Skip to main content


Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Archived » Service Oriented Architecture Tools Platform (STP) » STP BPMN Metamodel + Editor
STP BPMN Metamodel + Editor [message #374360] Thu, 06 December 2007 16:26 Go to next message
Maria Agustina Cibran is currently offline Maria Agustina CibranFriend
Messages: 13
Registered: July 2009
Junior Member
Hi all.

I am working on implementing an ATL transformation from STP BPMN to UML
activity diagrams. When inspecting the STP BPMN metamodel in detail, I
observe some inconsistencies/issues which pose some problems for realizing
such a transformation:

1) the representation of the lanes are a pure graphical concept: suppose
we have a bpmn model defining 'pool1' and two lanes within pool1, namely
'lane1' and 'lane2', and that 'lane1' contains 2 activities, namely
'activity1' and 'activity2'. Behind the scenes, the instance of the BPMN
metaclass 'Lane' that gets created for 'lane1', does NOT refer to the
activities that belong to that lane. In other words, the property
'Lane>>activities' is not set properly. This implies that the lanes are
JUST graphical and that the activities are not assigned to lanes but kept
in a "big bag" (the 'vertices' property of the pool). I find this strange,
specially considering that the metamodel does foresee the relation between
Activities and Lanes by providing an association between these two
metaclasses explicit. FYI I tried this with both the STP BPMN and the
Intalio editor.

In my transformation having this property set correctly is crucial, as I
need to create UML nodes in a UML activity diagram that are associated to
the corresponding partition.

I would appreciate if anyone can tell me why this is, if it is maybe a
current limitation of the STP editor and if it is the case, when it is
expected to fix this issue.

2) the representation of message and sequence flows seems to be
inconsistent in the metamodel: I observed that in the STP BPMN metamodel
message flows (represented by the MessagingEdge metaclass) are linked to
activities, whereas sequence flows are linked to vertices (Vertex
metaclass). Is there a reason for this distinction? Why not link both
message and sequence flows to activities?

Thanks for you help on these issues!
Kind regards.
Agustina
Re: STP BPMN Metamodel + Editor [message #374362 is a reply to message #374360] Thu, 06 December 2007 16:51 Go to previous messageGo to next message
Eclipse UserFriend
Originally posted by: atoulme.intalio.com

Hi,

please see comments inline:

Maria Agustina Cibran wrote:
> Hi all.
>
> I am working on implementing an ATL transformation from STP BPMN to UML
> activity diagrams. When inspecting the STP BPMN metamodel in detail, I
> observe some inconsistencies/issues which pose some problems for
> realizing such a transformation:
>
> 1) the representation of the lanes are a pure graphical concept: suppose
> we have a bpmn model defining 'pool1' and two lanes within pool1, namely
> 'lane1' and 'lane2', and that 'lane1' contains 2 activities, namely
> 'activity1' and 'activity2'. Behind the scenes, the instance of the BPMN
> metaclass 'Lane' that gets created for 'lane1', does NOT refer to the
> activities that belong to that lane. In other words, the property
> 'Lane>>activities' is not set properly. This implies that the lanes are
> JUST graphical and that the activities are not assigned to lanes but
> kept in a "big bag" (the 'vertices' property of the pool). I find this
> strange, specially considering that the metamodel does foresee the
> relation between Activities and Lanes by providing an association
> between these two metaclasses explicit. FYI I tried this with both the
> STP BPMN and the Intalio editor.
This is most probably broken. The relationship between an activity and a
lane is maintained through the edit policies IIRC.

Lanes are not important to us so far, so we haven't worked on them much.

As the BPMN editor as concentrated to have a simple meta model, with
good graphical capacities, we chose to emphasize the lane painting. We
based our work on the spec, where it is shown that an activity may be
owned by a lane, or be shared between two lanes, etc.
That's why we have chosen to show lanes as graphical elements.

Regarding the vertices property, I would like to know what you miss here
.. Can you tell us what you wanted to see there ?

>
> In my transformation having this property set correctly is crucial, as I
> need to create UML nodes in a UML activity diagram that are associated
> to the corresponding partition.
>
> I would appreciate if anyone can tell me why this is, if it is maybe a
> current limitation of the STP editor and if it is the case, when it is
> expected to fix this issue.

This is not an issue, just a work very much in progress. Currently we
are working regarding the internationalization of the BPMN modeler.

Feel free to join the effort and bring us to the next level! We will
welcome you contributions.
> 2) the representation of message and sequence flows seems to be
> inconsistent in the metamodel: I observed that in the STP BPMN metamodel
> message flows (represented by the MessagingEdge metaclass) are linked to
> activities, whereas sequence flows are linked to vertices (Vertex
> metaclass). Is there a reason for this distinction? Why not link both
> message and sequence flows to activities?
Sequence flows apply to subprocesses and activities, while messaging
edges only apply to activities. Please see the BPMN specification for
more information.
>
> Thanks for you help on these issues!
> Kind regards.
> Agustina
>
>

Thanks,

Antoine

--
Intalio, the Open Source BPMS Company

<a href="http://www.intalio.com">http://www.intalio.com</a>
<a href="http://bpms.intalio.com">Community website</a>
Re: STP BPMN Metamodel + Editor [message #374363 is a reply to message #374362] Fri, 07 December 2007 10:48 Go to previous messageGo to next message
Maria Agustina Cibran is currently offline Maria Agustina CibranFriend
Messages: 13
Registered: July 2009
Junior Member
Thanks a lot Antoine for your fast answer!

please see comments inline.

>> I am working on implementing an ATL transformation from STP BPMN to UML
>> activity diagrams. When inspecting the STP BPMN metamodel in detail, I
>> observe some inconsistencies/issues which pose some problems for
>> realizing such a transformation:
>>
>> 1) the representation of the lanes are a pure graphical concept: suppose
>> we have a bpmn model defining 'pool1' and two lanes within pool1, namely
>> 'lane1' and 'lane2', and that 'lane1' contains 2 activities, namely
>> 'activity1' and 'activity2'. Behind the scenes, the instance of the BPMN
>> metaclass 'Lane' that gets created for 'lane1', does NOT refer to the
>> activities that belong to that lane. In other words, the property
>> 'Lane>>activities' is not set properly. This implies that the lanes are
>> JUST graphical and that the activities are not assigned to lanes but
>> kept in a "big bag" (the 'vertices' property of the pool). I find this
>> strange, specially considering that the metamodel does foresee the
>> relation between Activities and Lanes by providing an association
>> between these two metaclasses explicit. FYI I tried this with both the
>> STP BPMN and the Intalio editor.
> This is most probably broken. The relationship between an activity and a
> lane is maintained through the edit policies IIRC.

can the code of the STP editor be modified so that the relation between a
lane and its activities is kept?

sorry, I'm not aware of the edit policies IIRC...(?)

> Lanes are not important to us so far, so we haven't worked on them much.

this is a pity... one of the reasons we have chosen Intalio in my project
is because of its support for lanes...

> As the BPMN editor as concentrated to have a simple meta model, with
> good graphical capacities, we chose to emphasize the lane painting. We
> based our work on the spec, where it is shown that an activity may be
> owned by a lane, or be shared between two lanes, etc.
> That's why we have chosen to show lanes as graphical elements.

the graphical capacities are very good indeed. And for me it is not a
problem that an activity is shared among lanes as long as the editor can
determine (based on the graphical position) to which lane(s) the
activities belong.

> Regarding the vertices property, I would like to know what you miss here
> .. Can you tell us what you wanted to see there ?

In order to be able to transform an STP BPMN model using a transformation
(e.g.ATL), all the relavant properties of the instances that get created
behind the scenes (by the editor) need to be well-set according to what
the metamodel specifies...The only thing I am missing is the link between
the lanes and their activities.

>>
>> In my transformation having this property set correctly is crucial, as I
>> need to create UML nodes in a UML activity diagram that are associated
>> to the corresponding partition.
>>
>> I would appreciate if anyone can tell me why this is, if it is maybe a
>> current limitation of the STP editor and if it is the case, when it is
>> expected to fix this issue.

> This is not an issue, just a work very much in progress. Currently we
> are working regarding the internationalization of the BPMN modeler.

> Feel free to join the effort and bring us to the next level! We will
> welcome you contributions.
>> 2) the representation of message and sequence flows seems to be
>> inconsistent in the metamodel: I observed that in the STP BPMN metamodel
>> message flows (represented by the MessagingEdge metaclass) are linked to
>> activities, whereas sequence flows are linked to vertices (Vertex
>> metaclass). Is there a reason for this distinction? Why not link both
>> message and sequence flows to activities?
> Sequence flows apply to subprocesses and activities, while messaging
> edges only apply to activities. Please see the BPMN specification for
> more information.

This is true indeed. Thanks for the clarification!

Kind regards,
Agus
Re: STP BPMN Metamodel + Editor [message #374364 is a reply to message #374363] Fri, 07 December 2007 11:36 Go to previous message
Eclipse UserFriend
Originally posted by: atoulme.intalio.com

Hi,

first technical details:

Edit policies are a GEF thing, that is reused by GMF.

When you manipulate a diagram, you use a tool, that issues a request.

Every edit policy is polled for a command.
If the commands cannot be executed, the action is forbidden. Otherwise
the commands are executed. All this is cool because it comes with
undo/redo support, and you can add behaviors to your shapes.

But, ehrmm, it's kinda big.

The link between lanes and activities is here, but not maintained: you
can get the lane activities by doing: lane.getActivities().

You need to create an edit policy for the activity parts, that will
change the lane activities depending on the position of the activity for
the lanes.

You will also need to code in your generator a way to compute the
coordinates of the activities according to their lanes. You can either
use ATL to compute the views, or our own generation API. Both will need
work to get this particular issue working.

Now, some more details about Intalio:
If you happen to be an Intalio subscription customer, you are at the
wrong place. Please contact your Intalio contact and we will try to find
a way to help you.

Intalio offers to develop features depending on customer requests
(community users and subscription customers).
To do that you can enter a D3 request on the community web site, and
fund the project. That way we will be able to allocate time and energy
to do that, and the result will be released in the BPMN component.
You'll need to contact our professional services for this.


Maria Agustina Cibran wrote:
> Thanks a lot Antoine for your fast answer!
>
> please see comments inline.
>
>>> I am working on implementing an ATL transformation from STP BPMN to
>>> UML activity diagrams. When inspecting the STP BPMN metamodel in
>>> detail, I observe some inconsistencies/issues which pose some
>>> problems for realizing such a transformation:
>>>
>>> 1) the representation of the lanes are a pure graphical concept:
>>> suppose we have a bpmn model defining 'pool1' and two lanes within
>>> pool1, namely 'lane1' and 'lane2', and that 'lane1' contains 2
>>> activities, namely 'activity1' and 'activity2'. Behind the scenes,
>>> the instance of the BPMN metaclass 'Lane' that gets created for
>>> 'lane1', does NOT refer to the activities that belong to that lane.
>>> In other words, the property 'Lane>>activities' is not set properly.
>>> This implies that the lanes are JUST graphical and that the
>>> activities are not assigned to lanes but kept in a "big bag" (the
>>> 'vertices' property of the pool). I find this strange, specially
>>> considering that the metamodel does foresee the relation between
>>> Activities and Lanes by providing an association between these two
>>> metaclasses explicit. FYI I tried this with both the STP BPMN and the
>>> Intalio editor.
>> This is most probably broken. The relationship between an activity and
>> a lane is maintained through the edit policies IIRC.
>
> can the code of the STP editor be modified so that the relation between
> a lane and its activities is kept?
>
> sorry, I'm not aware of the edit policies IIRC...(?)
>
>> Lanes are not important to us so far, so we haven't worked on them much.
>
> this is a pity... one of the reasons we have chosen Intalio in my
> project is because of its support for lanes...
>
>> As the BPMN editor as concentrated to have a simple meta model, with
>> good graphical capacities, we chose to emphasize the lane painting. We
>> based our work on the spec, where it is shown that an activity may be
>> owned by a lane, or be shared between two lanes, etc.
>> That's why we have chosen to show lanes as graphical elements.
>
> the graphical capacities are very good indeed. And for me it is not a
> problem that an activity is shared among lanes as long as the editor can
> determine (based on the graphical position) to which lane(s) the
> activities belong.
>
>> Regarding the vertices property, I would like to know what you miss
>> here .. Can you tell us what you wanted to see there ?
>
> In order to be able to transform an STP BPMN model using a
> transformation (e.g.ATL), all the relavant properties of the instances
> that get created behind the scenes (by the editor) need to be well-set
> according to what the metamodel specifies...The only thing I am missing
> is the link between the lanes and their activities.
>
>>>
>>> In my transformation having this property set correctly is crucial,
>>> as I need to create UML nodes in a UML activity diagram that are
>>> associated to the corresponding partition.
>>>
>>> I would appreciate if anyone can tell me why this is, if it is maybe
>>> a current limitation of the STP editor and if it is the case, when it
>>> is expected to fix this issue.
>
>> This is not an issue, just a work very much in progress. Currently we
>> are working regarding the internationalization of the BPMN modeler.
>
>> Feel free to join the effort and bring us to the next level! We will
>> welcome you contributions.
>>> 2) the representation of message and sequence flows seems to be
>>> inconsistent in the metamodel: I observed that in the STP BPMN
>>> metamodel message flows (represented by the MessagingEdge metaclass)
>>> are linked to activities, whereas sequence flows are linked to
>>> vertices (Vertex metaclass). Is there a reason for this distinction?
>>> Why not link both message and sequence flows to activities?
>> Sequence flows apply to subprocesses and activities, while messaging
>> edges only apply to activities. Please see the BPMN specification for
>> more information.
>
> This is true indeed. Thanks for the clarification!
>
> Kind regards,
> Agus
>
>

Antoine

--
Intalio, the Open Source BPMS Company

<a href="http://www.intalio.com">http://www.intalio.com</a>
<a href="http://bpms.intalio.com">Community website</a>
Re: STP BPMN Metamodel + Editor [message #605688 is a reply to message #374360] Thu, 06 December 2007 16:51 Go to previous message
Eclipse UserFriend
Originally posted by: atoulme.intalio.com

Hi,

please see comments inline:

Maria Agustina Cibran wrote:
> Hi all.
>
> I am working on implementing an ATL transformation from STP BPMN to UML
> activity diagrams. When inspecting the STP BPMN metamodel in detail, I
> observe some inconsistencies/issues which pose some problems for
> realizing such a transformation:
>
> 1) the representation of the lanes are a pure graphical concept: suppose
> we have a bpmn model defining 'pool1' and two lanes within pool1, namely
> 'lane1' and 'lane2', and that 'lane1' contains 2 activities, namely
> 'activity1' and 'activity2'. Behind the scenes, the instance of the BPMN
> metaclass 'Lane' that gets created for 'lane1', does NOT refer to the
> activities that belong to that lane. In other words, the property
> 'Lane>>activities' is not set properly. This implies that the lanes are
> JUST graphical and that the activities are not assigned to lanes but
> kept in a "big bag" (the 'vertices' property of the pool). I find this
> strange, specially considering that the metamodel does foresee the
> relation between Activities and Lanes by providing an association
> between these two metaclasses explicit. FYI I tried this with both the
> STP BPMN and the Intalio editor.
This is most probably broken. The relationship between an activity and a
lane is maintained through the edit policies IIRC.

Lanes are not important to us so far, so we haven't worked on them much.

As the BPMN editor as concentrated to have a simple meta model, with
good graphical capacities, we chose to emphasize the lane painting. We
based our work on the spec, where it is shown that an activity may be
owned by a lane, or be shared between two lanes, etc.
That's why we have chosen to show lanes as graphical elements.

Regarding the vertices property, I would like to know what you miss here
.. Can you tell us what you wanted to see there ?

>
> In my transformation having this property set correctly is crucial, as I
> need to create UML nodes in a UML activity diagram that are associated
> to the corresponding partition.
>
> I would appreciate if anyone can tell me why this is, if it is maybe a
> current limitation of the STP editor and if it is the case, when it is
> expected to fix this issue.

This is not an issue, just a work very much in progress. Currently we
are working regarding the internationalization of the BPMN modeler.

Feel free to join the effort and bring us to the next level! We will
welcome you contributions.
> 2) the representation of message and sequence flows seems to be
> inconsistent in the metamodel: I observed that in the STP BPMN metamodel
> message flows (represented by the MessagingEdge metaclass) are linked to
> activities, whereas sequence flows are linked to vertices (Vertex
> metaclass). Is there a reason for this distinction? Why not link both
> message and sequence flows to activities?
Sequence flows apply to subprocesses and activities, while messaging
edges only apply to activities. Please see the BPMN specification for
more information.
>
> Thanks for you help on these issues!
> Kind regards.
> Agustina
>
>

Thanks,

Antoine

--
Intalio, the Open Source BPMS Company

<a href="http://www.intalio.com">http://www.intalio.com</a>
<a href="http://bpms.intalio.com">Community website</a>
Re: STP BPMN Metamodel + Editor [message #605691 is a reply to message #374362] Fri, 07 December 2007 10:48 Go to previous message
Maria Agustina Cibran is currently offline Maria Agustina CibranFriend
Messages: 13
Registered: July 2009
Junior Member
Thanks a lot Antoine for your fast answer!

please see comments inline.

>> I am working on implementing an ATL transformation from STP BPMN to UML
>> activity diagrams. When inspecting the STP BPMN metamodel in detail, I
>> observe some inconsistencies/issues which pose some problems for
>> realizing such a transformation:
>>
>> 1) the representation of the lanes are a pure graphical concept: suppose
>> we have a bpmn model defining 'pool1' and two lanes within pool1, namely
>> 'lane1' and 'lane2', and that 'lane1' contains 2 activities, namely
>> 'activity1' and 'activity2'. Behind the scenes, the instance of the BPMN
>> metaclass 'Lane' that gets created for 'lane1', does NOT refer to the
>> activities that belong to that lane. In other words, the property
>> 'Lane>>activities' is not set properly. This implies that the lanes are
>> JUST graphical and that the activities are not assigned to lanes but
>> kept in a "big bag" (the 'vertices' property of the pool). I find this
>> strange, specially considering that the metamodel does foresee the
>> relation between Activities and Lanes by providing an association
>> between these two metaclasses explicit. FYI I tried this with both the
>> STP BPMN and the Intalio editor.
> This is most probably broken. The relationship between an activity and a
> lane is maintained through the edit policies IIRC.

can the code of the STP editor be modified so that the relation between a
lane and its activities is kept?

sorry, I'm not aware of the edit policies IIRC...(?)

> Lanes are not important to us so far, so we haven't worked on them much.

this is a pity... one of the reasons we have chosen Intalio in my project
is because of its support for lanes...

> As the BPMN editor as concentrated to have a simple meta model, with
> good graphical capacities, we chose to emphasize the lane painting. We
> based our work on the spec, where it is shown that an activity may be
> owned by a lane, or be shared between two lanes, etc.
> That's why we have chosen to show lanes as graphical elements.

the graphical capacities are very good indeed. And for me it is not a
problem that an activity is shared among lanes as long as the editor can
determine (based on the graphical position) to which lane(s) the
activities belong.

> Regarding the vertices property, I would like to know what you miss here
> .. Can you tell us what you wanted to see there ?

In order to be able to transform an STP BPMN model using a transformation
(e.g.ATL), all the relavant properties of the instances that get created
behind the scenes (by the editor) need to be well-set according to what
the metamodel specifies...The only thing I am missing is the link between
the lanes and their activities.

>>
>> In my transformation having this property set correctly is crucial, as I
>> need to create UML nodes in a UML activity diagram that are associated
>> to the corresponding partition.
>>
>> I would appreciate if anyone can tell me why this is, if it is maybe a
>> current limitation of the STP editor and if it is the case, when it is
>> expected to fix this issue.

> This is not an issue, just a work very much in progress. Currently we
> are working regarding the internationalization of the BPMN modeler.

> Feel free to join the effort and bring us to the next level! We will
> welcome you contributions.
>> 2) the representation of message and sequence flows seems to be
>> inconsistent in the metamodel: I observed that in the STP BPMN metamodel
>> message flows (represented by the MessagingEdge metaclass) are linked to
>> activities, whereas sequence flows are linked to vertices (Vertex
>> metaclass). Is there a reason for this distinction? Why not link both
>> message and sequence flows to activities?
> Sequence flows apply to subprocesses and activities, while messaging
> edges only apply to activities. Please see the BPMN specification for
> more information.

This is true indeed. Thanks for the clarification!

Kind regards,
Agus
Re: STP BPMN Metamodel + Editor [message #605693 is a reply to message #374363] Fri, 07 December 2007 11:36 Go to previous message
Eclipse UserFriend
Originally posted by: atoulme.intalio.com

Hi,

first technical details:

Edit policies are a GEF thing, that is reused by GMF.

When you manipulate a diagram, you use a tool, that issues a request.

Every edit policy is polled for a command.
If the commands cannot be executed, the action is forbidden. Otherwise
the commands are executed. All this is cool because it comes with
undo/redo support, and you can add behaviors to your shapes.

But, ehrmm, it's kinda big.

The link between lanes and activities is here, but not maintained: you
can get the lane activities by doing: lane.getActivities().

You need to create an edit policy for the activity parts, that will
change the lane activities depending on the position of the activity for
the lanes.

You will also need to code in your generator a way to compute the
coordinates of the activities according to their lanes. You can either
use ATL to compute the views, or our own generation API. Both will need
work to get this particular issue working.

Now, some more details about Intalio:
If you happen to be an Intalio subscription customer, you are at the
wrong place. Please contact your Intalio contact and we will try to find
a way to help you.

Intalio offers to develop features depending on customer requests
(community users and subscription customers).
To do that you can enter a D3 request on the community web site, and
fund the project. That way we will be able to allocate time and energy
to do that, and the result will be released in the BPMN component.
You'll need to contact our professional services for this.


Maria Agustina Cibran wrote:
> Thanks a lot Antoine for your fast answer!
>
> please see comments inline.
>
>>> I am working on implementing an ATL transformation from STP BPMN to
>>> UML activity diagrams. When inspecting the STP BPMN metamodel in
>>> detail, I observe some inconsistencies/issues which pose some
>>> problems for realizing such a transformation:
>>>
>>> 1) the representation of the lanes are a pure graphical concept:
>>> suppose we have a bpmn model defining 'pool1' and two lanes within
>>> pool1, namely 'lane1' and 'lane2', and that 'lane1' contains 2
>>> activities, namely 'activity1' and 'activity2'. Behind the scenes,
>>> the instance of the BPMN metaclass 'Lane' that gets created for
>>> 'lane1', does NOT refer to the activities that belong to that lane.
>>> In other words, the property 'Lane>>activities' is not set properly.
>>> This implies that the lanes are JUST graphical and that the
>>> activities are not assigned to lanes but kept in a "big bag" (the
>>> 'vertices' property of the pool). I find this strange, specially
>>> considering that the metamodel does foresee the relation between
>>> Activities and Lanes by providing an association between these two
>>> metaclasses explicit. FYI I tried this with both the STP BPMN and the
>>> Intalio editor.
>> This is most probably broken. The relationship between an activity and
>> a lane is maintained through the edit policies IIRC.
>
> can the code of the STP editor be modified so that the relation between
> a lane and its activities is kept?
>
> sorry, I'm not aware of the edit policies IIRC...(?)
>
>> Lanes are not important to us so far, so we haven't worked on them much.
>
> this is a pity... one of the reasons we have chosen Intalio in my
> project is because of its support for lanes...
>
>> As the BPMN editor as concentrated to have a simple meta model, with
>> good graphical capacities, we chose to emphasize the lane painting. We
>> based our work on the spec, where it is shown that an activity may be
>> owned by a lane, or be shared between two lanes, etc.
>> That's why we have chosen to show lanes as graphical elements.
>
> the graphical capacities are very good indeed. And for me it is not a
> problem that an activity is shared among lanes as long as the editor can
> determine (based on the graphical position) to which lane(s) the
> activities belong.
>
>> Regarding the vertices property, I would like to know what you miss
>> here .. Can you tell us what you wanted to see there ?
>
> In order to be able to transform an STP BPMN model using a
> transformation (e.g.ATL), all the relavant properties of the instances
> that get created behind the scenes (by the editor) need to be well-set
> according to what the metamodel specifies...The only thing I am missing
> is the link between the lanes and their activities.
>
>>>
>>> In my transformation having this property set correctly is crucial,
>>> as I need to create UML nodes in a UML activity diagram that are
>>> associated to the corresponding partition.
>>>
>>> I would appreciate if anyone can tell me why this is, if it is maybe
>>> a current limitation of the STP editor and if it is the case, when it
>>> is expected to fix this issue.
>
>> This is not an issue, just a work very much in progress. Currently we
>> are working regarding the internationalization of the BPMN modeler.
>
>> Feel free to join the effort and bring us to the next level! We will
>> welcome you contributions.
>>> 2) the representation of message and sequence flows seems to be
>>> inconsistent in the metamodel: I observed that in the STP BPMN
>>> metamodel message flows (represented by the MessagingEdge metaclass)
>>> are linked to activities, whereas sequence flows are linked to
>>> vertices (Vertex metaclass). Is there a reason for this distinction?
>>> Why not link both message and sequence flows to activities?
>> Sequence flows apply to subprocesses and activities, while messaging
>> edges only apply to activities. Please see the BPMN specification for
>> more information.
>
> This is true indeed. Thanks for the clarification!
>
> Kind regards,
> Agus
>
>

Antoine

--
Intalio, the Open Source BPMS Company

<a href="http://www.intalio.com">http://www.intalio.com</a>
<a href="http://bpms.intalio.com">Community website</a>
Previous Topic:STP BPMN Metamodel + Editor
Next Topic:SCA Composite Tools commited
Goto Forum:
  


Current Time: Thu Mar 28 10:39:26 GMT 2024

Powered by FUDForum. Page generated in 0.04113 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software

Back to the top