Why was wikipedia, and the public at large, better served by deleting these entries?
Anything I am not 95% sure about I add a notability tag to or simply leave alone. There are a number of articles that I did in fact find interesting citations for while trying to decide notability. Napier88 is a good example of such a language -- the papers on it received hundreds of citations, yet I had never heard of the language. I've been working on an entire rewrite of Napier88 in my spare time, which has understandably become even more "spare" recently :)
http://en.wikipedia.org/wiki/Categorical_list_of_programming...
What's useless about that?
http://en.wikipedia.org/wiki/List_of_object-oriented_program...
I don't see many pet projects here...
OK, so some of them might not even pass my standard of notability (which would be almost embarrassingly lax).
But it strikes me that you're using notability to solve a user interface issue -- that you find it difficult to navigate the programming language section.
More generally, it strikes me as horribly inefficient to rely on what one might call "outrage-driven notability" to correct for what were, you may now realise, erroneous selections for deletion.
Maybe I'm a dolt for blaming you; you're acting rationally within a broken system.
Surely a simpler solution to that is to make a list that you feel describes the languages you consider important. Maybe "List of programming languages notable in academia" (which would include things like FeatherweightJava but probably not regular Java). People who want to find languages that meet that criteria can find them all in one place.
Meanwhile, people like me who are interested in new languages, or languages that experiment with syntax but don't push the academic envelope, don't have that knowledge outright taken away from them by you.