| Real-Time M2M [message #489] |
Tue, 11 July 2006 18:42  |
Eclipse User |
|
|
|
Originally posted by: irbull.cs.uvic.ca
Hi all,
I am currently doing my PhD in something I call model driven
visualization. The idea is that we use MDE technologies to help design
and customize visualization tools. With the help of the MDE community
(planetmde and others) I have formally defined Meta Models for several
visualization tool (including charts, graphs, maps, etc...). Using
these metamodels (and the associated viewers), I am able to write
transformation rules from existing MetaModels (Business models, UML,
etc..) to my view models and render the results.
I am currently looking at the possibility of Real-Time model
transformations, that is, when input models change I update the view. I
understand that iterative model transformations are inherently difficult
which is why I am am studying them :) I have some theories on minimal
change sets (that is the minimal number of rules that must be executed
for a change to propagate).
I am wondering if there is any interest in the M2M community for this
work. Most of my work has been with EMF and ATL, and I am not that
familiar with MOF and QVT.
A brief overview is available at: http://www.modeldrivenvisualization.com.
but the site is a little dated :)
Cheers,
ian
|
|
|
| Re: Real-Time M2M [message #503 is a reply to message #489] |
Tue, 18 July 2006 10:04  |
| Eclipse User |
|
|
|
Originally posted by: bezivin.univ-nantes.fr
Ian:
There is a master student working on this subject in Nantes.
His name is Mikael Barbero.
I will ask him to describe his current
work on incremental transformations.
All the best,
Jean
> I am currently looking at the possibility of Real-Time model
> transformations, that is, when input models change I update the view. I
> understand that iterative model transformations are inherently difficult
> which is why I am am studying them :) I have some theories on minimal
> change sets (that is the minimal number of rules that must be executed
> for a change to propagate).
>
> I am wondering if there is any interest in the M2M community for this
> work. Most of my work has been with EMF and ATL, and I am not that
> familiar with MOF and QVT.
>
> A brief overview is available at: http://www.modeldrivenvisualization.com.
>
> but the site is a little dated :)
>
> Cheers,
> ian
|
|
|
| Re: Real-Time M2M [message #528 is a reply to message #503] |
Mon, 24 July 2006 07:29 |
| Eclipse User |
|
|
|
Originally posted by: mikael.barbero.free.fr
Hi,
As Jean Bézivin said, I am currently working on this kind of subject.
The main subject of my master thesis is about incremental
transformation. I deeply studied the state of the art in different
domains and especially in XML transformation (XSLT).
For now, i have some basis ideas for a loosely coupled incremental
models transformation with the ATL engine. Thus, all the needed stuff
(elements matching, differencing, and "patching") is generated as ad hoc
transformations (by some HOTs - Higher Order Transformations). Models
matching, differencing and patching are already prototyped.
Currently, I'm writing some research reports about my models
differencing heuristic. They will be published as soon as possible. My
matching and differencing approach will also be published on the GMT
website at the end of the August.
I'm very interested in this field of research, and your project seems to
be attractive. Are you looking for a PhD applicant or research engineer
in order to work on this project ? I would be grateful if you can tell
us more about your wishes.
Thanks in advance,
Best regards,
Mikaël Barbero
> Ian:
>
> There is a master student working on this subject in Nantes.
> His name is Mikael Barbero.
> I will ask him to describe his current
> work on incremental transformations.
>
> All the best,
>
> Jean
>
>> I am currently looking at the possibility of Real-Time model
>> transformations, that is, when input models change I update the view. I
>> understand that iterative model transformations are inherently difficult
>> which is why I am am studying them :) I have some theories on minimal
>> change sets (that is the minimal number of rules that must be executed
>> for a change to propagate).
>>
>> I am wondering if there is any interest in the M2M community for this
>> work. Most of my work has been with EMF and ATL, and I am not that
>> familiar with MOF and QVT.
>>
>> A brief overview is available at: http://www.modeldrivenvisualization.com.
>>
>> but the site is a little dated :)
>>
>> Cheers,
>> ian
>
>
|
|
|
Powered by
FUDForum. Page generated in 0.03344 seconds
] 
Search
Help
Register
Login
Home
