is there / would there ever be a jit-compiler for sugar?

Points of views and questions about the OLPC project. What should it be? How could it be better? Where is it needed most? Ask your questions here and let your opinion be known.

is there / would there ever be a jit-compiler for sugar?

Postby manymny » Fri Sep 18, 2009 5:21 am

perhaps there is a better forum for asking this question, but i specifically thought ektoric might have some insight about it, or at least good speculation, either way i thought it might be interesting outside pm.

a jit compiler is a tool that makes interpreted languages (like python, which sugar runs on) and javascript run considerably faster. the ff 3.5.x browser is much faster than 3.0 because so many websites use javascript (and use more of it now than ever) and its jit compiler (which i think is completely new to 3.5) at least doubles the speed of javascript.

python is an interpreted language- i have nothing against that, and i think there are jit compilers for making python run faster. do you think sugar will ever use one? if it doesn't already.

it seems there are two ways of this happening, i'm not entirely sure both are possible. the ideal and more likely way is that it would become a native feature of sugar: jit python would become the rule. (despite the word "compiler," all code would still be interpreted, you could still change the source as easily as you do now) the other way (if it's possible) would be if someone made a sort of jit "plugin" so that when integrated with sugar, it would perform faster. that'd be cool too- if an extra step at least.
teachers that continue teaching should be expected to continue learning; it's an absurdity to have 19th century expectations of 21st century educators. granted they shouldn't have their time wasted (more than students,) or be paid insulting salary.
Posts: 26
Joined: Mon Aug 17, 2009 11:26 am

Re: is there / would there ever be a jit-compiler for sugar?

Postby mburns » Thu Sep 09, 2010 12:58 am

Google sponsors the Unladen Swallow project which is an attempt to optimize Python with, among other things, a JIT compiler. Once it reaches maturity, expect it to become widely adopted.
Support Team
Posts: 105
Joined: Wed Dec 12, 2007 3:09 pm
Location: Corvallis, Or

Return to Open Discussion

Who is online

Users browsing this forum: Google [Bot] and 1 guest