Python MSVC++ binaries considered evil

[Cliff Wells]

On Sat, 26 Jan 2002 00:13:12 GMT
Mark Hammond wrote:

> Cliff Wells wrote:
> > What I was trying to say is that for a Windows user to actually do the
> > compilation with a free compiler, it is nowhere near as simple as it is
for
> > say Linux, where gcc is the standard.  Obviously Python itself is very
> > portable.  I also won't disagree that VC++ may in fact be the best
Windows
> > compiler (probably for reasons best left to the antitrust courts).
> 
> You are still missing Tim's point.  Most professional Windows developers 
> tend to have MSVC.  This means that for most professional Windows 
> developers it is in many ways *simpler* than Linux - no configure or 
> install.

I don't disagree.  In fact I _have_ VC++ (not that I use it much).  I'm not
arguing that it isn't very simple _if_ you have VC++.  But, if you have
Python on Windows, you will need to shell out at least $100 if you ever
want to compile C modules for Python, or Python itself for that matter. 
Personally, I would dump VC in 0.1 seconds if the Python source included
Makefiles for FreeBCC (or some other free compiler).  What I am saying is
that for people without VC it is not so simple.  If there were no free
compilers available for Win32 this would be moot (and not so long ago this
was the case, more or less), but it isn't the case today.  Besides, I
wasn't aware that Python was targeted exclusively at professional
programmers.

> Python is that way on Windows because people found it not working for 
> them, and contributed build changes back until it did.  This exact thing 
> has started happening with the free compilers now too, which is
excellent.

Yes, hopefully this entire discussion will become irrelevant.

> However, asking for it to happen wont make it happen.  You asked for it 
> to happen, and Tim told you why the main Python guys are not making it 
> happen.

Again, I didn't ask.  If I say "wow, nice car" it doesn't mean I asked you
to buy it for me ;-)

> ...
> > Since I was replying to your post, maybe you misinterpreted my
intention,
> > but I wasn't suggesting that PythonLabs should be responsible for it. 
This
> > list _is_ "the community" so I was discussing it here.  
> 
> Discussions don't make it happen either ;-)  The precedents have been 
> set, and basic support is already in the code base - it just needs work.

I only halfway agree with you.  Discussions can be the catalyst for
something to happen.  If the discussion in this thread hadn't occurred, I
can pretty much guarantee that I wouldn't be downloading FreeBCC this
weekend to give it a try - it simply wasn't on my mind.

> > This is of course the classic problem with the Windows platform:
everyone
> > uses (or believes everyone else does) Microsoft tools so everyone who
> > doesn't want to drop $$$ for them is left out (or at least left looking
for
> > pre-built binaries).  
> 
> Nor complaining...

You are correct, and I will stop complaining about MS immediately. 
Actually... that's a lie. ;-)

> > I guess we can hold our breath for the day when MS
> > includes a basic C/C++ compiler and libraries with the OS.  
> 
> Or, you can hold your breath waiting for the day for someone in the 
> Python community to make it happen.  Why not you?

I said I was going to look at it in my last post... can I have a day or
two?  Besides, that particular remark was aimed at MS, not the Python
community.  I have an unconscious tendency to take jabs at MS at random
intervals, if you haven't noticed.

> Microsoft today have the best Windows compiler. It was not that many 
> years ago when it had the *only* 32-bit Windows compiler.  I believe 
> they have a right to charge money for such a tool, while applauding any 
> (free or otherwise) competitors that offer a reasonable choice.

Hm. I thought Intel had the first 32-bit compiler, but I'm probably wrong. 
Regardless, my main point is that in order for people to contribute to
open-source Python on Win32 there is a cost of entry.  This seems likely to
reduce the number of potential contributors (besides the other reasons
discussed earlier).  I know that if I didn't need VC++ at work I wouldn't
have it at all, as much for political reasons as cost.  This affects not
only the Python core, but probably to a greater extent extension writers
who hope to gather support for their projects - they post their code on SF
or somewhere and Windows users can't help because they know they can't
compile it without VC++.

A common sentiment among open-source developers is if someone wants a
feature added or a bug fixed, the best thing is to try to do it yourself. 
On Linux (or any platform where gcc actually works) that's entirely
reasonable, but expecting people to shell out $$$ to Microsoft in order to
be able to help on Windows isn't (IMVHO).

>  > Still, I think I'll look at it this weekend.
> 
> Excellent - that's the spirit! :-)

I should have put that at the _top_ of my post and maybe I could have
avoided your earlier suggestions that I'm too lame do it myself. Sheesh!
:-P

BTW, thanks for your excellent contributions to Python.

Regards,

-- 
Cliff Wells
Software Engineer
Logiplex Corporation (www.logiplex.net)
(503) 978-6726 x308
(800) 735-0555 x308

"Then with your new power you'll accomplish all sorts of cool stuff 
 in no time, and We'll All Be Sorry.  At that point you can either 
 gloat a bit, and then relent, or go ahead and send the robot army 
 after us." - Quinn Dunkan