Sjur Moshagen wrote:
Hello,
Another "user"/lurker POV:
Den 15. aug. 2006 kl. 13.46 skrev Ross Gardler:
I disagree. The sitemap is a way of *configuring* this complex
processing, it is not the processing itself. The sitemap has become
an XML programming language and I hate it for that reason.
To me, the sitemap setup/programming is clean and easy, I really like
it, irrespective of whether it is "configuring" or "programming".
The quote is taken out of context. I agree with your observation in
general terms, but not in the context of the original discussion.
One other place where I think Forrest is using Cocoon is i18n. This is
one of Forrest's weaker points - it is hard to set up, and there's
really no comprehensive documentation (despite several efforts by a
number of people). Cocoon provides a set of tools to set up i18n, and I
find i18n in Cocoon to be easier to use than in Forrest, despite it
being based on Cocoon.
Implementing proper i18n needs to be a requirement of the new Forrest -
whether this is an argument for the present Cocoon-based solution or in
favour of a new engine is beyond me:-)
I totally agree that i18n is poor in Forrest. This is not a symptom of
Cocoon more a symptom that none of the active developers have a need to
develop it further. That is, it works sufficiently for those active
developers who use it.
The fact that a number of people have tried and failed to improve on our
current usage (including some very experienced Cocoon developers) is
just one example of the problems I am trying to describe.
Having said that, unless a developer who actually needs i18n features
becomes active this will not change any time soon, whether we continue
on the current road or try a new one. My personal belief is that a
simpler Forrest will allow such a developer to become active. Although I
have not put much thought into how i18n would work. Perhaps we can
discuss this if the community agree to my prototype going into SVN (when
it is a little more advanced).
Ross
