At 07:31 AM 7/8/2007 -0400, Jim Fulton wrote: > > I've tried to encourage people to use this, but they get stuck on the > > word "paste", so there's not many other people who consume or produce > > these entry points except for use with Paste or related packages > > (Pylons, etc). I'm not sure what to do about that, except perhaps to > > reset people's opinions with this rewrite. > >Well, this touches a nerve with me. I have a similar problem with >people rejecting out of hand anything that happens to live in the >zope or even the zc namespace. Similarly, at PyCon, I try to always >give at least talk that I think will be generally interesting to the >Python community at large, yet many people tend to assume that these >are Zope specific. I think this behavior is extremely unhealthy for >the Python community. Paste deploy is the only effort I've seen to >provided a much-needed glue layer for WSGI. It doesn't matter one >bit to me what it's called. It tries to fill a basic need. Now I'm >all for competition. If someone really wanted to come up with >something better, then I wouldn't mind seeing someone try, but >nothing else is happening AFAICT. I certainly have other things I >want to work on. Paste Deploy is a really good start and, FWIW, the >Zope community is embracing it.
Just a side note, but this is one reason my CheeseShop packages are named things like "BytecodeAssembler", "DecoratorTools", "SymbolType", "ProxyTypes", etc. -- even though the actual modules contained in those packages are things like peak.util.assembler, peak.util.decorators, etc. When breaking out a formerly-monolithic package, there's usually little reason to continue the "monolithy" in package names. Understandably, Zope is different because you had an early dependency management system that *did* use the imported package names to declare dependencies. And, with the number of core Zope packages dwarfing even my array of CheeseShop packages, it's also understandable that you'd prefer not to remember two sets of names. Not that I'm all that bright here regarding names, mind you; I just went from having a package called RuleDispatch to adding a new one called PEAK-Rules... d'oh! Still, bearing this rule in mind would be good for any future monolith breakups. At least the rate of *new* monoliths seems to be decreasing, though, and the spattering of new multi-part packages on the CheeseShop seems likely to increase the public awareness that things named similarly don't necessarily require every possible piece of a brand's offerings to be installed. _______________________________________________ Web-SIG mailing list Web-SIG@python.org Web SIG: http://www.python.org/sigs/web-sig Unsubscribe: http://mail.python.org/mailman/options/web-sig/archive%40mail-archive.com
