RE: [gmt-dev] State of the GMT project

["Ghica van Emde Boas" <emdeboas@xxxxxxxxxxxx>]

Hi Christophe,
It has been quiet for a while in the GMT space. All the possibilities for
problems for an open source project apply here: lack of time of the
committers because of daytime jobs, some parts of the project did not evolve
in the right direction, differences of opinion, difficulties to bridge time
zones for efficient communication, etc.

Recently the organization of GMT was changed slightly. The originators of
the project, Jorn and myself, are still there. This is a very important area
to explore!!
New committers are added to our project, most notable a group of the
university of Nantes has promised us a QVT implementation, real soon now.
Their project was presented as ATL by Jean Bezivin in our workshop on
Generative Techniques in the context of Model-Driven Architecture at OOPSLA
2003.

We plan to have a similar workshop this year, check it out at:
http://www.softmetaware.com/oopsla2004/mdsd-workshop.html

The content of GMT should be kept tool related, as opposed to methodology
discussions. We have started to put information about Model-Driven Software
Development as a methodology on a new website: www.mdsd.info.
MDSD has a wider meaning than MDA, and hopes to be totally open, in the
spirit of open-source. There is also a yahoo discussion group:
http://groups.yahoo.com/group/MDSD/, that started as the result of a BOF
session at OOPSLA.

Back to GMT. At OOPSLA last year we presented the vision of an open tool-set
interconnected by a workflow layer. We presented a small prototype featuring
an impression of such a layer. A document and the .ppt is in the CVS area of
GMT somewhere. I will add pointers to these, to allow them to be found :-)
The main requirement for a tool to be a GMT node is that it can understand
some form of model as input (presumably XMI), and that it does a
transformation into some other form model or code or text.

A number of candidates can be used as a node in this workflow development
process:
- GME from the Vanderbilt university
http://www.isis.vanderbilt.edu/Projects/gme/
- ATL as mentioned above mailto:Frederic.Jouault@xxxxxxxxxxxxxxxxxxx
- FUUT-je, currenly on the GMT site, a template based generation tool
- Architectureware, a metamodeling and generation tool
http://www.architectureware.de/

Also we are looking at domain specific frameworks that could fit in the
process. The first of those is Time Concious Objects,
http://www.softmetaware.com/tco/overview.html.
A second one could be a case study about the applicability of MDSD to the
PHP/MySQL area that I am doing.

Last but not least, we are having discussions with a number of people who
have contacted us about how they could participate. We would be very happy
if you would join us!

Regards,
Ghica van Emde Boas
Bronstee.com Software & Services b.v.
e-mail: emdeboas@xxxxxxxxxxxx,
tel: +31 23 5474422,
or: +31 6 53368747 (mobile)
fax: +31 23 5473347


> -----Original Message-----
> From: gmt-dev-admin@xxxxxxxxxxx [mailto:gmt-dev-admin@xxxxxxxxxxx]On
> Behalf Of Christophe Ney
> Sent: Wednesday, March 17, 2004 6:30 AM
> To: gmt-dev@xxxxxxxxxxx
> Subject: [gmt-dev] State of the GMT project
>
>
> Hi,
>
> ObjectWeb is in the process of submitting a proposal for
> the Web Tool project and several partners have MDA interest.
>
> I am trying to figure out what the stage of GMT is.
> Can someone update me on this?
>
> Thanks,
> Christophe
>
> Christophe Ney
> Executive Director
> ObjectWeb Consortium
> http://www.objectweb.org
>
> _______________________________________________
> gmt-dev mailing list
> gmt-dev@xxxxxxxxxxx
> http://dev.eclipse.org/mailman/listinfo/gmt-dev
>
> --
> This message has been scanned for viruses and dangerous content
> by MoveNext MailScanner, and is believed to be clean.
>
>