On 1/31/06, Sylvain Hellegouarch <[EMAIL PROTECTED]> wrote: > CherryPy can't go faster than it goes now because let's be honest there > are only 3 to 5 regular developers on it and it is not enough. I do > wonder then why people don't join much? Well I suppose it's because they > have other things on their agenda, guess what, same here ;) We also have > other projects that we need to take care of because at the end of the > day we get our money from those and not from CherryPy. I wish I could be > paid for working only on it but this is not the case. CherryPy is a > project I've been enjoying for a while now and I have plenty of ideas > for CP 3 but I can't stop sleeping, can I?
Thanks for raising this point. Cleanly making CherryPy support Paste takes effort, particularly because there are existing users to consider. If people want to help with that effort, I'm sure the help wouild be welcomed. RhubarbTart is interesting, but it's not a drop-in replacement for CherryPy in TurboGears. If people want to work on that on a branch, that's fine, too. But, ultimately, if people want Paste or WSGI integration, the quickest way to make that happen is to help out with the code. Christian Wyglendowski wanted to use WSGI middleware and successfully wrapped CherryPy's server to do so. Ian Bicking wanted to see Paste support in CherryPy and made CherryPaste and helped with RhubarbTart. Those are the kinds of efforts that get things done faster. What TurboGears needs most right now is a new release. That's my #1 goal. I'm keeping an eye on what happens after this next release, though, which is why these discussions are interesting. > This is fairly sad. Therefore if the community here pushes for a > solution that has a better WSGI and paste handling, I don't care because > I want TG to succeed. If it means to leave CherryPy aside, I don't mind > either. But remember that I wouldn't be surprised that by the end of the > summer you end up changing again, and again, and again... While I certainly agree that there will always be complaints about how one thing or another works, I don't think there will actually be very many changes of this sort. It takes time and effort for something new to build up and prove itself, and the CherryPy team has already done that. Generally, it's a lot easier to refactor and modify something you've got to suit new needs than it is to completely write something from scratch (Perl 6? Even Mozilla had a long, bumpy ride before it finally got to where it is today.) Of course, in some languages, it's easier to rewrite from scratch... but most of the Python code I've encountered is pretty readable. > Anyway, for now enjoy your pie whichever it is :) :) Kevin
