Dropbox Announces Their Own Open-Source Python

[abral]

An intern at google tried that and it's called unladen swallow [1].

it seems alot more promising to look into the openjdk world. Some people are working on truffle a specializing ast implementation.
There is an experimental branch that implements this for python3 ([2]).

Actually LLVM is a pretty bad JIT. Some people tried to make an OpenJDK backend that uses LLVM to make a plattform independent JIT called Shark. It was very slow ([3]).


[1]: https://code.google.com/p/unladen-swallow/
[2]: https://bitbucket.org/ssllab/zippy
[3]: http://icedtea.classpath.org/wiki/ZeroSharkFaq

They've also tried to use LLVM for a JavaScript JIT and it was pretty bad.

[Pajn]

Keep telling yourself that. Maybe one day it'll be true.

Have you ever used Python/Django and Node.JS?
I have. What I chooses Django for I would never ever even think about Node for
and the same is true in reverse.

[quasipedia]

Once again, a team picks up a fancy and trendy language without giving consideration to performance, only to realize later that they need C++ kind of performance. But guess what, a better compiler won't be much helpful if your code is highly dependent on dynamic objects and garbage accumulation (collection as many prefer to call it).

Once again, a team picks up a language that allows them to deliver quickly and well, and being able to do so they outperform their competition and establish them as the market leader, to the point they actually can dedicate time and money to contribute back to the language community with a new interpreter... I think this story is pretty similar to that of that other little site you may have come across time to time... what was called? Ah, yes.... Facebook!

[sarmad]

Once again, a team picks up a fancy and trendy language without giving consideration to performance, only to realize later that they need C++ kind of performance. But guess what, a better compiler won't be much helpful if your code is highly dependent on dynamic objects and garbage accumulation (collection as many prefer to call it).

[thebear]

Keep telling yourself that. Maybe one day it'll be true.

Not everything revolves around the web you know. Python is actually used in Scientific Computing quite extensively
and it's gaining momentum in this area. Python in it's own is not quite as useful as
what you can easily do with Python/Cython/C++/C/Fortran etc (see e.g. xdress for how easy it can be to interface C++).
The use cycle is usually: 1. Prototype in Python 2. Profile to identify hot-spots. 3. Move those parts (usually <5% of codebase) to ${FAVOURITE_COMPILED_LANGUAGE}.

[TheBlackCat]

Take a look at MIT's Julia Language. I know Python and Matlab are the current kings on this land, but Julia is very promissing and VERY fast.

I know about Julia, but whether it will ever replace Python is still an open question. It has a core set of very dedicated developers, but its community is very, very far from Python's at this point. That may change, but there is no guarantee it will change, especially if python gets a unified JIT system.

[quasipedia]

Please consider the option of NOT responding to trolls. Any kind of rational response will just add fuel to the fire. If you ignore them they will eventually get bored and go away.

To be honest, the OP to me seemed more genuinely clueless about what he was talking about, than trolling. But yeah... I was surprised too by the amount of replies he got!

Let me try too: C is dead, C is dead!!! Whoever uses C nowadays is stupid, PHP is sooooo much better!!!

[bison]

You clearly have no idea how python works if you think any of those languages you mentioned are competing with it - you're comparing apples to oranges. Ruby is the only language that is similar in usage/purpose to python, and it is notably less popular. Also, the jump the 3.0 wasn't a bad idea. While breaking the compatibility was a little annoying, python supplies scripts to convert your code. I had a roughly 2500 line code that I converted from 2.7 to 3.3 by hand and it only took me an hour to do.

People don't switch to python 3.x because if they didn't write their code for it, there's a good chance it's going to be a pain to convert. As long as people are actively using 2.7, it isn't going to die, and if it works for them and is still being maintained, I don't really see that as a problem.

Please consider the option of NOT responding to trolls. Any kind of rational response will just add fuel to the fire. If you ignore them they will eventually get bored and go away.

[dh04000]

You're the clueless one if you think Python has any chance int he next 5 years to remain relevant.

I am a research scientist, and python is used ubiquitously for writing various scientific tools and programs as well as for interface applications between various machines and computers. Considering we are still using machines and computers from the days of early Unix, DOS, and OS/X, then I see all of our python interface machines seeing use for the next 30 years easy.

EDIT: before anyone gets annoyed and yells at me for our use of older machines, keep in mind that a good fluorimeter or spectrometer bought 30 years ago is still very useful today, but the software to run it only works on OS's of that time. Also, a new one costs $100,000 to $250,000.

Moral of the story, the software writers pick up a language, make a peice of software and then move on. But us "USERS" of thier software actually use this long forgotten works for decades at a time as long as its still useful.

[skriticos]

Bullshit. Google's Go directly competes with python on the web server side. Javascript is directly competing with Python for server apps now too. Virtually every company I work with is switching away from Django to node.js and other options.

You're the clueless one if you think Python has any chance int he next 5 years to remain relevant.

I'm using both Python and Go and they do have vastly different strengths and design goals:

1. Python: It's has ridiculously strong data structures and is a godsend for anything data analysis. Performance is not so great, but it's excellent to create quick prototypes. The standard library is very well made and designed to get things done fast. Python is a capable scripting language.

2. Go: It's strength is in concurrency and speed. The data structures are a *lot* less flexible than Python being statically typed (ever tried to do calculations with mixed types like integer and float32/64.. prepare for a lot of type-conversions). It's shines if you want to build a web server quickly. Technically it's a systems programming language.

Now Django and node.js are frameworks. They are nice, but not a panacea to anything. Raw Python is a very useful tool on it's own and is not going anywhere. Nothing even remotely as elegant for analyzing data is anywhere on the horizon. JavaScript is nice because it targets the browser which has a lot of benefits (no installation required), but needs a lot of catching up when it comes to accurately crunching data or maintaining big projects.

So my point is, that I strongly disagree that Python is going anywhere or even getting significantly weaker. I don't see credible alternatives either.