It occurs to me that just as there is value in micro-formats (structured data embedded in HTML using existing HTML syntax and well-known conventions), there is also value in micro-models: models embedded in code (e.g., Java) using existing Java syntax (e.g., annotations).
The key advantages are model and code are in a single place and are [...]
541 * w00t: I’ve just heard that David Hearnden has submitted his thesis “Deltaware: Incremental Change Propagation for Automating Software Evolution in the Model-Driven Architecture” which, amongst many other things, contains a formal semantics for the model transformation language implemented by Tefkat (in an appendix).
David Hearnden is having trouble with EMF and damn-it if I’m not having exactly the same problem and debugging experience. Debugging this stuff is like trying to debug interpreted code by single-stepping through the interpreter!
[update: It seems that the problem is not with the serialisation, but that the Resource "contains" the objects twice. [...]
This afternoon Kerry and I spent some time looking at a particular aspect of Tefkat that was troubling us. We had a couple of rules of the following form:
RULE one(f, b)
FORALL Foo f
MAKE DYNAMIC Bar b // implicitly FROM one(f)
;
RULE two(f, b) EXTENDS [...]
I’ve just committed a bunch of changes to Tefkat without a log message! I’m really annoyed about that partly because I spent a good deal of time determining exactly what all the changes did, and partly because now that I’ve made this mistake (I swear that Eclipse never prompted me for the message) I [...]
The other day I was wondering about why we even need things like MOF/EMF (ignoring the cross-language story). That is, why don’t modern OO languages include associations (and their containment/multiplicity constraints) as first-class concepts.
Today I browsed over to this article by Dave Thomas (of smalltalk fame), a sometime DSTC visitor, and found this sentence:
Closely [...]
No builds available yet, but you can grab the code (LGPLed) from subversion if you’re really keen:
svn co https://svn.sourceforge.net/svnroot/tefkat/trunk tefkat
Note that I expect there will be a complete package rename (to replace the vestigal com.dstc) before the first release.
Here are some quick responses to this paper:
Not accessible via ant, can not be composed into larger transformation chains.
An Ant Task is in the works, but Tefkat can be called from the command line so it is invokable from Ant and Make. To be fair, this is not documented and the current distribution is in [...]
Here’s some feedback on Tefkat that I found in a paper discovered via a persistent Google search for “Tefkat”. I don’t have time to address all the issues right now, but I thought I should post them anyway. I will note that several criticism apply to very early versions of Tefkat and are [...]