Talk:DWIM

I think the edits removing content about DWIM being a hypothetical language were justified. The idea is patently silly, and I can't find any references to support the idea that anybody uses DWIM that way. Bigpeteb 17:59, 20 January 2007 (UTC)
 * Agreed, useless material which just seems like a bad joke, deleted. --80.217.189.62 16:03, 4 October 2007 (UTC)

It's stated that its use is humorous, but I cannot find where's the humour in it... --euyyn (talk) 13:56, 24 January 2008 (UTC)


 * I looked awhile, but couldn't find a reference. This term was well-known by computer scientists 30 years ago. (Even those who knew nothing about LISP. That's the reason, by-the-way, I have a certain question whether Teitelman originated it.) It's humorous because programmers and users know that a computer program cannot be swayed by their emotional appeal, so saying "DWIM" is intentionally and pointlessly "talking to a brick wall".


 * Alpha Ralpha Boulevard (talk) 05:09, 19 May 2008 (UTC)


 * I get the joke. A partner and I wrote a compiler that accepted Pascal, simple PL-1, FORTRAN, and several other such languages, tracking and tagging language semantics through the parse tree and doing "reasonable" interpretations of programmer errors (tagging them in the listing). You could, if you wanted, write (or import), FORTRAN code, require Pascal declarations, and get smart bounds checking. It became a monster when we tried to add ICON, and died. We did call it DWIM, though. ICON is great, btw. I wonder, if I started with ICON, then added .... htom (talk) 03:59, 8 March 2011 (UTC)


 * I seem to recall DWIM (Do What I Mean) in a list of joke assembler mnemonics in the mid 1970s when I was at Cornell. Other entries included DWPI (Do What Programmer Intended) and the now-familiar HCF (Halt and Catch Fire). So I'm pretty sure that it had a humorous meaning, at least at one time, but I don't want to add it unless I can find a contemporary reference. Karn (talk) 08:14, 16 November 2017 (UTC)


 * See the Jargon file entry (reference 6) to understand the humor. When the term was coined, it was generally recognized that DWIM was sarcastic. The whole domain of software engineering demonstrated how hard it is to properly specify a problem. Compare to malicious compliance, in which a human follows literal instructions (usually bureaucratic) to demonstrate how they fail to DWIM for their manager. Also note the use of ChatGPT to write software, and how it can be thought of as a DWIM function. What we considered impossible 50 years ago is now routine. -- SpareSimian (talk) 04:14, 13 December 2022 (UTC)

Article changes
This article has various small problems:

1) The citation slightly contradicts what the article says (Teitlman did not necessarily coin the phrase, and did not claim that he did.)

2) The POV is a programmer's, but the term is common among general computer users. (And according to some sources, to the public at large, in non-computer-related situations.)

3) Some statements are unprovable, hyperbole, or just wrong. "With this the user can nearly always get the desired result..."

4) Replacing the words "acronym" with "initialism" and "typo" with "thinko" is confusing, and makes distinctions that are not explained. The Wiki article on the "thinko" link directs to a Wiki article that does not mention the word.

5) The article has much unfounded speculation. Saying that DWIM originated in the "early computer age" is either too vague (when was the "early computer age"?) or untrue (the Wiki article states modern computers came about in the 1940s). It would be more straightforward to note that the term has been around for decades.

6) EMACS is a poor example, because most readers will never have heard of it, and probably less than 1% used it enough to make a good example. (I've used it, and was unaware of -- or didn't remember -- the comment-dwim function.) Examples that many readers will be familiar with appear on the Typographical error and Office Assistant.

7) The link to Occam's Razor is confusing. That is a rule of thumb about scientific explanation, whereas DWIM has to do with human factors.

Alpha Ralpha Boulevard (talk) 22:32, 14 May 2008 (UTC)