http://ecologywithoutnature.blogspot.com/2011/01/ian-bogost-on-computers-languages.html

Two comments from my post that are relevant here:

(1) Propositional logic is neither English nor French. It's not a language. Neither is programming code. It's done IN language, but it ISN'T language, just like the words you're reading here are done IN pixels, but they AREN'T pixels. Dig?

(2) [commenter posted some code] No amount of scary looking stuff that I don't understand touches my argument. I can show you a bunch of recipes for lobster bisque that would pop your mind, dude. They are written in English. I see "map return." I see "option." I see "Nondet" which is presumably an abbreviation, perhaps involving some sense of "determine." And the OED, for which I work, includes some coding senses of words such as “run” that I also see here.

I don't know what "Thou still unravish'd bride of quietness" means. Do you? But I can translate it into French.

25 years ago, when I was in AP Computer in high school, we were allowed to answer questions on the AP exam in pseudo-code. It wasn't the compilable syntax they were after, it was logic flow independent of language.

Tim the pixel bit is particularly nice.

I don't really have an opinion on the humanities-curriculum question (I think languages are over-emphasized in humanities curricula, and that programming languages are over-emphasized in computer-science curricula). But the sharp distinction doesn't make that much sense to me.

Propositional logic tends to obscure it because of its hyper-formalism and the fact that people rarely encounter anything but elementary examples written in it. But if you look at any actual programming language used by people to express nontrivial computational ideas, they have idioms, trends, fossilized expressions, phrasings that are considered "correct but awkward", etc.; writing code that compiles and is correct according to the formal language definition is only the start of what it means to be "fluent" in a particular computer language.

I don't think that means they should count as languages for the purpose of humanities curricula, but I think the sharp distinction being drawn here between "languages", which express meaning, and "codes", which are purely dry formal specification, is a bit simple.

I don't entirely understand the point about translation. Are we talking about translating computer code into English? If so, I agree that that's not the same kind of translation; as Morton says, it's just taking an algorithm and giving it an English gloss. It's more like explaining how a thing works than translation. But what about translating a significant idiomatic C program into idiomatic Haskell--- not just mechanically translating the C program word-for-word so it's "C in Haskell", like a bad French-to-English translation that simply uses a dictionary to replace every word in sequence, but translating an idea that was initially written "natively" in C so that it appears to a Haskell user to be native Haskell? I'm not sure that's the same kind of translation as English to French, but it sure seems like translation of some variety, and runs into some of the same issues and tradeoffs.

I mostly avoid these debates, though, because they always (the linked article not excepted) seem to have a reactionary-humanist quality to them; the usual, "these silly computer scientists trying to compare their dry formalisms to our infinitely subtle, complex, deep ideas, languages, philosophies!".

Actually it is conceivable that you could have a language that is a computer that uses humans or other language using beings as the computational medium. I'm also not sure, and perhaps i am too bound in semiotics here, that all language isn't code, and that it should be possible in theory, though not with our current systems i think, to build a computer that can interpret foreign languages.

I met my foreign language requirement for my degree with Spanish, but mostly they accepted spanish because I also was demonstrably passable in ancient greek statistics, and several computer languages. Here in the social sciences, statistics used to be acceptable as a foreign language of sorts... heh, i think that rule has changed.

@Mark
These silly computer scientists trying to compare their dry formalisms to our infinitely subtle, complex, deep ideas, languages, philosophies.

Yeah, that's common, but I don't see it at all in this case. And there are clearly things that computer languages and natural languages have in common. But it's pretty silly to suggest that they are interchangeable.

@Jeremy
Actually it is conceivable...
Isn't that a separate matter though?

It is and it isn't separate. if the argument is proceeding from difference, then when we could collapse some of these distinctions in short order, and if we do that, then as best as i can tell the only argument we have is that humans with language currently have cultures and it is the link between the language-culture and the ways of knowing/thinking that is important in learning other languages. But if that is the case then a computer language like lisp or haskell as embodying a mode of thought might be something to consider, whereas a more generic language might not.

I've posted my thoughts HERE on this, rather than a laborious comment.

Computer "languages" are not languages at all, period. They are convenient systems of notation, fancy versions of the sequences of bits that make up raw machine code. They are not intended for communicating with other people; they have no tense besides the present imperative; they are capable of discussing nothing but numbers.

Somebody somewhere is being a dumbass.

I personally don't think that computer "languages" should qualify for language requirements, but I really like this comment by Mark N:

But what about translating a significant idiomatic C program into idiomatic Haskell--- not just mechanically translating the C program word-for-word so it's "C in Haskell", like a bad French-to-English translation that simply uses a dictionary to replace every word in sequence, but translating an idea that was initially written "natively" in C so that it appears to a Haskell user to be native Haskell?

This, to me, is something well worth investigating. It reminds me of the example Ian and Nick Montfort point to in Racing the Beam (forgive me if I misstate something, as my copy is in my office) of a programmer who uses $60 both as the assembler command to exit a subroutine as well as the first colour in a sprite in order to save memory.

There does indeed appear to be a certain amount of nativeness to programming languages in terms both of efficiency as well as elegance, and it would be a shame to throw that particular baby out with the language requirement bathwater. I don't myself believe that this makes programming "languages" quite the same as, say, German or French, but I do think it is worth investigating notions of nativeness in programming languages all the same.

"Subtle, complex, deep": no. Not at all. In fact the difference in these terms is between rigor (software code has to be pretty unambiguous) and sloppiness (I can say "end" to you and you can take it any which way, pretty much).

Non-computational languages are not necessarily subtle, complex or deep, even if they are very old. They could be sloppy, vague and redundant.

But the real issue is, as Dave Mark puts it, the difference between "logic flow" competence and what medievals call "grammar" (i.e not logic). You need both of course. But you do logic "in" language, just like you do military strategy. Still, a war is not a piece of language and neither is logic.