Re: pygame and python 2.5

On Feb 11, 4:24 am, Steve Holden <s...@xxxxxxxxxxxxx> wrote:
mensana...@xxxxxxx wrote:
On Feb 11, 1:35?am, Steve Holden <s...@xxxxxxxxxxxxx> wrote:
[...]
After all, they have already given freely and generously, and if they choose
not to give more on top of that, it's really up to them.
Right. Get people to commit and then abandon them. Nice.
Anyone who committed to Python did so without being battered by a
multi-million dollar advertising campaign.

Multi-million dollar ad campaigns mean nothing to me.
I committed to Python because it's a great language.
I've dabbled in perl, Visual BASIC, UBASIC, REXX, Java,
Scheme, C and C++ but Python is the one I use.

Yes, but your decision must surely have been an informed one, and there
must surely be reasons why Python remains your choice.





The Python Software
Foundation has only recently dipped its toes in the advocacy waters,
with results that are still under evaluation. And the use of the
Microsoft "free" VC6 SDK was never a part of the "official" means of
producing Python or its extensions, it was a community-developed
solution to the lack of availability of a free VS-compatible compilation
system for extension modules.

I agree that there are frustrations involved with maintaining extension
modules on the Windows platform without having a copy of Visual Studio
(of the correct version) available. One of the reasons Python still uses
an outdated version of VS is to avoid forcing people to upgrade. Any
such decision will have fallout.

Such as anyone who tries to get in the game late.

I'm afraid it does seem to work out like that, yes.

An update is in the works for those
using more recent releases,

That's good news, although the responsible thing
to do was not relaease version 2.5 until such issues
are resolved.

Well that would be an issue for the release team. I'm not sure what
Anthony Baxter (the release manager) would have to say in response to
this point.

Possibly something like:

"I realize you're a Windows user, and a Windows user with
an AOL email address at that, so it may come as a shock
to learn that the computer industry doesn't start and
finish on Windows. I don't see why the needs of Windows
users like yourself should come ahead of the needs of
users on Mac OS, Linux, Solaris, etc." - Steven D'Arpano

I would hope that it would instead be that the needs of
all users are equal.


but that won't help users who don't have
access to Visual Studio.

That can be solved by throwing money at the problem.
But money doesn't help when the solution is on the
far side of the moon.

I see your problem, but I don't know what I can do to help you.

Well, that was the point of this, to get people to
see the problem.

There
were also, as I remember it, issues with the updated version of Visual
Studio being non-conformant with standards in some significant way, but
I never took part in the discussions on those issues.

Yes, it's
occasionally very frustrating to the rest of us, but that's life.
As the Kurds are well aware.
I really don't think you help your argument by trying to draw parallels
between the problems of compiler non-availability and those of a
population subject to random genocide.

You missed the point of the analogy.

Perhaps because it wasn't a very good one?





The US government suggested to the oppressed tribes
in Iraq that they should rise up and overthrow
Saddam Hussein at the end of the first Gulf War.
And what did the US government do when they rose up?
Nothing. They were left to twist in the wind.

Try to keep things in perspective, please.

See if you can see the similarity.

I buy into Python. I spend a lot of effort
developing a math library based on GMPY to use
in my research. I discover a bug in GMPY and
actually go to a lot of effort and solve it.
But _I_ can't even use it because I've been
left to twist in the wind by the fact that
Python 2.5 for Windows was built with an
obsolete compiler that's not even available.

Luckily, unlike the Kurds, my situation had
a happy ending, someone else compiled the fixed
GMPY source and made a 2.5 Windows version
available. But can anyone say what will happen
the next time?

Presumably not. I presume you have been reporting your bugs through the
Sourceforge project to keep the developers in touch with the issues you
have found?

Last time I tried, it didn't work and e-mail to the
maintainer didn't get any response.

Normally a package's maintainers will produce updated
installers,

Unless they have stopped doing Windows developement as
part of their job as is the case with GMPY. Luckily,
there's someone out there who does create Windows
binaries.

but this behaviour is unreliable and (no pun intended)
patchy sometimes.





The best I feel I can do is raise these things on occasion,
on the off-chance that I manage to catch the attention of
someone who is
altruistic, knowledgeable, and who has some spare time on
their hands!
Someone who, say, solved the memory leak in the GMPY
divm() function even though he had no way of compiling
the source code?
Just think of what such an altruistic, knowedgeable
person could do if he could use the current VC compiler
or some other legally available compiler.
Your efforts would probably be far better spent trying to build a
back-end for mingw or some similar system into Python's development
system, to allow Python for Windows to be built on a regular rather than
a one-off basis using a completely open source tool chain.

No, as I said elsewhere, I'm not a software developer,
I'm an amateur math researcher. My efforts are best spent
as an actual end user to find and report bugs that the
developers never see. Remember, a programmer, because he
wrote it, only _thinks_ he knows how the program works.
Whereas I, the user, _know_ how it works.

The fact that the current maintainers of the Windows side of Python
choose to use a commercial tool to help them isn't something I am going
to try and second-guess. To do so would be to belittle efforts I would
have no way of duplicating myself, and I have far too much respect for
those efforts to do so.

And I respect those efforts too. What I don't respect
is irresponsible behaviour.

There are published ways to build extension modules for Windows using
mingw, by the way - have you tried any of them?

Yeah, and got nowhere.

It's much harder than sniping on a newsgroup,

That figures. You try and contribute and you get
accused of being a troll.

I wasn't accusing you of being a troll, rather bemoaning your (in my
opinion) less-than-constructive tone.

The squeaky wheel gets the grease.

The points you raise are
important, and I do feel that there ought to be easier solutions for
people in your position.

That's all I'm asking for is for others to appreciate
the situation.


but you earn rather more kudos.

Guess what kudos I got for solving the GMPY divm()
problem? None. How much effort would it have been
to mention my contribution in the source code
comments (as was the case for other contributers)?
Not that I'm bitter, after all, I'm altruistic.

I'm sure if you've made a contribution to the code you only have to ask
for your name to be added as a contributor to be mentioned in the source.

That wasn't important, I'm not that petty.
It was simply a real-world example.


By the way, on the sci.math newsgroup I promote
Python every chance I get. One fellow thanked me
profusely for recommending Python & GMPY and asked
for some help with a program he was having problems
with. We worked it out fine but his problem made me
suspect there may be more bugs in GMPY. What's my
motivation for tracking them down?

The satisfaction of a job well done? What's my motivation for acting as
a director of the Python Software Foundation when I get accusations of
irresponsibility?

I apologize. But I hope you see how this appears from
the outside, that the PSF doesn't give a rat's ass about
Windows users with AOL addresses. Sure, that's wrong,
but calling people who bring up these points whiny leeches
doesn't do anything to dispell that notion.

Anyway, thanks for taking the time to help maintain gmpy.

Thanks, I try to help as much as I can. I'm a little
sensitive about gmpy because without it, I would have
to abandon Python and I don't want to abandon Python.


This thread is starting to make me think that there's a case to be made
for somehow providing supported build facilities for third-party
extension modules.

And the untouchables would greatly appreciate it.


This wouldn't be a simple project, but since there's a Windows buildbot
for Python there's no reason why the same couldn't be done for
extensions. I'll raise this with the PSF and see what the response is:
then your carping will at least have had some positive effect ;-)

Stick with it, and let's try to make things better.

Ok.


regards
Steve
--
Steve Holden +44 150 684 7255 +1 800 494 3119
Holden Web LLC/Ltd http://www.holdenweb.com
Skype: holdenweb http://del.icio.us/steve.holden
Blog of Note: http://holdenweb.blogspot.com
See you at PyCon? http://us.pycon.org/TX2007


.

Relevant Pages

  • Re: - E02 - Support for MinGW Open Source Compiler
    ... No Python abilities are required, either, for that matter. ... And the fact that building *any* Windows native program without commercial ... compiler suites installed *somewhere*. ... Also, vcproj2make has a dependency on PyXML, which isn't necessarily obvious ...
    (comp.lang.python)
  • Re: pygame and python 2.5
    ... I committed to Python because it's a great language. ... modules on the Windows platform without having a copy of Visual Studio ... developing a math library based on GMPY to use ... obsolete compiler that's not even available. ...
    (comp.lang.python)
  • Re: about presicion
    ... > Windows between any two releases of Python and will presumably ... What I meant was the OP can't compare Decimal and gmpy unless he has ... > file) compiled for the version of Python you're using. ... Would the command line c compiler from the MS SDK suffice or do you ...
    (comp.lang.python)
  • Re: pygame and python 2.5
    ... I realize you're a Windows user, and a Windows user with an AOL email ... developing a math library based on GMPY to use ... obsolete compiler that's not even available. ...
    (comp.lang.python)
  • Re: pygame and python 2.5
    ... from source code to Windows? ... version of Python, due to Python limitations. ... - Recompiling such an extension requires you to have a C compiler set ...
    (comp.lang.python)