Home » Modeling » EMF "Technology" (Ecore Tools, EMFatic, etc) » MXF proposal collaboration
|
| Re: MXF proposal collaboration [message #132484 is a reply to message #132422] |
Fri, 23 January 2009 15:36   |
 Michael Soden
Messages: 27
Registered: July 2009 |
Junior Member |
|
|
Hi Marc,
comments below.
Marc Pantel schrieb:
> Dear all,
>
> We have just read your very interesting proposal for creation of the MXF
> project.
>
> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
> development of the model animation tools for the TOPCASED project
> (http://www.topcased.org) together with the ATOS Origin (Eclipse
> commiters) and Anyware Technologies (Eclipse commiters) entreprises in
> relation with the CEA LIST and INRIA Triskell teams inside the TOPCASED
> project.
>
> We have designed for TOPCASED a meta-modeling design pattern for
> integrating execution concerns in any domain specific modeling languages
> based on Ecore. We rely on this pattern to propose a modular and generic
> model execution framework. This pattern has been experimented with
> several kind of action languages (Java language, Kermeta action
> language, endogenous model transformation with ATL and SmartQVT) and has
> thus proven its genericity. It has been applied for implementing the
> UML2 statemachine and the SAM (Structured Analysis Models from TOPCASED)
> model animators using SmartQVT as action language in the generic
> framework.
Sounds very close to what we're heading for. Is there any more
description available for "pattern" you are using? Actually, collegues
of mine at the Humboldt University Berlin have build a similar thing
(see http://eprovide.sourceforge.net/). Its worth looking at, too.
> Another execution kernel for UML2 state machine providing
> Self-* facilities, the MOCAS framework
> (http://sourceforge.net/projects/mocasengine) has also been integrated
> in TOPCASED relying on our model execution framework.
>
That's really interesting. Is the execution kernel for MOCAS somehow
related to the TOPCASED "pattern" you mentioned above?
> We are thus very interested by participating and committing ourselves in
> the MXF project.
Cool, I add you to the list of interested parties.
Best wishes,
Michael
>
> Best regards,
> Marc Pantel
> ACADIE team IRIT
> Leader of the TOPCASED WP2.10 work package on execution semantics for
> modeling languages
>
|
|
|
| Re: MXF proposal collaboration [message #132498 is a reply to message #132484] |
Mon, 26 January 2009 00:26 |
|
Hi Michael,
I've just read the proposal and we (Atos Origin) find it very interesting.
As Marc said Atos Origin had the opportunity to develop the TOPCASED
model simulation engine with two usages on SAM automatons and UML
statecharts with a full M2M approach.
Now that there have been initiatives to merge Papyrus and TOPCASED
approches in UML editors (MDT Papyrus) and in model simulation (one
meeting, first talks), I have the feeling that your proposal comes at
the right time to gather the efforts of all people involved in model
simulation/execution/debugging within the Eclipse Modeling community.
So Atos Origin is ready to participate by giving back our experience and
by contributing the TOPCASED simulation engine (if accepted by the
TOPCASED steering committee)
Best regards
raphaël Faudou
Atos Origin
Main developpers of the TOPCASED WP2.10 work package on execution
semantics for modeling languages
Michael Soden a écrit :
> Hi Marc,
>
> comments below.
>
> Marc Pantel schrieb:
>> Dear all,
>>
>> We have just read your very interesting proposal for creation of the
>> MXF project.
>>
>> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
>> development of the model animation tools for the TOPCASED project
>> (http://www.topcased.org) together with the ATOS Origin (Eclipse
>> commiters) and Anyware Technologies (Eclipse commiters) entreprises in
>> relation with the CEA LIST and INRIA Triskell teams inside the
>> TOPCASED project.
>>
>> We have designed for TOPCASED a meta-modeling design pattern for
>> integrating execution concerns in any domain specific modeling
>> languages based on Ecore. We rely on this pattern to propose a modular
>> and generic model execution framework. This pattern has been
>> experimented with several kind of action languages (Java language,
>> Kermeta action language, endogenous model transformation with ATL and
>> SmartQVT) and has thus proven its genericity. It has been applied for
>> implementing the
> > UML2 statemachine and the SAM (Structured Analysis Models from TOPCASED)
> > model animators using SmartQVT as action language in the generic
> > framework.
> Sounds very close to what we're heading for. Is there any more
> description available for "pattern" you are using? Actually, collegues
> of mine at the Humboldt University Berlin have build a similar thing
> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
>
>> Another execution kernel for UML2 state machine providing Self-*
>> facilities, the MOCAS framework
>> (http://sourceforge.net/projects/mocasengine) has also been integrated
>> in TOPCASED relying on our model execution framework.
>>
> That's really interesting. Is the execution kernel for MOCAS somehow
> related to the TOPCASED "pattern" you mentioned above?
>
>> We are thus very interested by participating and committing ourselves
>> in the MXF project.
> Cool, I add you to the list of interested parties.
>
> Best wishes,
> Michael
>
>>
>> Best regards,
>> Marc Pantel
>> ACADIE team IRIT
>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>> modeling languages
>>
|
|
|
| Re: MXF proposal collaboration [message #132531 is a reply to message #132498] |
Mon, 26 January 2009 18:16 |
Michael Soden
Messages: 27
Registered: July 2009 |
Junior Member |
|
|
Hi Raphaël,
that's cool. So I'll add Atos Origin as well to the list of interested
parties (beside FeRIA).
I checked briefly the TOPCASED website in order to find more information
about the simulation engine you mentioned. Unfortunately, there seems to
be only some french slides on the page. Could you give me a pointer to
more information?
Anyway, what I understood (with my poor french) is that you are having a
framework to integrate KerMeta, ATL and others. Did I get it right? Is
there more information about how you do that?
Many thanks and best regards,
Michael
raphaël Faudou schrieb:
> Hi Michael,
>
> I've just read the proposal and we (Atos Origin) find it very interesting.
>
> As Marc said Atos Origin had the opportunity to develop the TOPCASED
> model simulation engine with two usages on SAM automatons and UML
> statecharts with a full M2M approach.
>
> Now that there have been initiatives to merge Papyrus and TOPCASED
> approches in UML editors (MDT Papyrus) and in model simulation (one
> meeting, first talks), I have the feeling that your proposal comes at
> the right time to gather the efforts of all people involved in model
> simulation/execution/debugging within the Eclipse Modeling community.
>
> So Atos Origin is ready to participate by giving back our experience and
> by contributing the TOPCASED simulation engine (if accepted by the
> TOPCASED steering committee)
>
> Best regards
>
> raphaël Faudou
>
> Atos Origin
> Main developpers of the TOPCASED WP2.10 work package on execution
> semantics for modeling languages
>
>
>
>
> Michael Soden a écrit :
>> Hi Marc,
>>
>> comments below.
>>
>> Marc Pantel schrieb:
>>> Dear all,
>>>
>>> We have just read your very interesting proposal for creation of the
>>> MXF project.
>>>
>>> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
>>> development of the model animation tools for the TOPCASED project
>>> (http://www.topcased.org) together with the ATOS Origin (Eclipse
>>> commiters) and Anyware Technologies (Eclipse commiters) entreprises
>>> in relation with the CEA LIST and INRIA Triskell teams inside the
>>> TOPCASED project.
>>>
>>> We have designed for TOPCASED a meta-modeling design pattern for
>>> integrating execution concerns in any domain specific modeling
>>> languages based on Ecore. We rely on this pattern to propose a
>>> modular and generic model execution framework. This pattern has been
>>> experimented with several kind of action languages (Java language,
>>> Kermeta action language, endogenous model transformation with ATL and
>>> SmartQVT) and has thus proven its genericity. It has been applied for
>>> implementing the
>> > UML2 statemachine and the SAM (Structured Analysis Models from
>> TOPCASED)
>> > model animators using SmartQVT as action language in the generic
>> > framework.
>> Sounds very close to what we're heading for. Is there any more
>> description available for "pattern" you are using? Actually, collegues
>> of mine at the Humboldt University Berlin have build a similar thing
>> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
>>
>>> Another execution kernel for UML2 state machine providing Self-*
>>> facilities, the MOCAS framework
>>> (http://sourceforge.net/projects/mocasengine) has also been
>>> integrated in TOPCASED relying on our model execution framework.
>>>
>> That's really interesting. Is the execution kernel for MOCAS somehow
>> related to the TOPCASED "pattern" you mentioned above?
>>
>>> We are thus very interested by participating and committing ourselves
>>> in the MXF project.
>> Cool, I add you to the list of interested parties.
>>
>> Best wishes,
>> Michael
>>
>>>
>>> Best regards,
>>> Marc Pantel
>>> ACADIE team IRIT
>>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>>> modeling languages
>>>
>
|
|
|
| Re: MXF proposal collaboration [message #132608 is a reply to message #132484] |
Wed, 28 January 2009 18:15 |
Marc Pantel
Messages: 6
Registered: July 2009 |
Junior Member |
|
|
Hi,
>> We have designed for TOPCASED a meta-modeling design pattern for
>> integrating execution concerns in any domain specific modeling languages
>> based on Ecore. We rely on this pattern to propose a modular and generic
>> model execution framework. This pattern has been experimented with
>> several kind of action languages (Java language, Kermeta action
>> language, endogenous model transformation with ATL and SmartQVT) and has
>> thus proven its genericity. It has been applied for implementing the
> > UML2 statemachine and the SAM (Structured Analysis Models from TOPCASED)
> > model animators using SmartQVT as action language in the generic
> > framework.
> Sounds very close to what we're heading for. Is there any more
> description available for "pattern" you are using? Actually, collegues
> of mine at the Humboldt University Berlin have build a similar thing
> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
We are aware of this work as we attended the last three ECMDA. We think
that both approaches could be complementary in order to have a better
genericity/adaptability to various use.
We are currently writing a much improved description for submission.
But we can give some link to a quite short description we wrote for
ERTS'08 (http://www.combemale.fr/research/phd/2008/erts08-CCGMP.pdf). It
is a bit short but gives the main insights.
>> Another execution kernel for UML2 state machine providing
>> Self-* facilities, the MOCAS framework
>> (http://sourceforge.net/projects/mocasengine) has also been integrated
>> in TOPCASED relying on our model execution framework.
>>
> That's really interesting. Is the execution kernel for MOCAS somehow
> related to the TOPCASED "pattern" you mentioned above?
In fact, MOCAS was written independently outside of TOPCASED and has been
integrated later on by using our pattern.
>> We are thus very interested by participating and committing ourselves in
>> the MXF project.
> Cool, I add you to the list of interested parties.
Thanks.
Best regards,
Marc
> Best wishes,
> Michael
>>
>> Best regards,
>> Marc Pantel
>> ACADIE team IRIT
>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>> modeling languages
>>
|
|
|
| Re: MXF proposal collaboration [message #132620 is a reply to message #132531] |
Wed, 28 January 2009 18:28 |
Marc Pantel
Messages: 6
Registered: July 2009 |
Junior Member |
|
|
Hi,
> Hi Raphaël,
> that's cool. So I'll add Atos Origin as well to the list of interested
> parties (beside FeRIA).
> I checked briefly the TOPCASED website in order to find more information
> about the simulation engine you mentioned. Unfortunately, there seems to
> be only some french slides on the page. Could you give me a pointer to
> more information?
Here is a link to the public part of the TOPCASED model simulation
sub-project : http://gforge.enseeiht.fr/projects/topcased-ms/.
> Anyway, what I understood (with my poor french) is that you are having a
> framework to integrate KerMeta, ATL and others. Did I get it right? Is
> there more information about how you do that?
In fact, there are two aspects :
- on the one hand, TOPCASED is an Eclipse based integration of various
technologies and provides in a single bundles a lot of tools related to
MDE.
- on the other hand, the pattern we have proposed for defining the
execution semantics of a DSL is based on several points :
- extension of the metamodels for representing several aspects of the
runtime information related to model execution
- architecture of the execution engine based on two components, the
first one is generic and defines the discrete event based model of
computation we have used currently, the second one is DSL specific and
implements the execution semantics. This one has been defined using
different technologies: Java/EMF, Kermeta, endogenous transformation using
ATL and SmartQVT.
I hope this clears a bit the things. We will look for english available
documents and forward them to you.
Best regards,
Marc
> Many thanks and best regards,
> Michael
> raphaël Faudou schrieb:
>> Hi Michael,
>>
>> I've just read the proposal and we (Atos Origin) find it very interesting.
>>
>> As Marc said Atos Origin had the opportunity to develop the TOPCASED
>> model simulation engine with two usages on SAM automatons and UML
>> statecharts with a full M2M approach.
>>
>> Now that there have been initiatives to merge Papyrus and TOPCASED
>> approches in UML editors (MDT Papyrus) and in model simulation (one
>> meeting, first talks), I have the feeling that your proposal comes at
>> the right time to gather the efforts of all people involved in model
>> simulation/execution/debugging within the Eclipse Modeling community.
>>
>> So Atos Origin is ready to participate by giving back our experience and
>> by contributing the TOPCASED simulation engine (if accepted by the
>> TOPCASED steering committee)
>>
>> Best regards
>>
>> raphaël Faudou
>>
>> Atos Origin
>> Main developpers of the TOPCASED WP2.10 work package on execution
>> semantics for modeling languages
>>
>>
>>
>>
>> Michael Soden a écrit :
>>> Hi Marc,
>>>
>>> comments below.
>>>
>>> Marc Pantel schrieb:
>>>> Dear all,
>>>>
>>>> We have just read your very interesting proposal for creation of the
>>>> MXF project.
>>>>
>>>> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
>>>> development of the model animation tools for the TOPCASED project
>>>> (http://www.topcased.org) together with the ATOS Origin (Eclipse
>>>> commiters) and Anyware Technologies (Eclipse commiters) entreprises
>>>> in relation with the CEA LIST and INRIA Triskell teams inside the
>>>> TOPCASED project.
>>>>
>>>> We have designed for TOPCASED a meta-modeling design pattern for
>>>> integrating execution concerns in any domain specific modeling
>>>> languages based on Ecore. We rely on this pattern to propose a
>>>> modular and generic model execution framework. This pattern has been
>>>> experimented with several kind of action languages (Java language,
>>>> Kermeta action language, endogenous model transformation with ATL and
>>>> SmartQVT) and has thus proven its genericity. It has been applied for
>>>> implementing the
>>> > UML2 statemachine and the SAM (Structured Analysis Models from
>>> TOPCASED)
>>> > model animators using SmartQVT as action language in the generic
>>> > framework.
>>> Sounds very close to what we're heading for. Is there any more
>>> description available for "pattern" you are using? Actually, collegues
>>> of mine at the Humboldt University Berlin have build a similar thing
>>> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
>>>
>>>> Another execution kernel for UML2 state machine providing Self-*
>>>> facilities, the MOCAS framework
>>>> (http://sourceforge.net/projects/mocasengine) has also been
>>>> integrated in TOPCASED relying on our model execution framework.
>>>>
>>> That's really interesting. Is the execution kernel for MOCAS somehow
>>> related to the TOPCASED "pattern" you mentioned above?
>>>
>>>> We are thus very interested by participating and committing ourselves
>>>> in the MXF project.
>>> Cool, I add you to the list of interested parties.
>>>
>>> Best wishes,
>>> Michael
>>>
>>>>
>>>> Best regards,
>>>> Marc Pantel
>>>> ACADIE team IRIT
>>>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>>>> modeling languages
>>>>
>>
|
|
|
| Re: MXF proposal collaboration [message #620601 is a reply to message #132422] |
Fri, 23 January 2009 15:36 |
Michael Soden
Messages: 27
Registered: July 2009 |
Junior Member |
|
|
Hi Marc,
comments below.
Marc Pantel schrieb:
> Dear all,
>
> We have just read your very interesting proposal for creation of the MXF
> project.
>
> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
> development of the model animation tools for the TOPCASED project
> (http://www.topcased.org) together with the ATOS Origin (Eclipse
> commiters) and Anyware Technologies (Eclipse commiters) entreprises in
> relation with the CEA LIST and INRIA Triskell teams inside the TOPCASED
> project.
>
> We have designed for TOPCASED a meta-modeling design pattern for
> integrating execution concerns in any domain specific modeling languages
> based on Ecore. We rely on this pattern to propose a modular and generic
> model execution framework. This pattern has been experimented with
> several kind of action languages (Java language, Kermeta action
> language, endogenous model transformation with ATL and SmartQVT) and has
> thus proven its genericity. It has been applied for implementing the
> UML2 statemachine and the SAM (Structured Analysis Models from TOPCASED)
> model animators using SmartQVT as action language in the generic
> framework.
Sounds very close to what we're heading for. Is there any more
description available for "pattern" you are using? Actually, collegues
of mine at the Humboldt University Berlin have build a similar thing
(see http://eprovide.sourceforge.net/). Its worth looking at, too.
> Another execution kernel for UML2 state machine providing
> Self-* facilities, the MOCAS framework
> (http://sourceforge.net/projects/mocasengine) has also been integrated
> in TOPCASED relying on our model execution framework.
>
That's really interesting. Is the execution kernel for MOCAS somehow
related to the TOPCASED "pattern" you mentioned above?
> We are thus very interested by participating and committing ourselves in
> the MXF project.
Cool, I add you to the list of interested parties.
Best wishes,
Michael
>
> Best regards,
> Marc Pantel
> ACADIE team IRIT
> Leader of the TOPCASED WP2.10 work package on execution semantics for
> modeling languages
>
|
|
|
| Re: MXF proposal collaboration [message #620602 is a reply to message #132484] |
Mon, 26 January 2009 00:26 |
|
Hi Michael,
I've just read the proposal and we (Atos Origin) find it very interesting.
As Marc said Atos Origin had the opportunity to develop the TOPCASED
model simulation engine with two usages on SAM automatons and UML
statecharts with a full M2M approach.
Now that there have been initiatives to merge Papyrus and TOPCASED
approches in UML editors (MDT Papyrus) and in model simulation (one
meeting, first talks), I have the feeling that your proposal comes at
the right time to gather the efforts of all people involved in model
simulation/execution/debugging within the Eclipse Modeling community.
So Atos Origin is ready to participate by giving back our experience and
by contributing the TOPCASED simulation engine (if accepted by the
TOPCASED steering committee)
Best regards
raphaël Faudou
Atos Origin
Main developpers of the TOPCASED WP2.10 work package on execution
semantics for modeling languages
Michael Soden a écrit :
> Hi Marc,
>
> comments below.
>
> Marc Pantel schrieb:
>> Dear all,
>>
>> We have just read your very interesting proposal for creation of the
>> MXF project.
>>
>> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
>> development of the model animation tools for the TOPCASED project
>> (http://www.topcased.org) together with the ATOS Origin (Eclipse
>> commiters) and Anyware Technologies (Eclipse commiters) entreprises in
>> relation with the CEA LIST and INRIA Triskell teams inside the
>> TOPCASED project.
>>
>> We have designed for TOPCASED a meta-modeling design pattern for
>> integrating execution concerns in any domain specific modeling
>> languages based on Ecore. We rely on this pattern to propose a modular
>> and generic model execution framework. This pattern has been
>> experimented with several kind of action languages (Java language,
>> Kermeta action language, endogenous model transformation with ATL and
>> SmartQVT) and has thus proven its genericity. It has been applied for
>> implementing the
> > UML2 statemachine and the SAM (Structured Analysis Models from TOPCASED)
> > model animators using SmartQVT as action language in the generic
> > framework.
> Sounds very close to what we're heading for. Is there any more
> description available for "pattern" you are using? Actually, collegues
> of mine at the Humboldt University Berlin have build a similar thing
> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
>
>> Another execution kernel for UML2 state machine providing Self-*
>> facilities, the MOCAS framework
>> (http://sourceforge.net/projects/mocasengine) has also been integrated
>> in TOPCASED relying on our model execution framework.
>>
> That's really interesting. Is the execution kernel for MOCAS somehow
> related to the TOPCASED "pattern" you mentioned above?
>
>> We are thus very interested by participating and committing ourselves
>> in the MXF project.
> Cool, I add you to the list of interested parties.
>
> Best wishes,
> Michael
>
>>
>> Best regards,
>> Marc Pantel
>> ACADIE team IRIT
>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>> modeling languages
>>
|
|
|
| Re: MXF proposal collaboration [message #620605 is a reply to message #132498] |
Mon, 26 January 2009 18:16 |
Michael Soden
Messages: 27
Registered: July 2009 |
Junior Member |
|
|
Hi Raphaël,
that's cool. So I'll add Atos Origin as well to the list of interested
parties (beside FeRIA).
I checked briefly the TOPCASED website in order to find more information
about the simulation engine you mentioned. Unfortunately, there seems to
be only some french slides on the page. Could you give me a pointer to
more information?
Anyway, what I understood (with my poor french) is that you are having a
framework to integrate KerMeta, ATL and others. Did I get it right? Is
there more information about how you do that?
Many thanks and best regards,
Michael
raphaël Faudou schrieb:
> Hi Michael,
>
> I've just read the proposal and we (Atos Origin) find it very interesting.
>
> As Marc said Atos Origin had the opportunity to develop the TOPCASED
> model simulation engine with two usages on SAM automatons and UML
> statecharts with a full M2M approach.
>
> Now that there have been initiatives to merge Papyrus and TOPCASED
> approches in UML editors (MDT Papyrus) and in model simulation (one
> meeting, first talks), I have the feeling that your proposal comes at
> the right time to gather the efforts of all people involved in model
> simulation/execution/debugging within the Eclipse Modeling community.
>
> So Atos Origin is ready to participate by giving back our experience and
> by contributing the TOPCASED simulation engine (if accepted by the
> TOPCASED steering committee)
>
> Best regards
>
> raphaël Faudou
>
> Atos Origin
> Main developpers of the TOPCASED WP2.10 work package on execution
> semantics for modeling languages
>
>
>
>
> Michael Soden a écrit :
>> Hi Marc,
>>
>> comments below.
>>
>> Marc Pantel schrieb:
>>> Dear all,
>>>
>>> We have just read your very interesting proposal for creation of the
>>> MXF project.
>>>
>>> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
>>> development of the model animation tools for the TOPCASED project
>>> (http://www.topcased.org) together with the ATOS Origin (Eclipse
>>> commiters) and Anyware Technologies (Eclipse commiters) entreprises
>>> in relation with the CEA LIST and INRIA Triskell teams inside the
>>> TOPCASED project.
>>>
>>> We have designed for TOPCASED a meta-modeling design pattern for
>>> integrating execution concerns in any domain specific modeling
>>> languages based on Ecore. We rely on this pattern to propose a
>>> modular and generic model execution framework. This pattern has been
>>> experimented with several kind of action languages (Java language,
>>> Kermeta action language, endogenous model transformation with ATL and
>>> SmartQVT) and has thus proven its genericity. It has been applied for
>>> implementing the
>> > UML2 statemachine and the SAM (Structured Analysis Models from
>> TOPCASED)
>> > model animators using SmartQVT as action language in the generic
>> > framework.
>> Sounds very close to what we're heading for. Is there any more
>> description available for "pattern" you are using? Actually, collegues
>> of mine at the Humboldt University Berlin have build a similar thing
>> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
>>
>>> Another execution kernel for UML2 state machine providing Self-*
>>> facilities, the MOCAS framework
>>> (http://sourceforge.net/projects/mocasengine) has also been
>>> integrated in TOPCASED relying on our model execution framework.
>>>
>> That's really interesting. Is the execution kernel for MOCAS somehow
>> related to the TOPCASED "pattern" you mentioned above?
>>
>>> We are thus very interested by participating and committing ourselves
>>> in the MXF project.
>> Cool, I add you to the list of interested parties.
>>
>> Best wishes,
>> Michael
>>
>>>
>>> Best regards,
>>> Marc Pantel
>>> ACADIE team IRIT
>>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>>> modeling languages
>>>
>
|
|
|
| Re: MXF proposal collaboration [message #620611 is a reply to message #132484] |
Wed, 28 January 2009 18:15 |
Marc Pantel
Messages: 6
Registered: July 2009 |
Junior Member |
|
|
Hi,
>> We have designed for TOPCASED a meta-modeling design pattern for
>> integrating execution concerns in any domain specific modeling languages
>> based on Ecore. We rely on this pattern to propose a modular and generic
>> model execution framework. This pattern has been experimented with
>> several kind of action languages (Java language, Kermeta action
>> language, endogenous model transformation with ATL and SmartQVT) and has
>> thus proven its genericity. It has been applied for implementing the
> > UML2 statemachine and the SAM (Structured Analysis Models from TOPCASED)
> > model animators using SmartQVT as action language in the generic
> > framework.
> Sounds very close to what we're heading for. Is there any more
> description available for "pattern" you are using? Actually, collegues
> of mine at the Humboldt University Berlin have build a similar thing
> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
We are aware of this work as we attended the last three ECMDA. We think
that both approaches could be complementary in order to have a better
genericity/adaptability to various use.
We are currently writing a much improved description for submission.
But we can give some link to a quite short description we wrote for
ERTS'08 (http://www.combemale.fr/research/phd/2008/erts08-CCGMP.pdf). It
is a bit short but gives the main insights.
>> Another execution kernel for UML2 state machine providing
>> Self-* facilities, the MOCAS framework
>> (http://sourceforge.net/projects/mocasengine) has also been integrated
>> in TOPCASED relying on our model execution framework.
>>
> That's really interesting. Is the execution kernel for MOCAS somehow
> related to the TOPCASED "pattern" you mentioned above?
In fact, MOCAS was written independently outside of TOPCASED and has been
integrated later on by using our pattern.
>> We are thus very interested by participating and committing ourselves in
>> the MXF project.
> Cool, I add you to the list of interested parties.
Thanks.
Best regards,
Marc
> Best wishes,
> Michael
>>
>> Best regards,
>> Marc Pantel
>> ACADIE team IRIT
>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>> modeling languages
>>
|
|
|
| Re: MXF proposal collaboration [message #620612 is a reply to message #132531] |
Wed, 28 January 2009 18:28 |
Marc Pantel
Messages: 6
Registered: July 2009 |
Junior Member |
|
|
Hi,
> Hi Raphaël,
> that's cool. So I'll add Atos Origin as well to the list of interested
> parties (beside FeRIA).
> I checked briefly the TOPCASED website in order to find more information
> about the simulation engine you mentioned. Unfortunately, there seems to
> be only some french slides on the page. Could you give me a pointer to
> more information?
Here is a link to the public part of the TOPCASED model simulation
sub-project : http://gforge.enseeiht.fr/projects/topcased-ms/
> Anyway, what I understood (with my poor french) is that you are having a
> framework to integrate KerMeta, ATL and others. Did I get it right? Is
> there more information about how you do that?
In fact, there are two aspects :
- on the one hand, TOPCASED is an Eclipse based integration of various
technologies and provides in a single bundles a lot of tools related to
MDE.
- on the other hand, the pattern we have proposed for defining the
execution semantics of a DSL is based on several points :
- extension of the metamodels for representing several aspects of the
runtime information related to model execution
- architecture of the execution engine based on two components, the
first one is generic and defines the discrete event based model of
computation we have used currently, the second one is DSL specific and
implements the execution semantics. This one has been defined using
different technologies: Java/EMF, Kermeta, endogenous transformation using
ATL and SmartQVT.
I hope this clears a bit the things. We will look for english available
documents and forward them to you.
Best regards,
Marc
> Many thanks and best regards,
> Michael
> raphaël Faudou schrieb:
>> Hi Michael,
>>
>> I've just read the proposal and we (Atos Origin) find it very interesting.
>>
>> As Marc said Atos Origin had the opportunity to develop the TOPCASED
>> model simulation engine with two usages on SAM automatons and UML
>> statecharts with a full M2M approach.
>>
>> Now that there have been initiatives to merge Papyrus and TOPCASED
>> approches in UML editors (MDT Papyrus) and in model simulation (one
>> meeting, first talks), I have the feeling that your proposal comes at
>> the right time to gather the efforts of all people involved in model
>> simulation/execution/debugging within the Eclipse Modeling community.
>>
>> So Atos Origin is ready to participate by giving back our experience and
>> by contributing the TOPCASED simulation engine (if accepted by the
>> TOPCASED steering committee)
>>
>> Best regards
>>
>> raphaël Faudou
>>
>> Atos Origin
>> Main developpers of the TOPCASED WP2.10 work package on execution
>> semantics for modeling languages
>>
>>
>>
>>
>> Michael Soden a écrit :
>>> Hi Marc,
>>>
>>> comments below.
>>>
>>> Marc Pantel schrieb:
>>>> Dear all,
>>>>
>>>> We have just read your very interesting proposal for creation of the
>>>> MXF project.
>>>>
>>>> We, at FeRIA (IRIT, LAAS and ONERA laboratories) are involved in the
>>>> development of the model animation tools for the TOPCASED project
>>>> (http://www.topcased.org) together with the ATOS Origin (Eclipse
>>>> commiters) and Anyware Technologies (Eclipse commiters) entreprises
>>>> in relation with the CEA LIST and INRIA Triskell teams inside the
>>>> TOPCASED project.
>>>>
>>>> We have designed for TOPCASED a meta-modeling design pattern for
>>>> integrating execution concerns in any domain specific modeling
>>>> languages based on Ecore. We rely on this pattern to propose a
>>>> modular and generic model execution framework. This pattern has been
>>>> experimented with several kind of action languages (Java language,
>>>> Kermeta action language, endogenous model transformation with ATL and
>>>> SmartQVT) and has thus proven its genericity. It has been applied for
>>>> implementing the
>>> > UML2 statemachine and the SAM (Structured Analysis Models from
>>> TOPCASED)
>>> > model animators using SmartQVT as action language in the generic
>>> > framework.
>>> Sounds very close to what we're heading for. Is there any more
>>> description available for "pattern" you are using? Actually, collegues
>>> of mine at the Humboldt University Berlin have build a similar thing
>>> (see http://eprovide.sourceforge.net/). Its worth looking at, too.
>>>
>>>> Another execution kernel for UML2 state machine providing Self-*
>>>> facilities, the MOCAS framework
>>>> (http://sourceforge.net/projects/mocasengine) has also been
>>>> integrated in TOPCASED relying on our model execution framework.
>>>>
>>> That's really interesting. Is the execution kernel for MOCAS somehow
>>> related to the TOPCASED "pattern" you mentioned above?
>>>
>>>> We are thus very interested by participating and committing ourselves
>>>> in the MXF project.
>>> Cool, I add you to the list of interested parties.
>>>
>>> Best wishes,
>>> Michael
>>>
>>>>
>>>> Best regards,
>>>> Marc Pantel
>>>> ACADIE team IRIT
>>>> Leader of the TOPCASED WP2.10 work package on execution semantics for
>>>> modeling languages
>>>>
>>
|
|
|
Goto Forum:
Current Time: Sat Apr 27 04:57:12 GMT 2024
Powered by FUDForum. Page generated in 0.03887 seconds |
] 
Search
Help
Register
Login
Home
