Marc This sounds interesting, and I look forward to reading about it. I have a question: Are you suggesting that people who want to use executable UML should...
I have a question: Are you suggesting that people who want to use executable UML should build their own model compiler? Or perhaps that knowing about how a...
MessageMarc That's a good point about the early days of 3GL compilers -- people were sceptical. I think it is an excellent idea to "de-mystify" model based...
Responding to Balcer... ... Actually, I don't think it is something every shop can do. In fact, my concern would be that it would, indeed, convince observers...
... I know many of the details of how both my computer and car work. This is not because I have any intention of hand assembling either. But it allows me to be...
The same points I would have made had I been typing faster...! -- Marc Balcer ... Not necessarily. To start a chain of analogies, there are a lot of people...
I have a simple (but long) question. Objects can be created synchronously and asynchronously. We can create objects asynchronously by generating creation...
A quick summary answer: The book, even though it uses the BridgePoint action language (OAL), is based upon the UML action semantics. In the action semantics,...
Shayne Cool question sir. But BridgePoint, being a superior product; could be ahead of the book. cheer's me up, Regards Cavan _____ From: shayneflint...
hey all I am just trying to learn abt Execuatble Uml....and i am looking for a open source tool on that..please let me know if there is any thanks vidhya ...
Wow 10 days with no messages! Everyone must be waiting with bated breath for the next installment of "Let's Build a Model Compiler." :-) Well the last...
... Compiler." :-) That hurts ;-). ... seems ... the ... When Steve and I were writing the chapters where we had to address bridges, we went around and around...
... Okay. This makes sense in light of the other answers. My original thought was about whether there were different mappings. ... I had considered this....
Responding to Kamphuis... ... Since no one else has jumped on it, I will take a stab. The "Executable UML" does define an MDA profile for translation....
... The short answer is No. We haven't done that work. The longer answer is that profiles are very informal. Most of the substance is expressed in English,...
Stephen J. Mellor
steve@...
May 11, 2003 1:54 pm
464
Do any of you have a rule of thumb when modeling Security aspects of a system? I say 'aspect' since normally it is _implemented_ using aspected orientated...
... of ... of ... comes ... set ... that ... Funny you should say, "aspect," because in my vocabulary, an aspect of a software system is (can be modeled as) a...
... We tend to separate out into multiple application domains (e.g. online bookstore + warehouse) where we have a large system. To take the example above - one...
Generally I've modeled security as a seperate domain pretty much along the lines suggested by Marc. There have been times when I felt that security policy was...
In my previous job, we found that security, rather than being a monolithic domain was actually composed of several subdomains (or whatever it is they are...
Responding to Durow... ... In addition to Balcer's comments, I would point out that the notion of AOP is particularly relevant for implementing things like...
In the shopping cart example, once a state entry has been run - the class is no longer 'in a state'. E.g. when we have finished running the actions for 'Adding...
Responding to Durow... ... I think this is getting hung up on a naming convention. FWIW, I always name states in the past tense in the pure Moore tradition of...
The differing views of what a state is were thoroughly discussed early this year. You might want to check the archive on yahoogroups. The months with the...
... According to Executable UML (p 152), "an object is in exactly one state at a time." So even though the state procedure has completed, the object is still...
Tsubasa Imai
imai@...
May 18, 2003 5:48 pm
474
Oh, great. Here we go again... ;-) A summary of events so far: 1. State names that are based solely on the state procedure name are perfectly OK from a...
... I agree with this -- it corresponds to my understanding of state. But perhaps the interesting question here is: Should state be the *only* thing that...