On Sat, 2 Mar 2002, Chris Withers wrote:
kosh@aesaeion.com wrote:
- A site with DTMLed templates tend to be a big bunch of templates and ZPT seems to be fewer with many more Python Script objects
That is because DTML is often abused that is not an inherent flaw in DTML.
Au contr�re, that is DTML's BIGGEST inherent flaw. You can write nice DTML, if you try pretty damn hard, but it's soooooo easy to get into foot shooting land.
You can write bad python pretty easily also but that is not a flaw of python. Many people have been encouraged to write bad DTML from lots of sources on zope.org. Even the zope book has had some pretty bad examples which does not help the issues.
Dreamweaver and that kind of editing doesn't matter to me at all since neither we or any of our customers use that.
But that doesn't mean other people don't use it. I didn't like the idea of Dreamweaver until I started using it with WebDrive over WebDAV. Now I'm a convert ;-)
Well I don't run a windows OS and do all my devel on linux. With KDE 3.0 webdav will be transparent across any kde app so I will probably use webdav more but I will still use it with DTML.
Why do you think ZPT is longer laster or more scalable?
It's gonna be supported longer, it was better designed for the task, etc. I could go on but you won't believe me anyway and I'm close enough to RSI-land anyway...
I doubt support for DTML can be pulled for a good while. Talk to TheJester he has some very good arguments for why DTML needs to stay around. It has been discussed a fair bit on #zope and the channel is fairly well divided. I think ZPT was designed for a task of allowing web designers to work on stuff and insulate them from the code. The problem I see is that it took so long to get python script in zope. It encouraged people to put lots of logic in DTML since external methods are a pain by comparison. If examples where updated that seperatd the logic into python scripts and the presentation in DTML it would work well.