Re: [amalgam-dev] Amalgamation for Helios

[Cédric Brun <cedric.brun@xxxxxxx>]

The 'glue and integration' part was mainly about  - may be - providing
more comprehensive welcome page, help topics and - why not - the amalgam
examples directly within the modeling package.

For the discovery technology, the idea is to provide an accessible and
simple UI  to help the user discovering what modeling components he
needs, something quite similar to the connector discovery from mylyn :

http://tasktop.com/blog/eclipse/growing-open-source-ecosystems-install

It's quite obvious that discovering what project one needs among all
those "Three Letters Acronyms" is a real difficulty for the end user.
Moreover the modeling package size is just insane and providing a base
package with all the modeling component pre-installed make it - if
possible - really difficult to get UI consistency.

So far the implementation is going pretty well, it is based on P2 and on
the Mylyn code extended by an EMF model ;). Basicaly you can define an
EMF model describing the components discovery with information like
"summary, license, update-site" and then install those thanks to P2.
 Once we've got that we can start thinking about a "Modeling Package
Diet" to provide a package with more consistency.

Cédric


Artem Tikhomirov a écrit :
> Cedric,
>
>  I didn't understand what 'glue and integration' code you're talking
> about, as well as the discovery technology you mentioned. Could you
> please shed more light on this?
>
> Regards
> Artem
>
> On Wed, Dec 9, 2009 at 10:28 AM, Cédric Brun <cedric.brun@xxxxxxx
> <mailto:cedric.brun@xxxxxxx>> wrote:
>
>     Hi Amalgamers,
>
>     To provide a better Eclipse Modeling Package no doubt we'll need glue
>     and integration code - just like the other amalgam distributions.
>     My understanding is that if we want to provide this code in the
>     modeling
>     package it should be part of the Helios release and of the
>     update-site -
>     and as such be declared soon.
>     As I was planning to include  a discovery technology for modeling I'm
>     all for it.
>
>     Any objection ?
>
>
>     Cédric
>
>
>     _______________________________________________
>     amalgam-dev mailing list
>     amalgam-dev@xxxxxxxxxxx <mailto:amalgam-dev@xxxxxxxxxxx>
>     https://dev.eclipse.org/mailman/listinfo/amalgam-dev
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> amalgam-dev mailing list
> amalgam-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/amalgam-dev
>