Yesterday, Saturday, November 19, 2016, the Modeling Craftsmanship Camp took place in Hanover. A group of approximately 35 people interested in modeling spent their Saturday exchanging ideas, expanding their knowledge, and networking. It was a ton of fun! I would like to sincerely thank the organizers, Stephan Dankers and Michael Jastram. It was a fantastic event in a fantastic location with participants from all industries.
Once again, it was interesting to see how others approach modeling and which methods they use. Time and again, one discovers that there are topics that everyone struggles with. For example, dealing with variants. The more one delves into the topic, the more problems arise, and the methods from textbooks (e.g., VAMOS) quickly reach their limits in reality.
There was consensus that much of the literature on SysML lacks truly practical examples. Either trivial systems are modeled that aren't needed in reality, or there are models for systems with which one has no contact whatsoever. One option should therefore be to share aspects of models that have proven valuable in one's own company as best practice templates or to create a poster based on them. This is a topic I'm currently working on myself. But attending barcamps like these is also a perfect way to see how and what others are modeling. In one session, the moderator modeled an "intelligent mailbox" with the participants. Requirements, use cases, and a system context were modeled. And all this within 45 minutes. Not bad, in my opinion. In another session, the Fundamental Modeling Concept (FMC) was presented, which offers a very simplified language for describing systems with just four model elements. At an event like this, the topic of tools cannot be ignored. The use of requirements tools on the one hand and modeling tools on the other was naturally a topic. For anyone interested in this, I recommend the article Coexistence between requirements and modeling toolsAre models images, or do models have a purpose? This question was examined. And a very sensible answer for me was: If you store information in the model and can obtain new information from it, then it is a model. This definition probably leads, alarmingly often, to models being merely images that are only readable by humans (or well-trained engineers).
Barcamp sponsor Andreas Willert said he sees the Barcamp format as the future of conferences. And I agree. I'm finding it increasingly difficult to get excited about traditional conferences. The frontal approach of conferences, coupled with often boring and low-quality presentations, simply puts me off. If you feel the same way, I definitely recommend attending a Barcamp.
If you missed the event, don't worry. The next events are already planned. On December 3rd, the nord.systemscamp.org in Hamburg and in February 2017 the munich-openspace.de on the topics of functional safety, requirements engineering and systems engineering.
If you follow my blog closely or subscribe to the newsletter, you won’t miss anything.
For me, an event like this, combined with a city trip, is like a little vacation from which I come home full of energy and full of new ideas.
Best regards
Goran Madzar