I was reading a public blog post among the Zope honchos. And they were agreeing 
that ZPT was a mistake. Some said that Zope ought to have used something like 

Jim Fulton had complained that he had just spent an entire morning setting up 
tests for a ZPT page, and that he, Fulton, was supposed to be good at this 

One good thing about Ruby on Rail's RHTML is that it clearly separates RHTML 
calls with side effects but no returned value ("<%") from those that return a 
value but have no side effects on the page ("<%="). The developer of Eiffel 
says that in good O-O design routines either have side effects or they return a 
value, but not both. 

As for me, I am lazy. ZPT does things in a way that few other frameworks do, 
although Ruby's Amrita2 sort of resembles ZPT. Displaying phantom  text in a 
static page  which will not be there dynamically is not superior to DTML. It is 
just more confusing. I have never learned ZPT well and I may never do so.

Of course, I am not a Zope developer guy, just a user. As for _ being 
confusing, if I need to figure something out about the namespace variable then 
I look it up.

-----Original Message-----
From: Dieter Maurer [mailto:[EMAIL PROTECTED]
Sent: Tuesday, January 09, 2007 2:01 PM
To: Andreas Jung
Cc: Mark, Jonathan (Integic); zope@zope.org
Subject: Re: [Zope] Isn't DTML more like what other frameworks do?

Andreas Jung wrote at 2007-1-8 19:08 +0100:
> ...
>The question is: what is easier to learn and to understand - DTML or ZPT?
>Can you explain the "nonsense" of the _ namespace in DTML to a newbie?


There is only one caveat: that two positional arguments need to
be passed for recursive calls of DTML objects ("None" and "_").

>Can you explain the sequence-item magic with all special cases to a newbie?

There is no need. Newbies do not need to know *all* special cases
only the most essential ones.

The myriads of TALES expression types are almost as bad the
"sequence-item" special cases.

>ZPT is another approach to generate HTML. It's more logical, easier to learn
>and read.

This is arguable...

Zope maillist  -  Zope@zope.org
**   No cross posts or HTML encoding!  **
(Related lists -
 http://mail.zope.org/mailman/listinfo/zope-dev )

Reply via email to