Here's what Jonas says about himself:
"Born in St.Gallen, Switzerland in 1980. Spent one year at Old Lyme highschool in Connecticut, USA. Moved back to St.Gallen to start off as a primary school teacher. Made my degree as an engineer in informatics in 2005. Since then I've been working as a software engineer at the University of applied sciences in Buchs and St.Gallen. Projects include a knowledge based system, process optimization software (GlobalOptimize) and modelling and simulation software (GlobalSimulate), all based on Eclipse. I've been working with Eclipse and building RCP-Applications since 2003."
And here is a little more about the institute. Of course, we're very grateful to have their support for Jonas's and others' efforts as well:
"Institute for modelling and simulation, FHS St.Gallen.. a group of ten people, working for the University of Applied Sciences in St.Gallen, Switzerland. We do research and provide services in managing complex processes by using modelling and simulation techniques. We're using static, system dynamic, discrete event and agent based models. Most solutions are custom made. But we also distribute two general purpose software solutions: GlobalOptimize and GlobalStorehouse. Besides, we're working on a multi-paradigm open source simulation software called GlobalSimulate."
Jonas and the institute's contributions look to be really important to the project. While they have done a lot of work in Agent-Based Modeling, their major software contribution will be in the area of Systems Dynamics modeling, which is an extremely important component of any comprehensive modeling toolkit. Even though the project name is "Agent Modeling Platform", the project really is about much more than agents.
I'll say more about this in future posts, but we already support innovative approaches to Model Execution and Simulation in general, but we haven't been clear about this potential and there are some relatively small improvements to the platform that could support a much broader set of use cases. For example, would you like to be able to take one of your EMF models, define behavior for it using an EMF-based meta-model -- including visual editors and all the other EMF services, and then execute behavior against it dynamically with compiled Java performance? If that sounds good to you, we already have most of the pieces in place. (The only part that is really missing is to get a complete EMF to AMF mapping, as you can already do the same for Java objects using other APIs) Let us know and we'll make that a priority.
Outside of even more general usages, AMP has the ambition to provide support for an entire array of Modeling and Simulation techniques as well as operations research, analytics and other domains. As we grow that vision within Eclipse, we very much want your ideas and input, but more than that we want your real live working code! If any of the above is interesting or important to you or your enterprise, you could be the next new AMP committer.
As a personal note, while many projects start out with a number of committers, AMP has been just me , so when I say "we" in various online conversations, most of the time I've really mean "me". And I've felt like a complete bozo saying "we" all the time. That's not to say that we (see what I mean?) haven't had a lot of input and effort from others, but it is a big difference to have someone who will actually be working on the code day to day. They say that the real measure of success of an Open Source project is the day that you can walk away and know that it will succeed without you and so today is a happy day indeed. Don't worry though Jonas, I'm not leaving yet. :)
0 comments:
Post a Comment