Hi Dietmar.. comments below.
On Oct 1, 2009, at 8:43 AM, Dietmar Gombotz wrote:
How is your Plan on developing acore – as i see the project plan is very ambitious. I would give more time for meta model changes as the acore.
Yes, especially if the release candidate was supposed to be out four months ago! ;) I need to fix that. I did flesh the plan out a while ago so it probably needs adjusting again.
But the acore bit isn't actually as much of a challenge as the related stuff. That's because most of the infrastructure and related tools for MetaABM can be migrated very easily. On the other hand, there are aspects of AXF and AGF that need to be engineered new. The changes that I have been considering for the metamodel are not enormous, but they are significant. And I have been discussing them with a number of people but informally -- I keep trying to get those discussions to take place on the list but the issues are abstract enough that it is hard to get people to engage. I'm going to make another try at getting some public discussion going. My working assumption is that the basic approach is sound and that there are a few aspects that need to be re-designed, and so the bottom line at this point is that we need input about anything that might be really broken or need to be significantly rethought. What really needs to be determined is wether there are significant enough issues to make a decision to delay things.
Note that just because the meta-model update is finished does not mean that all of the related tools and implementations are complete. For example the meta-model might support features that are not yet available on any target platform. And on the other hand, there are enough things that really should be changed before people adopt the tool more broadly that it makes sense to move forward with this as quickly as possible. In other words, even if there is a delay we'll still need an "Acore 0" model. It can't remain as "metaabm"; there is a branding issue there as well.
It is a bit of a chicken and egg thing -- having the meta-model perfect drives adoption, but the meta-model can't be perfect until many people adopt it. So it is quite possible that we'll go through another round of this in year 2 though I think that should be avoided if at all possible. (One thing I really want to get a handle on is how we might do meta-model versioning. i.e. if it is possible to change meta-model structure over multiple versions without changing the file extensions themselves.)
I understood that you want to finish the meta model very soon and then only develop the whole suite – or was this wrongly understood.
No, it's all happening in parallel. For example, if you look at the CVS log you can see that there has been a lot of work lately on the testing and execution infrastructure.
- 1) I never participated in an eclipse project, did some tweaks and fixes in some projects on sourceforge, and had some involvement with the Java Community Process in the last years. I know the tools Eclipse uses and am quite well acquainted with the platform itself (played around with SWT some time ago) - 2) See 1 – but was always more a fixer of bugs when they arose in my project – not a dedicated contributer – this is my first go on such a thing - 3) Depending on my clients – but surely a couple of hours a week (2-5 seems like a realistic time frame) – it also depends if you count the time that i do research as active time – might be double the time - 4) Seems not a big problem (documentation is always the problem it seems! )
Great!! It's a particular weakness of Eclipse projects for some reason and AMP is no exception. The neat thing is that Eclipse has very nice support for built-in documentation. One of the things I've been playing around with are Cheat Sheets. They seem like a really powerful way to walk users through the system, and I'm thinking that a lot of the stuff that would ordinarily be in tutorials might be better as cheat sheets with a link from the page.
- 5) I do not see much problem, expect that have not yet been involved to much in the eclipse projects – Actually if i don’t pass i do think that beeing a contributer surely is a good first step into the eclipse proces
Yes, the big issue is really the IP release and company sign-off stuff. Any issues there can be killers. But not being involved in Eclipse projects before hand is no problem.
For starters: I will dive into the documentation during the weekend and find out what i think is amiss there. Furthermore i will check out the bugzilla. Let’s see where we go from there.
Cool. On the documentation, let me know if you find some. ;)
Seriously, there is some for MetaABM and that can be a starting point for what needs to happen for AMP -- though it is all outdated enough that it probably makes sense to start fresh. Please see:
What I am trying to do now is a bit of triage in terms of providing users some quick paths into the tool itself. I only have a raw outline and I'll be working more on it over the next week..
cheers,
Miles
No questions yet from my point Mit freundlichen Grüßen Dietmar Gombotz --- ---- ---- ---- ---- ---- ---
Dietmar Gombotz, BBSc
Software, Systems, Services Hinweis:
Diese Nachricht und eventuelle Dateianhänge sind vertraulich und nur für den (die) darin angegebenen Adressaten bestimmt. Jeder Gebrauch, jede Weitergabe oder Vervielfältigung durch andere als den (die) bestimmten Adressaten sind verboten. Sollten Sie diese Sendung irrtümlich erhalten haben, verständigen Sie bitte unverzüglich per E-Mail den Absender und löschen Sie die Sendung samt allfälligen Dateianhängen aus Ihrem System. Please note:
This message and any attachment thereof are confidential and may be privileged or otherwise protected from disclosure. If you are not the intended recipient any use, distribution, copying or disclosure is strictly prohibited. If you have received this message in error, please notify the sender immediately by return e-mail and delete this message and any attachment thereof from your system Von: Miles Parker [mailto:milesparker@xxxxxxxxx]
Gesendet: Mittwoch, 30. September 2009 08:28
An: Dietmar Gombotz
Cc: amp-dev@xxxxxxxxxxx
Betreff: Re: Eclipse AMP - Contributing Delighted to hear from you. I'm really looking forward to seeing a team of people come together around this project and it sounds as though you have exactly the right expertise. I hope you don't mind but I've cc'd the developer list (w/o your contact details) for project transparency and as it could be helpful for future developers. Any further follow-up to the amp-dev@xxxxxxxxxxx as well. I'm especially interested to hear your thoughts on MetaABM. As the project plan relates, there is going to be a new meta-model "acore" that will try to learn from MetaABM experience. There are a number of things that I'm already wanting to work on, and some areas that need puzzling out. It's really important to have a good critical look at the current model and very timely as I'm giving a paper on it at the end of October. And I am very glad that you want to work on implementation issues as that is the area that we really need to work on over the next six months or so. And community building is the most important thing of all right now, see below. A few questions to start off not to put you on the spot -- just for context. 1. Do you have any official or unofficial involvement in Eclipse projects now? i.e. bugzilla, blogs, etc.. 2. Any other open source stuff? 3. What kind of time do you think you might have to devote to the project? 4. Are you able to work on documentation, examples and user support as well? (Please say yes. :)) 5. Looking at the guidelines below do you see any issues with getting past Eclipse foundation committer process? (Even if you do, you could still be very involved in project.) As you're probably aware, the Eclipse way to become involved in projects is (quite) a bit more complex than in most Open Source projects...this is the first time I've been through this process as lead as well so I've had to read up on it today. I want to do things right so that we have good consistency in process for the future. I hope that you won't find it too much of an impediment. Please see the links below for some general guidelines. From what I can gather from the Eclipse guidelines, what should happen is for you to take on developer role for a bit. Here are some suggestions for how to get involved. -Obviously, try to get the thing built and installed, etc.. :) -Check out the bugzilla. There isn't a lot there yet, but... add anything and everything you think of that could be done better or needs to be done or whatever. even better, take a look at some of the outstanding bugzilllas and work on patches for them. -Discuss details of meta-model design, API and ideas for what *you'd* like to see in AMP on dev-list. (More user focussed stuff might make sense on amp newsgroup.) -Write documentation and example models! I'm especially wanting to get some cheat sheets and other internal documentation going. More than anything, right now we need to get people engaged in the project and that means getting them up and running and using the tool as quickly as possible, so I would say that this is the best way to contribute right now. There is actually a lot of stuff there now, but it's worthless if no-one knows how to use it. -Along with this, any input on website, screencasts, and build issues is most welcome. On Sep 29, 2009, at 8:24 AM, Dietmar Gombotz wrote:
With great pleasure I have read that your Project has been accepted into the Eclipse Suite Personally I am based on two areas. Primarily I own a small software company in Vienna (Austria) where I focus on specific software needs for Government and Businesses. Secondary I am a student and since recently junior researcher at the Vienna university of technology, where I focus on using Agent Based Models in the Computational Economics area. I am currently working on two projects (Master Thesis and a Paper) regarding Agent Based Modelling and would like to contribute to your project. (the paper is to analyse meta-ABM, criticise and propose improvements) I contact you to see how I could support your efforts. As an experienced software developer in the Java are I see my focus primarily on helping in the Implementation area. -> I have worked with MOF/Ecore twice already (University) and would like to build up more know-how in the MDA area. As a German speaker I could also provide some background on building up a community within the German speaking parts of Europe. Let me know if you think I might be able to help,
|
