[Python-Dev] Is Python moving too fast? (was Re: Is python commercializationazing? ...) (original) (raw)

Tim Peters tim_one@email.msn.com
Sun, 27 Aug 2000 07:13:35 -0400

[Skip Montanaro]

I began using Python in early 1994, probably around version 1.0.1.

And it's always good to hear a newcomer's perspective . Seriously, it was a wonderful sane sketch of what's been happening lately. Some glosses:

... From my perspective as a friendly outsider, ...

Nobody fall for that ingratiating ploy: Skip's a Python Developer at SourceForge too. And glad to have him there!

... 3. Python is now housed in a company formed to foster open source software development. I won't pretend I understand all the implications of that move ... but there is bound to be some desire by BeOpen to put their stamp on the language.

There is more desire on BeOpen's part-- at least at first --to just pay our salaries. Many people have asked for language or library changes or enhancements in the past based on demanding real-world needs, but until very recently the only possible response was "huh -- send in a patch, and maybe Guido will check it in". Despite the high approachability of Python's implementation, often that's just too much of a task for the people seeking solutions. But if they want it enough to pay for it, or aren't even sure exactly what they need, they can hire us to do it now (shameless plug: mailto:pythonlabs-info@beopen.com). I doubt there's any team better qualified, and while I've been a paid prostitute my whole career, you can still trust Guido to keep us honest . For example, that's how Python's Unicode features got developed (although at CNRI).

I believe that there are key changes to the language that would not have made it into 2.0 had the license wrangling between CNRI and BeOpen not dragged out as long as it did.

Absolutely. You may have missed some of the endless posts on this topic: we were going to release 2.0b1 on July 1st. I was at Guido's house late the night before, everything was cooking, and we were mere hours away from uploading the 2.0b1 tarball for release. Then CNRI pulled the plug in an email, and we've been trying to get it back into the outlet ever since. When it became clear that things weren't going to settle at once, and that we needed to produce a 1.6 release too with only the stuff developed under CNRI's tenure, that left us twiddling our thumbs. There were a pile of cool (but, as you said later, old!) ideas Guido wanted to get in anyway, so he opened the door. Had things turned out as we hoped, they would have gone into 2.1 instead, and that's all there was to that.

... All this adds up to a system that is due for some significant change.

Sure does. But it's working great so far, so don't jinx it by questioning anything .

... Once 2.0 is out, I don't expect this (relatively) furious pace to continue.

I suspect it will continue-- maybe even accelerate --but shift. We're fast running out of any feasible (before P3K) "core language" idea that Guido has ever had a liking for, so I expect the core language changes to slow waaaaay down again. The libraries may be a different story, though. For example, there are lots of GUIs out there, and Tk isn't everyone's favorite yet remains especially favored in the std distribution; Python is being used in new areas where it's currently harder to use than it should be (e.g., deeply embedded systems); some of the web-related modules could certainly stand a major boost in consistency, functionality and ease-of-use; and fill in the blank _________. There are infrastructure issues too, like what to do on top of Distutils to make it at least as useful as CPAN. Etc etc etc ... there's a ton of stuff to be done beyond fiddling with the language per se. I won't be happy until there's a Python in every toaster .

although-perhaps-light-bulbs-don't-really-need-it-ly y'rs - tim