On Wed, 2008-04-09 at 12:25 -0700, Mike Orr wrote: > On Wed, Apr 9, 2008 at 11:24 AM, Cliff Wells <[EMAIL PROTECTED]> wrote: > > > > > > On Wed, 2008-04-09 at 08:50 -0700, Mike Orr wrote: > > > > > The problem with C libraries is a lot bigger than just Google. It > > > frustrates users on Windows and Macintosh to no end, and many of them > > > give up trying to install Pylons/lxml/ToscaWidgets/wxPython and go on > > > to something else. Precompiled binaries don't always exist, are too > > > old, hard to find, or built with the wrong C compiler or Unicode > > > width. > > > > > > > I think this is less true that it used to be. Microsoft now makes a > > free version of Visual Studio available which is sufficient to compile > > Python extensions on Windows: > > > > http://www.microsoft.com/express/download/ > > But is every Pylons user willing to/smart enough to/allowed to install > and run the compiler?
I'd think that if they were allowed to install Python (which also doesn't ship with Windows), they'd be allowed to install some Microsoft tools. Granted, some places might have policy issues with this, but I'm unconvinced that it's Pylons' place to work around specific policy issues (especially ones that we can only speculate about). As far as "smart enough", well... don't get me started. Anything that raises the bar enough to slow the expansion of crappy, insecure applications on the web is a +1 from me ;-) > More to the point, it's pretty far from what > they wanted to do in the first place, which is to write a web > application. If Pylons doesn't do it conveniently, they'll go to > something else. Arguing whether it's Microsoft's fault or not doesn't > change the situation. One can say we don't care about those users, > but I don't think Pylons wants to do that. Pylons is not TurboGears. It's clearly geared toward the DYI type. I can think of several things that are far more basic to web development and far more difficult in Pylons than compiling a C extension: authentication and authorization, managing your database connection, etc. 0.9.7 will even make using a template engine a bit less "automatic". I obviously don't have much of a problem with the above (otherwise I wouldn't be here), but let's be consistent in the level of difficulty we can expect Pylons users to face. Installing the Microsoft development tools is a click-through install, as it is on most Linux systems. I believe on Mac it's an additional CD. Building the extension is usually as simple as "python setup.py build" or similar. If it isn't, then that should be addressed upstream. I don't expect any extension author would object to patches that made his program more easily installable. If running a build script is too hard for people, I can absolutely assure you that using AuthKit is completely beyond their abilities. I wouldn't want to rely on whatever auth they might cook up on their own either ;-) That being said, I don't disagree with the basic premise that unless there's a serious benefit, C extensions should be avoided. I just dislike the idea that they are under a blanket exclusion. I mean, I'd probably have to build psycopg or the MySQL adapter if I were on Solaris or some other lesser supported platform. Someone, somewhere is going to be forced to build *something* on some platform. I'd rather see it documented than pretend like it's not going to happen or let it limit Pylons' performance or features. Regards, Cliff --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "pylons-discuss" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en -~----------~----~----~----~------~----~------~--~---
