Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Pakhi – Bangla Programming Language

github.com

Post reply on HN
Print view View original

31–40 of 69 posts

Re: Show HN: Pakhi – Bangla Programming Language

#31

Post by tsegratis »

FlyingSnake wrote: post #14

I wonder if it’s possible for the programming languages to take advantage of the unique features in some languages, e.g. Agglutination in Hungarian and Turkish. Has any study being done to that effect? I love the concept, this project has the potential to bring a whole new demography to the fold of programming.

An interesting concept from Turkish is each word can directly give you all (or single) objects that do or can respond to that action

  yap -- make
  yapan -- the one who makes
  yapanlar -- all those who make
  yapılan -- then one being made
  yapılanlar -- all those being made
This would be like functions having a standard way to return lists of all objects they're related to

Or direct solving of some complex queries like 'Who can open this door?'

In Turkish that would be open_this_door_an, where '_an' is the standard way to get a list of all objects capable of doing it

Re: Show HN: Pakhi – Bangla Programming Language

#32

Post by mtalantikite »

tsegratis wrote: post #28
FlyingSnake wrote: post #14

I wonder if it’s possible for the programming languages to take advantage of the unique features in some languages, e.g. Agglutination in Hungarian and Turkish. Has any study being done to that effect? I love the concept, this project has the potential to bring a whole new demography to the fold of programming.

Since RPN is front page, let me note that Turkish can be viewed as an RPN language -- the verb comes at the end (although agglutination etc make it not quite so clear cut) The agglutination is very similar to properties on structs object.time.by.ability.duration Except that there is a fixed standard order, and every object (word) responds in the same way. -- Very similar to an inheritance hierarchy from a single obje…

Bangla also is a subject-object-verb language and I hadn’t thought of it like RPN before!

Re: Show HN: Pakhi – Bangla Programming Language

#34

Post by mtalantikite »

rayiner wrote: post #21

Cool project! A bit off topic, but I’m a bad Bangladeshi and never learned the alphabet. Any sources you recommend to learn? I’m not great with (human) languages to begin with, but the character set is a real hurdle.

I’m currently learning Bangla and there are a few basic apps for iOS that present the letter along with it’s name, pronunciation, and how it is to be written (as in how to piece together the strokes to write the letter). It’s definitely geared towards kids, but the one I’m using is called BorKnow.

Also I’m using a book by Hanne Ruth Thompson called Beginner’s Bengali that does a good job at introducing the script.

Re: Show HN: Pakhi – Bangla Programming Language

#35

Post by hvocode »

lvass wrote: post #19

I can barely understand anything in that page though I wish you success and offer this Edsger W. Dijkstra quote, in good faith, for reflection. "Projects promoting programming in 'natural language' are intrinsically doomed to fail."

Not sure how that’s relevant, given that it appears to be expressing the usual language constructs but not in English or via a Latin alphabet. A couple minutes with Google translate made that apparent.

Re: Show HN: Pakhi – Bangla Programming Language

#36

Post by vr46 »

PaulRobinson wrote: post #7

While I 100% applaud the effort - and I absolutely love the idea of languages in non-English languages - you should be aware that phonetically (and if you drop the 'h'), the name is a racist term in the UK at least.

Yes, non-white Englishman here cringed when I saw the name having been on the receiving end of the term many times.

Re: Show HN: Pakhi – Bangla Programming Language

#37

Post by ksaj »

PaulRobinson wrote: post #7

While I 100% applaud the effort - and I absolutely love the idea of languages in non-English languages - you should be aware that phonetically (and if you drop the 'h'), the name is a racist term in the UK at least.

I think it is we who decided it is a racist word. I was a guest instructor in a university class in Islamabad, and people there used that word quite casually. It surprised me how often I heard it, but also that it was never treated in a negative manner. An important differentiator is that they weren't using it as a label to call each other maliciously, but more to denote that something is from Pakistan.

Perhaps at least in this case, intent has everything to do with its interpretation. I still don't use the word because that nuance is too easily missed.

Probably both cultures (which apparently are directly related, in terms of overlapping histories) are aware of words like this, and not likely to confuse them for ill intent. A little like how Canadian Anglophone children make fun of the French word for 'seal', but Francophones who also use the English swear word that it sounds like clearly know when you're swearing, versus when you're talking about a seal.

Re: Show HN: Pakhi – Bangla Programming Language

#38

Post by gerdesj »

tsegratis wrote: post #31
FlyingSnake wrote: post #14

I wonder if it’s possible for the programming languages to take advantage of the unique features in some languages, e.g. Agglutination in Hungarian and Turkish. Has any study being done to that effect? I love the concept, this project has the potential to bring a whole new demography to the fold of programming.

An interesting concept from Turkish is each word can directly give you all (or single) objects that do or can respond to that action yap -- make yapan -- the one who makes yapanlar -- all those who make yapılan -- then one being made yapılanlar -- all those being made This would be like functions having a standard way to return lists of all objects they're related to Or direct solving of some complex queries like 'Wh…

  * yap: make
  * yapan: maker
  * yapanlar: makers
  * yapılan: made (I think, "then one being made" doesn't make sense to me)
  * yapılanlar: OK you got me there!  It may be another job for "made" or goods or a synonym eg: objets d'art.  Yes, I did just use three languages as my own in one sentence.
I'm not a linguist. I've managed to fail French and German and yet pass Latin (whatever the heck that is!) at O level (old UK exam at age 16.)

I don't think we have quite the same concept you are describing, in English as such but I think we might come close. You can generally take a verb and turn it into a closely related adjective, gerund and gerundive. Anyway, if there isn't a word available in English, we simply steal (sorry, borrow) someone else's - simples! I bet I routinely use a few Turkish words without even realising it. English is a proud mongrel - it was an amalgam from day 1.

(edit: I've forgotten how to do a list hereabouts - sorry, Ooh two spaces)

Re: Show HN: Pakhi – Bangla Programming Language

#40

Post by zokier »

UncleOxidant wrote: post #26

We English speakers are incredibly lucky that programming languages first emerged based on our native language. It's easy to imagine scenarios where the first programming languages were based on Mandarin or Bangla or Hindi keywords.

> It's easy to imagine scenarios where the first programming languages were based on Mandarin or Bangla or Hindi keywords.

I don't know. To me that seems pretty far fetched idea.

Russian, or German, maybe? But I don't really see how the computer revolution could have started in SE Asia without fairly major changes in history

Post reply on HN