On 9/21/06, Christophe Lombart [EMAIL PROTECTED] wrote:
On 9/21/06, Edgar Poce [EMAIL PROTECTED] wrote:
On 9/21/06, Christophe Lombart [EMAIL PROTECTED] wrote:
This project is not always clear for some people.
Please don't take wrong my personal opinions on this subject, I hope
you to take it as constructive feedback.
of course. you are welcome :-) It is always interesting to get
feedback. thanks for this mail.
I think that the project
goals sometimes are not clear because the framework part is unbalanced
with the end user part. I think graffito stands somewhere between a
framework for developers and an app for end users and the framework
constraints and indirection layers seem to be priorized over end user
features.
The Graffito foundation are still under construction. So, that's
difficult to add now a lot of user features. Furthermore, the OCM
framework implementation required a lot of efforts and time. that's
why there are not so many user features.
I think that most users who look for a CMS are mostly
interested in feature richness.
The feature list can be very long. Graffito is not only CMS oriented.
Later, it should be also possible to use it in other content oriented
application (document management, asset management, forums, ...). I'm
not sure it is possible to satisfy all user features. I think users
are looking for a platform that can be used for any kind of content
oriented application (ECM) - even if some personalization/dev is
required.
After building the foundation, we should add more flexibility in the
framework before making real applications (and user features).
I'm believing that flexibility is more important than the complete
feature list.
Eg. : use only the services/applications I need; have the possibility
to add custom/new applications or services, access to any kind of
content repo, extend the content object model, modify the
business/content rules, ... If you know Plone, Drupal, ... you
certainly know what I mean. I didn't find such tools in the real open
source java world.
I heard about them but I never used them. I've just taken a look to
some sites and docs and they look interesting. I think in the java
world a JSR-168 portal provides the environment to plug modules in a
very nice way.
If we have this kind of flexibility, the community can help to
implement user features and end-user applications. Is it make sense
for you ?
yes, thanks. However I think that in order to be more productive
regarding end user features the graffito foundation might reduce the
flexibility and leverage existing technologies.
e.g.
1. maybe not every content application need to be rewritten from scratch
2. maybe not all the deployment models (standalone, in a portal or
embedded) should be supported
3. maybe it shouldn't include its own persistence abstraction layer
but leverage a single persistence api.
4. maybe it doesn't need to include in the foundation a new service
framework but use one already written.
5. maybe it doesn't need to include a custom workflow api but support
a single workflow engine.
Again, what should the object model/API for doing it ? ;-)
I'm not sure, it would be interesting to open another thread and
request feedback to the community.
br,
edgar
br,
Christophe
