I whipped up a quick graffle generator: http://webobjects.mdimension.com/wolips/eomodelgraffle-latest.jar
screengrab of output (you will want to click one of the autolayouts -- it doesn't try layout):
java -jar eomodelgraffle-latest.jar -output /tmp/Test.graffle -modelgroup /Users/mschrag/Documents/workspace/CallLogs/Resources
It uses velocity templates for the graffle chunks, so you can make your own if you really want, though the layout within the group is done in code (because it has to compute offsets, etc). Could probably make that fancier, but whatever -- 1 hr.
> There may be personal preferences on how to do things, but I mean, how can you have an overview and discuss about the relationships and the quality and naming of them when you have just a list of entities in a tool?
Just how I work ... I can reword your sentence to say "just a pile of boxes and lines on 30 pages" and make it sound just as silly, I suppose. I find diagrams to be completely worthless. Maybe it's just me. When I go to a website for some new technology and it has an architecture diagram on it, my brain turns off. They just do nothing for me. I'd rather navigate EOModelDoc or just Entity Modeler. I can double-click a relationship and jump to its destination, i can open entities and see their subentities, all without filling my entire wall with a subway map. Maybe it's just me.
> I can't speak for others, but in my environment I have never found anyone questioning the necessity of a diagram view to discuss and review a model within a group of people. May be I am living in another world ,-)
I know lots of people like diagrams. I'm just not one of them.
> My question is, how much of an effort would it be to include a diagram view in Entity Modeler similar to the EOModeler, even if it seems we would be the only ones using them? Or is the answer anyway that will never happen and we have to look for another tool? At least I know then there is no hope and we have to leave and look for something else, although not what we would like.
It's enough effort that I haven't done it. If you care enough about it that you want to fund the development of it, we're always open to that, but I don't really plan on investing a huge amount of time in a diagramming component that will unquestionably be subpar compared to what Graffle can do. It doesn't make sense to me.
ms
This archive was generated by hypermail 2.0.0 : Wed Dec 02 2009 - 12:15:44 EST
