands, like cat, yacc, diff, and so forth, were pronounced as words.

It is a matter of bemusement to me that 30 years later, not only have some people invented a new pronunciation based on looking at the spelling (but being unaware or indifferent to its origin), but seem to want to defend it as somehow legitimate. Imagine for a moment that you're a French person, well-schooled in your history-steeped language, and you hear some American not only pronounce "cliché" as "clitch", but insist that it's a valid alternative pronunciation, in "common use", and ought to be cited as such in the dictionary. Wouldn't you be embarrassed for the poor American? --MCB 07:21, 11 December 2006 (UTC)

Inciting the editor wars

A recent editor added (regarding vim) equivalent in power to Emacs.

Now you've done it!

To Emacs partisans, nothing is as powerful as Emacs. The fusion power of a galaxy of suns powering Sagans worth of IBM Z/OS mainframe computers couldn't rival the power of Emacs. Surely, someone will be along to blast this statement soon.

Atlant 14:50, 23 Mar 2005 (UTC)

I'll quickly change it to roughly equivalent in power to Emacs, in that way, the statement becomes mostly harmless. ;-) Kim Bruning 17:16, 23 Mar 2005 (UTC)

Well done -- thanks!

Atlant 17:27, 23 Mar 2005 (UTC)

Alright, buster! Just WHERE does the Emacs even come close to the power of vi? - Anonymous, 12:02, 9 April 2006 (UTC)

Does this article contradict itself?

Did Bill Joy or Evans Hall write it?

Bill Joy wrote it while sitting in Evans Hall, a building at the University of California at Berkeley.

Atlant 13:04, 27 Apr 2005 (UTC)

Picture on main page?

Why is there a picture of the Vim startup screen on the vi article? The two really are different, and the picture that used to be up, which showed simply a blank screen with tildes, did a better job of portraying the minimalist nature of the editor. In fact, I do not believe that plain vi even has a startup screen.--Shutranm 18:50, 29 May 2005 (UTC)

I looked through the page history (not very thoroughly) and couldn't find a page version showing regular vi, so I made my own. And you're right that there's no startup screen; I'd better go amend the caption. - mako 19:56, 29 May 2005 (UTC)

license?

what is the license of vi(m)? Is it GPL? —Preceding unsigned comment added by Gflores (talk • contribs)

nvi (and presumably the original vi) are BSD-licensed. vim is under a custom licence called the vim licence. 207.245.124.66 15:21, 14 December 2005 (UTC)

Though Vim does have a seperate license, its license is GPL compatible, as noted on Comparison_of_text_editors. ~Linuxerist E/L/T 03:09, 20 May 2006 (UTC)

I can find no evidence that vi is currently or has every been licensed under the GPL or any other open source license. Although it was originally written for BSD, AT&T's copyrights on portions of the code (see USL v. BSDi) spurred a rewrite which became the BSD-licensed nvi. Unless someone can find evidence otherwise, it should be removed from Category:Free text editors. Hamitr 04:09, 16 February 2007 (UTC)

The ed(1) sources that the original vi used have been available under a four-clause BSD license since 2002, when Caldera released pre-System III Unix under those terms[2]. As mentioned in the article, the Heirloom Project has taken the vi codebase[3] and ported it modern Unix-like OSes, released on those same terms -- although not GPL-compatible, it meeets the FSF freedoms, DFSG, etc. Thus, both the original vi and most reimplementations are available as Free Software, so vi can rightly be categorised as free. Ianking 09:00, 16 February 2007 (UTC)

Thanks for the clarification. It should remain as is. Hamitr 14:21, 16 February 2007 (UTC)

Removing the Trivia Section

I think the Trivia Section material should be preserved; it's very useful. Integration into the main article is fine, but removal would discard valuable information. Where else would you verify that the keyboard upon which vi was written had no cursor keys, and indeed see a layout of the keyboard? Thanks 4.235.36.117 14:24, 26 April 2007 (UTC)

Image copyright

How can such a simple image be copyrighted? I can understand the program itself, but a bunch of tildes and slashes I do not think qualifies. Can I make my own copy (from scratch) and license it as public domain? -Henry W. Schmitt 08:24, 27 September 2007 (UTC)

No need. The image was mistagged; I've corrected its copyright information. Chris Cunningham 08:54, 27 September 2007 (UTC)

Thanks, but I'm not sure I understand this license at all.

This screenshot is of a program that has been released under a free software license. As a derivative work of that program, this screenshot falls under the same license.

Free software license: BSD license Copyright © belongs to the uploader, all rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: Redistributions of source code must retain the above copyright notice, with the name of the uploader, and this list of conditions; Redistributions in binary form must reproduce the above copyright notice, with the name of the uploader, and this list of conditions in the documentation and/or other materials provided with the distribution; Neither the name of the uploader nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

The copyright belongs to the uploader, but the license is the same as the software? --Henry W. Schmitt 09:20, 27 September 2007 (UTC)

Yes? This is pretty universal with free software / free content licences. As for the "series of tildes and slashes", the argument is that as an image file the work is just a binary blob: you can't copyright an apple, but you can certainly copyright a picture of one. Chris Cunningham 09:39, 27 September 2007 (UTC)

Comment on Howto/Manual

Even though someone make complain that there is "how to" information here, the material which is presented is intended to show how the orthogonal command+motion is built up. While there were contemporary editors which could combine things, the keystroke-oriented editors tended to be haphazard in design (teco, for instance), and did not spend an appreciable amount of the design to making groups of functions work together. Inevitably, an attempt to convey this purely in words will run into looks-like-a-manual. Tedickey 14:30, 20 October 2007 (UTC)

The biggest problem is that it's written in a tabulated, manual-style format. There's no need for that. The whole motions section could be contracted into a section of prose which describes what commands are available and how they were arrived at without having to lay them out for the reader in a manul-style format. I'd argue that moving the history section up above the command list would be a good start. Chris Cunningham 14:43, 20 October 2007 (UTC)

You suggest that the information be convoluted in dense impenetrable prose instead of the current easily browsable format? Why on earth? Mlewan 16:30, 20 October 2007 (UTC)

Yes, I definitely suggest dense impenetrable prose. Sigh. Chris Cunningham 20:23, 20 October 2007 (UTC)

As I read the comment, he objects to having tables (presumably bulleted lists also are not good form ;-). Regarding moving the history section - that might serve to better introduce H,J,K,L but would leave the Derivatives and Clones section even more contorted than it has become over the last few edits. Tedickey 16:56, 20 October 2007 (UTC)

The derivatives section is pretty short already; I don't think there would be any difficulty or loss of clarity in converting it to prose. Chris Cunningham 20:23, 20 October 2007 (UTC)

Let's assume that a block of text would be as easy to read as a table (something that is very questionable). The question remains why you prefer prose. Mlewan 20:39, 20 October 2007 (UTC)

Because Wikipedia prefers prose. "Easy to read" is highly subjective at any rate; a manual-style layout is opaque to most all non-Unixees. Chris Cunningham 00:48, 21 October 2007 (UTC)

"Because Wikipedia prefers prose"! What kind of argument is that? "Wikipedia" prefers nothing. Wikipedia is a software. Wikipedia readers have preferences - and they most certainly include clarity, not some cabbalistic belief in prose.

"Easy to read" is not subjective in the sense that you can say what you want. For each text and reader situation, you can measure what text structure is optimal. In no case I have heard of, impenetrable prose would be better than a logically structured list or table. For more reading check Readability and legibility, Readability, Usability, Knowledge visualization, Information Design, Graphic Design, Communication studies, Optimalweb, Lunatech Research, and the links from those. Mlewan 07:12, 21 October 2007 (UTC)

A program description is not necessarily a manual

I'm not going to go into any edit war here, but I strongly disagree with this edit.

There is a vast difference between a user manual and a description of how a program works. What sets vi apart from other programs is the way commands are called - with simple short key strokes. It would not make any sense to write in an article about Word, that the program uses ctrl-c to copy text, as thousands of other programs use the same shortcut. However, when it comes to vi, it is of interest to see how illogical a program that has been used by millions of people over several decades works. Clearly, it is a very efficient approach for people who learn it, even though it probably has caused more grey hairs than all efforts of squaring the circle the last few thousand years.

That is in fact almost the only interesting thing with the program. The fact that there is a text editor called vi is of no particular interest at all. There are plenty of them.

The only information that properly conveyed this information was removed in that edit. Mlewan (talk) 21:01, 27 October 2008 (UTC)

Well what you are talking about is then original research. If this program is illogical as you describe, then notable 3rd party sources will have written about it and prose can be written on that basis. --Cameron Scott (talk) 21:08, 27 October 2008 (UTC)

Prose is often obfuscation. It is one of Wikipedia's silliest and most madly applied rules to try to make text as convoluted as possible to degrade readability. Prose works in some circumstances. This is not one of them.

What I am talking about is not original research. It is giving facts - indisputable facts about how the program works. Research could be performed or inspired based on facts, if they were there. Now they are gone for no valid reason. Mlewan (talk) 21:22, 27 October 2008 (UTC)

Saying "facts cannot be OR" is an indication of a lack of understanding of what constitutes OR. Everything on Wikipedia is supposed to be factual: however, those facts must be drawn from extant secondary sources, not direct analysis of the subject.

The section contains a large amount of banal instructional content which is of zero value to anyone who isn't looking for a guide to using the program. As such, it has no value to an encyclopedia. Elimination of all the keystroke discussion with the exception of an overview of the basic positioning of the most common functions would be a good start in turning what is current a manual into a useful discussion of what makes vi's design unique. That's the route I'd go, and assuming that nobody else is willing to do the work it's how I'll proceed. Chris Cunningham (not at work) - talk 22:18, 27 October 2008 (UTC)

1991??

Quoth the article:

It is still widely used by users of Unix variants (about half the respondents in a 1991 USENET poll preferred it).

Heh, heh. Surely there must be a more recent and/or accurate set of numbers out there? --Jfruh 00:17, 29 July 2005 (UTC)

Ya use what propaganda ya gots, n'est pa?

Seriously, among the Unix crowd I hang around, vi (or its more-recent variants) is still very competitive as an editor.

Atlant 11:16, 29 July 2005 (UTC)

Oh, I know it is. It's just that if you want to reinforce stereotypes about said crowd, I can think of now better way to do it than to use a 14-year-old Usenet poll as if it were relevant evidence :) --Jfruh 13:15, 29 July 2005 (UTC)

Maybe these (or one of them) could replace the usenet poll reference? Linux Journal 2008 Reader's Choice Awards. It also won in 2005 and 2001 and probably most every other year. It also won the 2008 LinuxQuestions Members' Choice Awards and the 2007 and so on. Though these tend to go to the "vi-family" as not even the BSDs use the actual bona fide vi. —Preceding unsigned comment added by 74.162.131.216 (talk) 04:37, 3 August 2009 (UTC)

Full Form Of vi

Hi,

I wanted to know the full form of vi..... coz all commands of unix or linux has some or the other meaning n full form so i guess vi also shud hav sm....??? i read tht vi is a visual mode of ex or something but this wasnt really answer of my question!!!!!

~20 Aug 2009 VIKI —Preceding unsigned comment added by 116.72.26.243 (talk) 18:28, 20 August 2009 (UTC)

Interview with Bill Joy http://web.cecs.pdx.edu/~kirkenda/joy84.html In the "History" section there's something need to be cited.And above is the url you would find the info. Below is the text: "Mike Horton brought his editor along from Bell Labs called hed for "Horton's editor." He was disappointed when vi won out over it. But vi had momentum with the local users - and Mark, somewhat out of frustration, went out and actually supported vi. That was nice, because I didn't have the patience to do it anymore. Just putting the termcap entries in that people would mail me would take hours a week, and I was tired after three or four years." In the context,it's around 1979.For the interview is taken in the August 1984. Hope it helps.Sorry for that i actually don't have that much time and ability to do the citation. :) —Preceding unsigned comment added by 222.188.136.177 (talk) 00:59, 26 August 2009 (UTC)

lowercase title

Ok twice now I've tried to add the note "The correct title of this article is vi. The initial letter is shown capitalized due to technical restrictions." and it shows up in the preview but when I hit save it isn't there. Anyone know why this is happening? Thanks. Wolfrock 20:38, 31 March 2007 (UTC)

It is working. Note that you need javascript to have it working. -- ReyBrujo 21:20, 31 March 2007 (UTC)

You people are nuts. Yes, the name is normally written vi and it makes us feel like we're on the inside when we point that out. But come on. Do you know the Wikipedia article about chickens has the title Chicken? Do you know why? Because it's a title. Furthermore, if you write the word "chickens" at the beginning of a sentence it is written "Chickens" because words are capitalized at the beginning of a sentence. Just because vi is normally written as a lowercase word doesn't mean it has some kind of force field that repels the elementary rules of usage and grammar. The correct title of the article is Vi. At the beginning of a sentence you should write Vi. —Preceding unsigned comment added by 74.7.179.158 (talk) 15:42, 9 September 2009 (UTC)

Instructional material removed

Just a reminder that instructional material, user manuals and guides, etc., belong in Wikibooks rather than here. (Please see WP:NOT.) I userfied the command guide that Hydratab inserted as User:Hydratab/vi-commands so that it can be added to Wikibooks if desired. Cheers, --MCB 17:49, 20 June 2006 (UTC)

I've tagged the main usage section for this. Despite the disclaimer, it's still a bit in-depth. We shouldn't have to describe how a feature is invoked. Chris Cunningham 12:53, 24 September 2007 (UTC)

This wiki page is still going to the dogs, most of the article (at this time of writing) should just be scrapped. -- 68.119.62.140 (talk) 21:50, 23 March 2010 (UTC)

The Register article about vi

Check http://www.theregister.co.uk/2003/09/11/bill_joys_greatest_gift/, which reproduces part of a previous interview. Some data can be included here. -- ReyBrujo 16:37, 6 January 2007 (UTC)

In particular, the following quote could be added

People don't know that vi was written for a world that doesn't exist anymore - unless you decide to get a satellite phone and use it to connect to the Net at 2400 baud

— Bill Joy, Linux Magazine, 1999

167.107.191.217 (talk) 14:42, 16 April 2010 (UTC)

210.77.x.y

Clearly someone with a dynamically assigned address in the Chinese netblock 210.77/16 wants to link to their blog about vi. Is it possible to create a rangeblock on this specific article? -- Autopilot (talk) 17:36, 10 April 2011 (UTC)

I'd like to WP:AGF but it is edit-warring. I've made a request for page protection. Msnicki (talk) 17:44, 10 April 2011 (UTC)

Sounds good. TEDickey (talk) 17:44, 10 April 2011 (UTC)

Under the cursor vs. position specified by the cursor

From the vim docs:

["x]x                   Delete [count] characters under and after the cursor
                        [into register x] (not |linewise|).

How about "under and after the cursor"?—Trevor Caira 14:40, 8 August 2005 (UTC)

I'm sorry, but I think the article ("specified by the cursor position") is perfectly clear (at least in this regard) as it stands now. I'd also accept "indicated by the cursor". But trying to identify/describe what the cursor is and how it indicates the indicated character is doomed to fail as various terminals and termulators do this in varying ways: blinking blocks, reversed text, underlined text, "I-beam insertion points", and so on...

Atlant 14:51, 8 August 2005 (UTC)

I think that phrasing is vague (i.e. before or after the cursor?). Which character is indicated by the term "under the cursor" can be explained somewhere else in the article (maybe in a neighboring phrase or sentence), but it is the standard explaination of which character the command 'x' deletes, and once you understand which character is under your i-beam/block/underline, I believe it is the appropriate explanation of this command for Wikipedia. It may be worth noting that Google has: `Results 1 - 10 of about 4,380 for vi "delete the character under the cursor".'—Trevor Caira

How about: ' "x" to delete the character under the cursor (where the position "under the cursor" may be to the right of an I-beam, over an underscore, or under a block character depending on the terminal) '—Trevor Caira 22:34, 8 August 2005 (UTC)

Adding a comment six years later ..... The whole approach of the vim documentation is wrong. You start with looking at what the 'x' command by itself does, which is, it deletes the character under the cursor and moves to the next character. You then apply the general vi rule that: a number in front of a command means "repeat that command [number] times". Old_Wombat (talk) 05:24, 8 January 2012 (UTC)

Pronunciation

Citing Raymond's TAOUP might at first seem a WP:RS, but reading the paragraph noted, Raymond betrays his ignorance of vi (he's a known Emacs user) by commenting that the various implementations of vi provide the same features. Raymond is in the quote relevant to this topic regurgitating the comments from the Jargon File. For this topic, a more suitable reference (discarding the jokes about "six") would be Arnold & Lamb, or Arnold TEDickey (talk) 08:33, 24 May 2011 (UTC)

Not a WP:RS, but providing some useful material for further investigation is this wiki TEDickey (talk) 08:35, 24 May 2011 (UTC)

I would like to observe that people are claiming that the program is pronounced in a certain way based on writing in a number of books written from a local area in the USA. If anything, the sources only state that "The following people believe that the editor is pronounced in a certain way". The sources absolutely do not indicate that the editor is ACTUALLY pronounced in any certain way, as they do not refer to things such as international surveys (or even locality surveys). This misappropriation of sources is very invalid and anyone who sticks to them is ignoring reality. 115.64.159.41 (talk) 11:43, 26 June 2013 (UTC)

screenshot

The "Vim-example.png" is noticeably not vi because of the status line. That makes the text shown asserting that it is vi, misleading (probably the figure itself should be removed from Wikipedia) TEDickey (talk) 23:29, 22 February 2014 (UTC)

The replacement image [[File:NetBSD 6.1 vi C Hello World.png]] is suitable for this topic TEDickey (talk) 11:23, 23 February 2014 (UTC)

returning to insert mode

``and "i" to return to insert mode

This phrase makes it sound as if the default mode in vi is insert mode, which it is not. Perhaps it would be more appropriate to rephrase this paragraph to avoid suggesting this—Trevor Caira 00:31, 11 August 2005 (UTC)

This has been done already - some time in the 9 years (!) since this talk page entry was written 136.186.72.46 (talk) 00:01, 31 March 2014 (UTC)

Regarding Linux Journal survey

Hello,

The lede of the article says the following:

A 2009 survey of Linux Journal readers found that vi was the most widely used text editor among respondents, beating gedit, the second most widely used editor by nearly a factor of two (36% to 19%).

This seems to me highly misleading. The survey in question did not have vim as an option, and given that it's more commonly used than vi in my experience, it doesn't make sense to say that "vi was the most widely used text editor among respondents", since that number doubtlessly includes many vim users.

I tried to remove this sentence and was reverted (because it was "reported correctly", which is true, but it's still misleading). I then tried to add the caveat that the survey did not include vim as a separate option and was reverted again.

I think it should be removed or modified to point out the fact that it encompasses all vi variants. InverseHypercube (talk) 20:22, 6 April 2014 (UTC)

Your recommendation, then, is to pretend that the source said something other than what it does. If you don't have a source which says what you want it to, I suppose that must seem the right thing - to you. TEDickey (talk) 20:35, 6 April 2014 (UTC)

No, what I'm saying is that we should not include information that we know to be misleading. I suppose that you concede that the vi category includes many vim users? InverseHypercube (talk) 21:56, 6 April 2014 (UTC)

not necessarily: there's no control on the survey to ensure that only newbie Linux users respond. That would be necessary to guarantee your assertion. Knowledgeable users (particularly developers) would likely include people who use multiple platforms, e.g., the BSDs, undermining your assertion. Without a WP:RS giving actual numbers (none has been suggested), there's no way to determine how many are using vim, per se, or (as in the vast majority of users) using pretty much just the plain vi commands. Bear also in mind that surveys will likely be biased to report students more than the (much larger) fraction who are not. TEDickey (talk) 08:02, 7 April 2014 (UTC)

Rand Editor

• http://www.edm2.com/index.php/Rand_Editor
• http://www.rand.org/pubs/notes/N2239-1.html
• http://terminals-wiki.org/wiki/index.php/HP_2645A
• https://github.com/blakemcbride/Rand-E-Editor
• https://github.com/dspinellis/unix-history-repo
• https://www.twitter.com/enf

Xb2u7Zjzc32 (talk) 07:20, 19 December 2016 (UTC)

That's Cholmondeley pronounced Vee Eye

"The name vi is pronounced..." it says. I have used Vi for the last seventeen years and I don't believe I have ever heard it pronounced Vee Eye no matter what Eric Raymond wrote in two of the references provided in the article. I don't really care how people pronounce it but it is so annoying in Wikipedia articles when editors try to impose their vision of the world on the rest of us. No doubt Vim should be pronounced Vee Eye Em... — Preceding unsigned comment added by 81.187.233.172 (talk) 19:05, 6 June 2012 (UTC)

How is that different than you trying to impose your vision of the world? Simply find reliable sources giving alternate pronunciations. tedder (talk) 19:13, 6 June 2012 (UTC)

vee eye is the common, but not exclusive, pronunciation everywhere I've worked, although oddly vim seems to be most often pronounced as the single syllable "vim". -- Autopilot (talk) 01:42, 7 June 2012 (UTC)

I used to be a software developer. First time I was in charge of a multi-person project, I named the project using a term from some American SF-author´s novel. Sort of like a clockwork orang/orange insider joke. I never told anyone the solution of my own joke. I feel the need to tell you about the kind of mindset I have to presume in other developers: Are you aware of the existence of Romam numerals? When Knuth insist on the x in Tex to be pronounced as the Greek letter chi, he´s building up the punchline for the vi-joke; he wants programmers to have rudimentary knowledge about the Green and Latin symbols used in mathematics. This leads us to Uncle Kenny and the tea age eggs eleven eighty seven mentioned at the end of 'Last Friday Night' music video. Easter egg(s) never ever marked the spot: South Park season 10 episode 12 explained, the guys from Room A113 have been waiting for years for some outsider to figure out their little pun: What term did Hollywood invent as their synonmy for conspiracy theory? Area 5 1, quinque unum.

Unfortunately, there are no references regarding the idea of first coming up with a pun, then inventing some pseudo-explanation for the use of that pun word; except for possibly Lover Come Back (1961 film) and a connection between the Intel i5 processor and its codename.

Have you ever wondered why people bother with hiding easter eggs? It seems to be so much easier to just put them on the breakfast table. In order to create an easter egg, step 1, you have to think of a pun. Step 2, you have to invent some bogus explanation for using the pun, as a distraction: When you use the pun word, people ask about its meaning. If you give them a red herring answer, they stop asking. Until you provide a seemingly more reasonable explantion. Eg, before the age of the internet, there used to be a concept, Citizens Band, cb. Then you think about ways of transferring this concept into the information age. Then you look at the keyboard in front of you, and wonder: the key between c and b, how do you called it? When trying to decode other people´s puns, you have to guess their reasons for creating that easter egg: If you want the cooperation of other people for a project, you need to avoid the image of being selfish and arrogant; if your signature is too evident in that project, you failed. If you find a sufficiently cool way to hide it, if you do it in a way that prevents you from ever being able to convincingly prove the existence of the signature, then it just might word. So, isn´t a triple x not just sort of an extended form of a double x?

So, let´s talk about the Elvis-Lives-Conspiracy. Elvis, his songs, his ideology, became more respected after his supposed death. Do those people who support Elvis mostly want to profit from 'I am like Elvis/support me'. How much of their own money/image/ideological-cleanliness are people actually willing to risk in order to keep the idea alive that Elvis just pretended his disappearance in order to find out how the public responds? — Preceding unsigned comment added by 2001:4DD7:C4DB:0:20F2:4C91:FF53:2DF2 (talk) 21:44, 20 July 2017 (UTC)

Please note that the pronunciation "vee eye" is not proven to be prevailing by any sources given. They merely suggest that the authors of a few books prefer "vee eye". Autopilot, perhaps you should actually examine the "reliable sources" you attempt to appropriate for your particular viewpoint. 115.64.159.41 (talk) 11:44, 26 June 2013 (UTC)

xvi

xvi is a very small cross platform editor which might deserve a mention. You can find it at https://martinwguy.github.io/xvi/ and it's still maintained. 2001:985:d04c:1:a71:90ff:fe2e:dc90 (talk) 11:02, 13 June 2020 (UTC)