The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was Withdraw nomination because APL's functions are more like operators and almost entirely define its syntax, which is notable. And no consensus for Delete. --Cybercobra (talk) 19:21, 13 August 2009 (UTC)[reply]

APL function symbols (edit | talk | history | protect | delete | links | watch | logs | views) (delete) – (View log)

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

Topic not notable in and of itself. WP is not a programming manual; WP is not a directory of built-in functions in a programming language; WP:IINFO Cybercobra (talk) 04:51, 11 August 2009 (UTC)[reply]

Merge into APL would also be acceptable. --Cybercobra (talk) 13:56, 12 August 2009 (UTC)[reply]

Is merge something special or just cut and paste? If the dictionary could be included in a way that it was not immediately visible but could be revealed, similarly but opposite to the way the TOC can be hidden, I think that'd be great. But lengthening the APL (programming language) page by several screensful while we're hoping to remove half of it would not be a good idea. Phil Last (talk) 18:07, 12 August 2009 (UTC)[reply]

• Delete. Not an encyclopedia entry, practically a how-to guide. Hairhorn (talk) 05:24, 11 August 2009 (UTC)[reply]

• fairly emphatic Keep. I believe this is a valuable page for Wikipedia to have. It is quite hard to find this information on the internet (part of the reason I have contributed to this page), for example:

try Googling for "apl symbols", or "apl functions". See how many pages actually describe operators, and how many of them...

the canonical reference, ISO 8485:1989, costs ~260 USD.

As far as notability is concerned, I think that while this may not seem like it is enough to justify having a separate article, merging this content into the APL article itself would clutter the (already large) article too much. I do not believe that WP:NOTDIR applies, for example, we have separate articles for all of SQL's statements, articles on each C standard header (stdio.h, math.h, stdlib.h, etc), even articles upon printf and suchlike.

What I would like to see is this page extended to separate APL function symbols introduced in the standards from those specific to individual implementations. This would be closer to what other programming language pages do. —porges_(talk) 06:37, 11 August 2009 (UTC)[reply]

If you're going to expand the article, please put a coherent lead on it. Wikipedia is not a collection of articles for experts only. Hairhorn (talk) 07:13, 11 August 2009 (UTC)[reply]

WP:GHITS is not a good way to gauge notability. Where are the reliable independent sources? --Cybercobra (talk) 19:40, 11 August 2009 (UTC)[reply]

Huh? What does this have to do with my comment? I wasn't using Google to gauge notability. —porges_(talk) 09:41, 13 August 2009 (UTC)[reply]

• Keep. While the creating editor of this entry seems grossly to misunderstand the purpose of Wikipedia, and the article lacks anything much like an appropriate lede, not only is the subject legitimate but there is extensive content present that could in fact be part of an excellent article. —SlamDiego_←T 08:17, 11 August 2009 (UTC)[reply]

Comment why? What kind of improvements do you see possible for this article? What here is worth saving?--RadioFan (talk) 16:15, 11 August 2009 (UTC)[reply]

Could you explain why you consider the subject legitimate? --Cybercobra (talk) 21:07, 11 August 2009 (UTC)[reply]

• In reply to both of the above queries: APL isn't simply a programming language, but a important notation. (It has often been joked that it is a notation and not a programming language.) Providing a proper exposition of it qua notation is like providing a proper explanation of the IPA, and the tables in this article, though not as important as that of “Table of mathematical symbols”, are not unlike it. (I'm not here trying to be waxy, but to explain the rôle concisely by analogy.) —SlamDiego_←T 21:23, 11 August 2009 (UTC)[reply]

• Delete wikipedia is not a manual. I dont see how this article could be improved to be anything to look like anything but a manual RadioFan (talk) 12:21, 11 August 2009 (UTC)[reply]

• Keep per SlamDiego. The article needs work, but it shouldn't be removed. Irbisgreif (talk) 15:37, 11 August 2009 (UTC)[reply]

• Keep I also agree it should be kept, possibly with some work. I have used apl in the past but am not an apl partisan. The functions in apl were not determined on an ad-hoc basis; Iverson put a lot of thought into producing a small, very general, orthagonal set. These functions contribute as much to the unique character of apl as the "funny characters" and showing people the set of them provides valuable insight into the design and early former appeal of apl. It also greatly increases the value of the code samples in the main article since without descriptions of the functions, about all the code samples tell the reader is that apl code is short and cryptic. —Preceding unsigned comment added by 4.227.243.220 (talk) 19:00, 11 August 2009 (UTC) — 4.227.243.220 (talk) has made few or no other edits outside this topic. [reply]

• Strong Keep. I believe this is a valuable page for Wikipedia to have. A lot of people ARE working hard to make it better. —Preceding unsigned comment added by Clathwell (talk • contribs) 20:56, 11 August 2009 (UTC) — Clathwell (talk • contribs) has made few or no other edits outside this topic. [reply]
  • WP:ILIKEIT isn't a very good, objective argument. --Cybercobra (talk) 21:04, 11 August 2009 (UTC)[reply]
  • WP:ILIKEIT I can list the people who are working on it if thats what you're getting at - but you already know who - I'm guessing. I saw a guideline about not dishing new contributers too.--Clathwell (talk) 17:52, 11 August 2009 (EST)
    • I'm just pointing out it's an argument to avoid in deletion discussions. Your rationale for keep does not explain why or how the page is valuable, only that you believe it is; without such reasons, it falls under WP:ITSUSEFUL. Whether it's being worked on is not directly relevant to whether it should be deleted. --Cybercobra (talk) 22:04, 11 August 2009 (UTC)[reply]
• Keep. Roger Hui (talk) 21:27, 11 August 2009 (UTC)[reply]
  • Deletion discussions are not simple votes. Please provide a rationale for your vote. --Cybercobra (talk) 21:51, 11 August 2009 (UTC)[reply]
  • This page provides information about APL without making the main APL page overly large. If that page is not deleted then neither should this one. Roger Hui (talk) 00:13, 12 August 2009 (UTC)[reply]
  • I note that Python also has separate Python (programming language) and Python syntax and semantics pages (among other Python pages). For APL, the symbols are an integral part of the language, part and parcel of its syntax and semantics. Roger Hui (talk) 00:24, 12 August 2009 (UTC)[reply]
    • But those articles do not list such trivia Python's built-in functions (e.g. len(), hash(), etc.), which I find comparable to the content of this article. I do acknowledge that the nature of APL complicates such a distinction a bit. --Cybercobra (talk) 01:59, 12 August 2009 (UTC)[reply]
• Very strong keep deleting this article would be like saying we could have an article on Russian without one on the Cyrillic alphabet, or Greek without the Greek alphabet, or Arabic without the Arabic script. These articles, incidentally, may provide some examples of how this article could be extended and improved; the ideas behind the notation are important. --NSH001 (talk) 22:20, 11 August 2009 (UTC)[reply]
• keep. Truth to tell I didn't even know it existed but as there are pages for every other alphabet then it would be stupid not to have one for APL which, contrary to the belief of those who insist on emphasising the historic link between its name and the intials of a book title, is not solely a programming language but remains a mathematical notation which was its original purpose. Given that the page does exist it's a perfect excuse for me to remove at least one whole screen full of ancient history from the APL (programming language) page. Phil Last (talk) 07:43, 12 August 2009 (UTC)[reply]
• keep. The function symbols are unique to APL and a core property, that distinguishes APL from other languages. —Preceding unsigned comment added by ReinhardKarcher (talk • contribs) 07:49, 12 August 2009 (UTC) — ReinhardKarcher (talk • contribs) has made few or no other edits outside this topic. [reply]
• keep. the symbol set is not equivalent to "Python's built in functions" - without an understanding of the APL symbols you cannot use the langugage. This page should be regarded as a Rosetta Stone - without it the readers' understanding is crippled. crishog - 9:15, 12 August 2009 BST — crishog (talk • contribs) has made few or no other edits outside this topic. - very true, but I helped to devise the new wording of the main APL article. Also I have been an APL developer for over 30 years, so I would think I know what I'm talking about - crishog - 9:15, 13 August 2009 BST
• keep. The information on the page is needed - by providing a complete list of functions it enables viewers to think things like "Oh look matrix inversion is built in". It also shows the wide range of APL functions. It merits its own page because the "APL home page" seems rather cluttered and I support the recent efforts to reorganise it. ApMog (talk) —Preceding undated comment added 11:53, 12 August 2009 (UTC). ApMog (talk) 12:53, 12 August 2009 (UTC) — ApMog (talk • contribs) has made few or no other edits outside this topic. [reply]
• Comment - this is an example of a truly borderline case of what should be included. I'd go with whatever the consensus decides. Bearian (talk) 13:47, 12 August 2009 (UTC)[reply]
• keep. It not very easy to understand what APL is about unless you have a function list. DavidSol (talk) 02:37, 13 August 2009 (UTC)[reply]
  • It's not very easy to understand Python either without knowing the built-in functions, but WP doesn't list them anywhere because we're not a manual. Also, WP:ITSUSEFUL. --Cybercobra (talk) 03:06, 13 August 2009 (UTC)[reply]
    • I disagree - Python syntax and semantics covers what is on this page and more, I think, as it defines the behaviour for various Python operators instead of just listing them. Perhaps I could go through and add some further explanation for each operator, which would make it more useful? I'm just not sure why an article on Python syntax is much different than this page.DavidSol (talk) 15:35, 13 August 2009 (UTC)[reply]

Lexical vs symbolic programming languages (Keep)

Credit here to Cybercobra for pressing his point, which exposes a key distinction of APL.

In lexical programming languages, where the primitives are words, usually English words, the distinction between the core language and the libraries that install with it is of slight importance. To read a program in eg PHP, with its large vocabulary of core and library functions, I need the manual, nothing less.

APL is a symbolic language. Its primitives are more like an alphabet than a vocabulary. As mentioned, Iverson began APL as a mathematical notation, paying close attention to orthogonality and generality. The result has been great stability in the core notation. The relatively high level of abstraction of the alphabet and the consequent short length of many APL programs has lessened the value of accompanying libraries.

Most APL interpreters ship with libraries, but there is not the standardisation seen with eg C or PHP. The economics of writing software in APL frequently favours writing a program over finding one.

Consequently, unlike lexical PLs, one can read most APL programs knowing only the core alphabet, much as one can pronounce Serbian or Russian knowing only the Cyrillic script.

This article should be retitled 'Table of APL symbols', following Table_of_mathematical_symbols as a model. (Functions should be avoided because APL observes Heaviside's distinction between functions and operators – higher-order functions.) The model explains how to read the common mathematical notation. (Note that it already links to the Iverson_bracket from APL.) It may be helpful to recall here that as 'Iverson notation' this notation was in use before its implementation as a PL. (That implementation compelled a linearisation of the notation that Iverson credited as the source of much simplification and generalisation.)

If this article were not on WP, it could be posted on the [APL Wiki], the resource for practising and aspiring APL programmers. But, as a canonical guide to reading APL programs from the last five decades, WP is the better place for it. And, yes, it should list and describe the modest number of dialectical variations.

5jt (talk) 10:07, 13 August 2009 (UTC), Stephen Taylor, Editor, Vector, the Journal of the British APL Association[reply]

Note: Have moved/renamed the article to APL syntax. Assistance in correcting the nomenclature used in the article would be appreciated. --Cybercobra (talk) 19:21, 13 August 2009 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.