forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <stef...@apache.org>
Subject Re: analysis paralysis :-)
Date Wed, 27 Feb 2002 16:29:52 GMT
John Morrison wrote:
> 
> > From: Stefano Mazzocchi [mailto:stefano@apache.org]
> >
> > Steven Noels wrote:
> >
> > > where do we go from here?
> > >
> > > 1) 'force' the entire apache community into maintaining xml docs for
> > > their project website: should be no problem
> >
> > There is no alternative: separation of concerns is king if we want to
> > keep the docs scalable... otherwise this project will become the
> > bottleneck and I don't want this.
> 
> Without doubt they should use xml.  Question is - should they all *have*
> to use the same DTD/schema?

I would be a great plus if they did, don't you think?

But it's utopical to think they will do just because we tell them to do
it.

So we have to come up with an easy and appealing transition process.

> > > 2) force root@xml.apache.org to run tomcat/cocoon: seems to be a
> > > problem, in terms of:
> > >    - need: for static content, this is overkill
> > >    - platform: not sure whether machines are up to spec for this
> > >    - people: are we ready (teamwise) to support a high-visibility
> > > website
> >
> > These are all good points.
> >
> > My original idea was to run Forrest and Gump from 'gump.covalent.net'
> > (and have it renamed 'forrest.covalent.net'), this will not be a
> > high-visibility web site, but a medium visibility testbed to show the
> > potentials of the platform... in order to show root@apache.org what
> > functionality we could have if we allowed dynamic technology on Apache
> > (finally!)
> 
> Testbed first.

yes, totally

> > BTW, PHP is a official ASF project but runs entirely on their own
> > resources.
> >
> > moreover, all jakarta mail lists run on nagoya, which is a Sun-owned
> > machine (a big E4500, some 200K$ worth of hardware, sitting there
> > running Bugzilla, doing gump builds and sending jakarta email, with load
> > that never gets higher than 0.01, at least this is what Pier told me
> > since I don't have an account on that machine... but I know Sam does)
> 
> Just out of curiosity - what webserver/servlet engine do they have
> running?  (I've got to 'port' cocoon2 onto an E10K (not sure about the
> number))

I don't know but I think I can bet it's apache + tomcat :)

> > Sam, what's your view on this.
> >
> > > 3) incremental development:
> > >
> > > are we absolutely positively in need of a dynamic website with browser
> > > negotiation from the beginning? in need of user comments/write-ins and
> > > other dynamic features?
> > >
> > > or should we start incrementally and start out with the visible part,
> > > i.e. 'portal look & feel' and aggregation from different information
> 
> Is this going to be a portal?  News to me :)  Wouldn't mind if it is, but
> login may become a problem unless it's either an apache id or guest.

forget about that for now.
 
> > > sources - staying static for as long as possible?
> > >
> > > given the state of the current xml.apache.org wildlife park, I tend to
> > > say we should start slowly and try to gather people around the same
> > > website structure and look&feel, and withhold all dynamic features until
> > > we have a new xml.apache.org based on forrest cocoon generation
> > >
> > > and we should decide so upfront because it is difficult to start working
> > > otherwise
> >
> > Ok, I change my -1 on staticity to a -0, but I *want* a clean URI
> > space... this will require Forrest to produce .htaccess files with
> > mod_rewrite rules, but I think this isn't that hard to do. We could also
> > perform browser negotiation using those rewrite rules.
> >
> > What do you think?
> 
> I'm -0 overall...
> 
> +1 for clean URI - that's a given.
> static/dynamic... not so sure, after all dynamic's the whole point of
> Cocoon (for me at least).  I think I'd sooner go for dynamic and judge
> the performance, we can always wget -m and host that if it's not upto the
> job (and I'd be suprised if it isn't...).

AHHHHHHHHHHGGGGGGGGGGg!!!!

damn it, John, I spent 'months' designing the cocoon system to that the
same sitemap could be used for both dynamic *and* static generation
without the need for wget/teleport or stuff like that!

did you ever run cocoon from the command line on top of your dynamic
stuff?

That's why we need forrest: not even the people here know cocoon's
functionalities... go figure the rest of the ASF people :(

-- 
Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<stefano@apache.org>                             Friedrich Nietzsche
--------------------------------------------------------------------

Mime
View raw message