Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×

Comment Re:There is something called multi-process you kno (Score 1) 196

Of course I know about multiprocessing. Why have one copy of the interpreter and libraries loaded when you can have N, plus its so much more efficient to marshal data across process boundaries than to access a global shared memory block.

I've heard this processes are so much better because we can't do threads for so long. Kind of like if I cut off my right arm, its so much better to only have a left arm because you only need to move 5 fingers instead of 10.

Comment Re:And it still has the GIL (Score 1) 196

2. The only reason it's hard to fix is because certain parts of Python are overly dynamic. Since they broke backwards compatibility in Python 3 it would have been the perfect time to fix it. Instead they broke backwards compatibility for stuff 99% of the community doesn't give a fuck about and now nobody is upgrading even though Python 3 has been out for over 5 years.

That is really insightful, seriously. Python 3 did break backwards computability, this really would have been the time to fix some original design flaws, but they didn't, instead, they focused on stuff, like you said 99% of the people out there don't care about, hence why so many use 2.7 today and how many new projects are even started with 2.7.

There's nothing wrong with design flaws, we all make them, you just at some point have to go back and realize you made a mistake and fix it.

Comment Re:And it still has the GIL (Score 1) 196

And therein likes the problem: python is a "incredibly dynamic language" which makes any sort of performance difficult if not impossible. The problem is Python is so dynamic that its impossible to perform any sort of meaningful validation before the code is actually run.

Comment Re:And it still has the GIL (Score 1) 196

I never said it was easy removing the GIL, nor do I know how to do it and meet all of Guido's requirements.

The GIL is a design flaw of the language. If Python remained just a way to add quick scripting to existing programs, just like TCL, I would have no problem with its design. But I do have problems with Python becoming a systems language. Its far far far too dynamic for its own good, it should not encourage dynamically replacing bits of the runtime at runtime. The GIL really shows the age and intent of Python.

These sort of ultra dynamic language may be good at writing quick and dirty scripts, but such dynamic features make maintaing and understanding any large system a nightmare. After all, bugs are so much more fun to find months after you've released an app that right away that a static analyzer could have found.

Comment Re:And it still has the GIL (Score 0) 196

I could not agree more. Its virtually impossible to perform any sort of type inference in Python, hence there are no viable JITs. Basically the whole design of Python was so any part of the runtime can be overwritten at runtime, i.e. monkey patching.
I think the big problem with Python is all the hacker types who think it so cool to swap out bits bits of the runtime at runtime just because you can. Now this leads to some truly incomprehensible and unmaintainable code.
Dynamic typing is OK, at least its done correctly in JavaScript so one can actually perform type inference and JIT compile it.
I really wish some other languages like Scala would gain more traction.

Comment And it still has the GIL (Score 1, Insightful) 196

Yup, Global Interpreter Lock so Python is still fundamentally single threaded -- only a single thread can be executing any python code at any given instance.

Its 2014 and we still can't have a multi-threaded python, this is ridiculous.

If you read Guido's criteria for getting rid of the GIL, he lists so many things that are specific to the current single threaded system (which is evidently perfect) that the only solution that meets his criteria is the current system.

I guess the only solution is to either live with single threaded system or fork it.

Comment Lets define our own string, vector, list classes! (Score 1) 73

General rule of thumb is when a library defines their own types like string, vector, hash map, list, etc, ... run, don't walk away from it.

Seriously, WTF is wrong with just plain old STL???

Lets implement our own string class so we can be completely incompatible with everything else.

It seems like every first year CS student writes their own string and list classes (I know I did when I started).

Comment What "Windows Exodus" (Score 0) 628

There is no such thing as a Windows Exodus, just a bunch of windows users bitching that OMG, OMG, SOMETHING IS DIFFERENT!!! Like with every Windows version, there will lots of bitching and complaining, then eventually, they will be good little Windows users and simply accept whatever Microsoft tells them to use.

Comment Re:And you wonder why desktop Linux is a failure? (Score 1) 202

1. they dont' want input from users. they don't want users tweaking its innards. users are expected to update their workflows and expectations to the 'one true path.'

you know this because?

2. sure there is. if gnome 3 is going the flat-learning-curve/flat-power-curve route...

I agree, Gnome 3 with the defaults sucks, WTF could they not have made tweak tool part of the control panel??? or at least default install. But just grab tweak tool and some extensions and Gnome3 is nice.

3. no they couldn't, well, not as easily. gnome2 is say 90% of what modern users want. it's easier to add the 10% and get it working well, than rewriting half of gnome3 and resyncing their changes with every gnome release.

No need to 're-write' it, just use the components, and tie them together differently, basic simple javascript.

4. talk to the gnome3 devs.. their demagoguery is the problem. their slavish apple chasing attitude is another.

Again, you know this because?

Comment And you wonder why desktop Linux is a failure? (Score 0) 202

A thousand developers all going a thousand different directions.

? One of Gnome3 biggest failures is they provide very little documentation on how to customize / modify it. Gnome3 is a actually a dammed good design and provides for immense customizability. Want it to behave like Gnome2, all you have to do is write a bit of JavaScript and glue the bits together is a Gnome2 style, thats it. I truly have no idea why the Gnome3 developers want to hide all the great work they have done. Its so easy to write themes / extension, but why do we have to install these fucking 3rd party tools like Gnome Tweak Tool, WTF is it not built into Gnome3???

There is no fucking need to waste everyones time extending Gnome2.

If these MATE clowns would have just taken a look at Gnome3, they could have made it work exactly like Gnome2 without introducing all this insane complexity of maintaining another dammed desktop.

I'm sorry, but these MATE clowns really piss me off, they could have worked with the Gnome3 developers to fix the problems with Gnome3, but instead, they go off their own way, and create duplicate dead effort.

So, how many God dammed desktops does Linux have now? do we really need so many? The look / feel of just about any of these desktops can be achieved with a custom shell built on Gnome3, much like Mint. So what does this MATE shit do for developers like me, now they expect me to support GTK3 and GTK2?? Why can't these desktop developers learn to get along or at least take a look at Gnome3 and see what you can build on it.

Sorry for this rant, but this complete and total inability to get along, work with, or at least look at what other developers are doing is so fucking ridiculous. We don't need another desktop, we just need to fix the ones we have.

Slashdot Top Deals

Lots of folks confuse bad management with destiny. -- Frank Hubbard

Working...