Is Sanskrit the most suitable language for computer software?

True, but in that case compatibility with old code is one of the constraints on the problem. It’s just very rarely explicitly called out because you’re very rarely looking at switching languages.

Wow, I didn’t know that’s where Panini’s brilliance lay… I was sure his brilliance was in first grilling the sandwich on both sides at once.

His brilliance was truly without end, a renaissance man who created both modern linguistics and modern café cuisine. Widely considered history’s greatest linguist, and the discoverer of principles that underlie almost all of modern linguistics, he also first united pesto, cheese, and deli meats in a form previously not even understood.

Does anyone remember Loglan
If you’re looking for a spoken language, this would be among the best candidates.

However, it will be easier to teach humans to use a crisp subset of their native language for future Programming than to teach them a new language.
So unless the idea is that all programming is to be done by Sanskrit experts, count on there someday being many Human to Computer language subsets and enough raw processing power to try out dozens of variants in a short time to see which is the program action the user was hoping for.

Actually, this is incorrect. Many computer languages use constructs that range from being mildly inconvenient to complete pain-in-the-butt for programmers but are intended to make the language easier for the compiler/interpreter to parse. (I used to have a canonical example from Common Lisp, but it’s been many years since I’ve programmed in it, so I’ve forgotten what it was.)

But certainly, syntactic sugar such as “begin-end” blocks are more for the compiler’s benefit than the programmer’s, although they do have some use for increasing the readability of the code.

I know nothing about programming languages or Sanskrit. In fact, with all due respect, I don’t care to learn. What I do know is that India in recent years has produced a buttload of good programmers. That right there is enough to put the Sanskrit/programming idea in the same basket with the old accepted wisdom that Italian is the best language for music – which originated at a time when Italy was producing a buttload of good musicians.

What are they famous or something?

Umm, India has produced a boatload of good programmers in last few years, probably has more to do with having 1 Billion People and a much higher rate of being multilingual with English as a common second language. (keeping in mind English is the most in demand language for programming for now)

• 2005 est. 1,080,264,388 (2nd)
Official languages Hindi, English, and 21 other languages
see http://en.wikipedia.org/wiki/India

Indeed. The Russians did this with COBOL and Pascal at least, and probably other programming languages — that is, kept the grammar and semantics of the original language, while translating the keywords into Russian. Makes perfect sense to do that. (I’m sure the Russians had their own home-grown programming languages as well.)

      • Yea, it would be improved if they had a way to force it to keep character strings in the executable, so you could comment it, and it would just automatically insert jumps over all the comment characters.
  • Well now, this depends on what your objectives are. Do you mean a better programming language would be easier for people to remember? Or that a better language would be easier to troubleshoot? The problem with compiled code is that if you only have the machine code, you have no way of knowing what the source code was. Seems to me the first matter of business is to switch to a programming language that allows accurate two-way translation. Does any other language allow that?
    ~

Three issues here:

  1. Executable size. Nowadays it matters a lot less (unless you’re expecting folks to download your stuff over dialup), but back in the day when every byte mattered, this just wasn’t feasible.

  2. Run time. There’s going to be some overhead involved in passing over all of those character strings. Back then, it would’ve been a nightmare. Nowadays it would interfere badly with a pipelined architecture.

  3. Security. Want to make it easy for hackers and script kiddies to reverse engineer your code and find out exactly how it works? This is the way.

See my point about securtiy above. At least the most recent version of MS Visual Studio will generate assembly and machine code listings along with the source that you could use to map some things back (anonymous crash data, for instance). That’s a good solution because it leaves the developers with what they need to do their troubleshooting without making it easy for malicious programmers to discover the behavior of their target.

Right. Italian is the “language of music” because Italians established an early reputation in composing and describing music (as well as making musical instruments) over the course of many decades. Subsequent achievements by, say, the German composers happened against this historical background. The Italian vocabulary and notation systems already in place were then carried forward, even long after Italian contributions to music became much less significant. That’s because there wasn’t any need to reinvent the wheel — as people have been saying since, well, since the day after the wheel was invented.

English has so far been the dominant language of computer science and programming languages because most of the work in the field happened first in the UK and the US, and a large fraction of it continues there even now. There’s a huge amount of inertia behind this choice. I don’t see that changing, even after the US and UK stop being prolific contributors to the field (if or whenever that might happen), for the same reason that we’re still using Italian to talk about music — and French to talk about food and diplomacy, and German to talk about philosophy, and Latin to talk about the law, and so on.

Since Sanskrit offers no special attractions for designing a new computer language, the only way for it to become popular, let alone dominant, is for a lot of Indians who know Sanskrit to start designing fantastic new languages that really catch on. But any language that’s fantastic in Sanskrit would be pretty much just as fantastic in English, since the underlying mathematics are not tied to natural languages anyway.

II Gyan II, you will no doubt find Syntax interesting as it has a lot of Topology inherent.

What ever happened to the proposed study to teach a bunch of children exclusively Loglan. I suppose the problem was to teach their parents Loglan first. :stuck_out_tongue: :stuck_out_tongue: :stuck_out_tongue:

Is Sanskrit a living language?

It seems to be analogous to Latin: dead, but not entirely forgotten. But I’m not Indian, or Hindu, nor am I a linguist.

(And I’m not going to make a Doper acronym out of that disclaimer either.)

Sounds fairly evil, was this for real?

Supposedly. A few years back, the TeeVee showed a couple of villages in Bangalore that conversed in Sanskrit. There were also a few newspaper articles. Of course, such reporting is gospel, so the ‘supposedly’ at the beginning is blasphemy.

Either way, it doesn’t sound like your average Indian programmer would be fluent in Sanskrit. Thanks.

Aside from the obvious problem that they wouldn’t be able to communicate with anyone except loglan-speakers, I think this would end up being a massive handicap in the future when they try to learn other languages that aren’t so logical, since they won’t have experience with all the exceptions-to-the-rule that are par for the course with natural languages. Why not teach them the most difficult language possible, then say, “everything else is easier”? :slight_smile: