"Here's a challenge, then, for the internet: instead of spamming my Wikipedia talk page (which I don't really care about), why don't you work on fixing WP's notability guideline for programming languages?"
Amen to that. Others have said it many times before and it is still true: Wikipedia's notability guidelines would benefit from being fixed. It's rather annoying to note that what this guy was doing was within the letter of the law but yet seemed so wrong. There is no way that the likes of Factor or Clean should be deleted from Wikipedia so the question becomes how can the guidelines be changed so that they don't allow it to happen.
Sadly the mostly widely mis-used rule :) IAR is for daily usage, no real application in this case (because unfortunately the wider community will probably agree with deletions)
Factor is an interesting case. I tried to improve that article at one point by adding references, but I was unable to find any. As far as I can tell, it's almost never actually mentioned in the programming language community: not mentioned in a single book, even books about Forth-like languages, has no articles about it (in workshops, conferences, or anything), and its website has been mentioned in passing in four papers according to Google Scholar (the papers provide no real info on it besides mentioning in passing that it exists). It appears to have some level of notability among PLs aficionados, but if you were trying to discern that by reading the PLs literature, you wouldn't know it.
Clean is quite easy to find references for, though, so not sure how someone could've argued about that one in good faith.
It is something of a problem for things that are "known" in an area but don't have many references, because I do think Wikipedia's insistence on third-party sources is a good idea: an article based entirely on a few sources written by the subject and associates of the subject isn't great, like an article about a band that only cites the band's website, MySpace, and a few mentions on people's blogs. But how do you distinguish that case from the well-known-but-has-few-good-references case, and then how do you write an article about the latter? Here's a music example that might be analogous to some of these programming languages, a record label that's well-known in the postindustrial sphere, but has no good sources as far as I can find: http://en.wikipedia.org/wiki/Old_Europa_Cafe. How should one improve the article, and what should the information in it be referenced to?
The long-term solutions here are to improve the literature: someone needs to write a more recent book or survey article on emerging programming languages, and someone else needs to write a book on 1980s/90s independent postindustrial record labels. Then Wikipedia can just cite those books. Until then, since Wikipedia works best when it's a summary of the existing literature, it runs into trouble when the existing literature itself is deficient.
The answer is: changing stated Wikipedia policy is a painful, politicized process. Editing articles is annoying, and deleting articles is annoying and political, but trying to change deletion policy is the exact kind of direct frontal attack that gets you maximum political resistance and minimum effective gain.
Amen to that. Others have said it many times before and it is still true: Wikipedia's notability guidelines would benefit from being fixed. It's rather annoying to note that what this guy was doing was within the letter of the law but yet seemed so wrong. There is no way that the likes of Factor or Clean should be deleted from Wikipedia so the question becomes how can the guidelines be changed so that they don't allow it to happen.