On 25/09/2012 16:05, Jos Snellings wrote:
Hi Javier,

Thank you, this is clear.
1. I agree with the fact that the "blocks" mechanism (SSF) is more modular than mounting submaps.
2. That been said, I need to bear that in mind to estimate the work. Just need to assess if they
    translate into each other.

By the way, to my knowledge there is a production site with Apache Cocoon 3:

http://thesaurus.european-heritage.net

That is the one that I created.
is there another one?

Yep: http://cocoon.apache.org/live_sites_3.0.html

Do you want yours to be added?

Regards.

On Tue, Sep 25, 2012 at 3:58 PM, Javier Puerto <javier@apache.org> wrote:
Hi Jos

2012/9/25 Jos Snellings <jos.snellings@upperware.biz>
Dear all,

Considering the upgrade cost of an existing project to cocoon-3,

You have to bear in mind that it's still in beta and there's some components that's are not migrated yet but we have one production site working with Apache Cocoon 3.
 
I have the following question:

what happened to map:mount?

Was deprecated for 2.2.x version.
 

As in:
 <map:match pattern="**">
        <map:mount src="residuary-treatment.xmap" check-reload="no" uri-prefix="" />
 </map>

It was a handy modularity mechanism to reduce the size of sitemaps.

Is it no longer there?
Do we need it? Do we want it?

For Apache Cocoon 2.2.x and 3.0, the "block" is the new unit of modularization. You can see some examples here:
http://cocoon.apache.org/2.2/1291_1_1.html

It's based on Servlet Service Framework subproject. More info here:
http://cocoon.apache.org/subprojects/servlet-service/servlet-service-impl/architecture.html
-- 
Francesco Chicchiriccò

ASF Member, Apache Cocoon PMC and Apache Syncope PPMC Member
http://people.apache.org/~ilgrosso/