OpenMI Association Technical Committee meeting no 13
Date: January 16, 2008
Venue: Web-meeting
Participants:
Jan Gregersen, LICTEK (gregersen@lictek.dk) (chairman of the meeting)
~psi@dhigroup.com, DHI - Water & Environment (psi@dhigroup.com)
Adrian Harper, Wallingford Software (adrian.harper@wallingfordsoftware.com)
Onno
Jon
Stef
Objectives:
- Planning the period from now until the February OATC meeting.
- News from USA and planning transatlantic corporation
Documents:
http://sourceforge.net/projects/openmi
Agenda :
1. Welcome to Jon and news from USA
- Presentation round.
Since this is the first OATC meeting where Jon participates, we should start off with at short presentation round, where each person shortly presents him selves.
- Any news from USA.
We kindly ask Jon to provide any info from USA about the use of OpenMI.
2. Planning transatlantic corporation
Due to the distance between USA and Europe, it will not be possible for Jon to be physically present at the OATC meetings in Europe. What is the best way to deal with this? Some meeting to be held in USA, using skype during the OATC meetings, other suggestions...
3. planning next OATC meeting and the period from now until the meeting.
The next OATC meeting with the meeting participants physically present will be held in Denmark on February 4th to 6th (less that two weeks from now). During the last four moths we have been occupied with work on OpenMI 1.4 release. OpenMI 1.4 is now released and we can focus almost 100% on OpenMI version 2. At the last OpenMI Executive Committee meeting it was decided that we should have the draft OpenMI 2.0 architecture ready by August 2008, so we need to get in track with the development very soon. Personally I (Jan) feel that the last six moths has been a bit frustrating with respect to OpenMI 2.0 developments. OpenMI 1.4 developments have taken more time than first anticipated and it has been difficult to find the right way of working (should we focus on use cases, should we simply start programming, should we use UML or directly modify the interfaces, should we keep the SDK running during the development of the standard and so on..). When we developed OpenMI 1.0 funded by the HarmonIT project, each company participating in the developments had more that one person working full time on the project, which made it possible to do proper project management. Now the situation is different, everyone is only working part time on the developments, which makes it difficult to make good planning. Anyway, we have to deal with this and adjust our level am ambition to what is possible and find a way of working that fits with the current conditions.
Even though we (and people outside the OATC that understands the details of OpenMI) are happy about what OpenMI can do, we are still waiting for rest of the World to go for OpenMI. Even within our own organizations we face problems with convincing our own colleagues about the wonders of OpenMI. So my suggestion is to focus on the reasons for not using the current OpenMI to determine which changes are the most important.
That was many words so now some bullet points
- Use cases
We have already written some use cases for OpenMI version 2 (see OpenMI version 2 development ). However, there are too many use cases and most of the use cases still need some work before they can be useful. So my (Jan) suggestion is that each of us select one use case, which we feel is most important and make sure that this use case is fully described at latest by Tuesday next week. In this way we will have about four main use cases to work with at the next OATC meeting.
- Iterative development model
When we developed OpenMI 1.0 we followed at nice iterative development approach. One iteration was only completed when all unit tests were running. I (Jan) think that we should start using this approach again. This implies that the GUI, the SDK, and even some models must be kept updated with OpenMI standard developments. Adrian has recently reorganized the source code on source forge. This means that some changes must be made in order to make sure that the source code for verison 2 is at the stage where we last left it. We should have this in place by Tuesday next week also, which will allow everyone to prepare a bit before the meeting.
4. Task status
All tasks are listed on source forge. see. http://sourceforge.net/projects/openmi
5. Any other business
6. Next web meeting
see the calender
