Talk:GOTO

From rfd discussion

I would suggest that this is anything other than a token in certain programming languages (such as Basic) rather than standard English. Similarly, I might say "You should have a bool as your return type, not a void", but that does not make the C++ tokens "bool" and "void" English words. In fact, it would be more correct to write: "You should have a 'bool' as your return type, not a 'void'" (or to drop the indefinite articles altogether: "You should have 'bool' [...]"). So it is with "this is a goto", in my opinion. Is "goto" really used to mean "an unstructured branch"? &mdash Paul G 15:36, 12 July 2005 (UTC)

• Likewise for "printf" in dmh's discussion above. Incidentally, if "everyone knows what printf means", why include it in a dictionary? :)
• Well, I know that I am in a minority (possibly of one) in thinking that programming languages are "languages" and that we should have entries for GOTO, PERFORM, COMPUTE, IF etc. In this particular case, I would keep it (improve the definition though) and possibly add something like GOTO-less to describe structured programming. SemperBlotto 17:12, 12 July 2005 (UTC)

• In fact programming languages are not "language"s. The term language has several senses, as you will find in any dictionary. And the computing sense is quite different to the linguistics sense. If you feel otherwise, then please translate at your own choice either your comment or mine into whichever computer programming language you prefer. Alternatively, perhaps a sound file of two computers communicating with each other in one of these "language"s would also be very interesting. — Hippietrail 01:25, 14 July 2005 (UTC)

I too would like to know where programming language terms "belong." Is there an en.wikiprogrammingdictionary.org? Or are programming terms, operators, functions, common library calls, platform-specific system calls, keywords, html markup, etc. intended to be lumped together somewhere in Wikisource? Or is this simply information that people feel should be repressed and/or inaccessible? --Connel MacKenzie 21:50, 12 July 2005 (UTC)

Well, there is a proposal for a Wikicode project, but as far as I can discern, that pertains to source code for scripts and programs and the like. It could possibly be extended to defining what all the gobbledegook is, I should imagine. --Wytukaze 21:57, 12 July 2005 (UTC)

• Delete: Computer languages are another form of conlang. Do we really want to open that can of worms. Eclecticology 18:27, July 12, 2005 (UTC)

I'm very carefully not opening that can of worms. We bring in those few terms that happen to originate as programming language tokens but are also used in plain English. We leave the rest out. This is all covered by the existing formulation of CFI. Note the definition given, which is supported by usage in referreed journals. It's not "the C token goto" or the BASIC token, or whatever. It's "any construct leading to an unstructured jump" (or whatever I actually said). The example is meant to reflect this: setjmp/longjum is "essentially a goto", meaning there are several flavors. By contrast, I don't believe that cdadr has similarly become generic (but car and cdr probably have). -dmh 21:21, 12 July 2005 (UTC)

I agree with what dmh has said above. You even see those without any sort of programming background using certain constructions, though I'm not convinced goto has achieved that sort of ubiquity. Of course, that does not mean it has use with those who do have a programming background. Printf, however...

P.S. What on earth do we do with the practice of correcting typoes on IRC (etc.) with s/badword/goodword? --Wytukaze 21:57, 12 July 2005 (UTC)

I don't think total ubiquity is necessary. We should certainly include halyard and swage, even if only sailors and smiths use them. In this case, goto and printf pretty clearly have generic (i.e., non-keyword) senses in the computing community.

As for s/badword/goodword/, it's not a usage of badword to convey meaning, and it would tend to invalidate the usage of badword which is being replaced. Put another way, do the substitution first and then consider the result. -dmh 17:03, 13 July 2005 (UTC)

Indeed, I wasn't saying that total ubiquity was necessary at all; rather, that not everyone uses it yet. Just a note, if you will.

And on the matter of s//, I was actually referring to the method of substitution, rather than the substitution itself. I'm not sure whether it merits an entry, but it is another construction of considerable use and usefulness. --Wytukaze 18:07, 14 July 2005 (UTC)

• This is a borderline item, and I'm willing to give it the benefit of the doubt. It is clearly better known than most computer programming instructions. Should it be considered as "Translingual" rather than English? Why does it need to be in all upper case? In an earlier time computers didn't have lower case letters at all so that then the "GOTO" format was there because there was no choice. Is all capitals now necessarily the only acceptable format? Eclecticology 17:04:54, 2005-09-06 (UTC)

When I see a word in all caps, I immediately think it's an acronym or initialism and try to think of what it stands for; therefore, I would request this be listed (if at all) in lowercase). If listed as "English" please note that this is in computer language - I would hate a non-native speaker to think that it's okay to say "I will goto the store". Cheers, --Stranger 19:17, 6 September 2005 (UTC)

I disagree that 'goto' is in computer language. It's a technical term used in English (and possibly other languages). Come to think of it jump has a similar sense (The code was so full of jumps there was no way you could possibly debug it, in which "goto" could substitute for "jump"). We should be clear that goto is not the same as go to, but this is no different from noting that pile up is not the same as pileup, though in both cases they are related — computing doesn't really figure in.

For me what seals the deal is uses in computing articles of the form "Construct X is tantamount to a goto." where goto clearly refers to the construct in general, not its manifestation in any particular programming language. However, I will not now nor probably ever provide a single citation to support this, and if no one else does, the term must be deleted for lack of documentation. -dmh 20:19, 6 September 2005 (UTC)

---