Yes, his focus is almost entirely on Simula. I like it, because I think I have the Simula mindset as well (coming from Norway, that is). And in Simula, simulations were important (not exactly surprising given the name).
SmallTalk is something entirely different, and I think it is good that the author didn't try to talk about that as well. (Which only reminds us that OO is quite hard to define.) While I partly agree that it is important to talk about SmallTalk if one wants to get at the origin of OO-programming, that was exactly where the author wanted to avoid going. SmallTalk was probably the first language where the "programming" of OO became more important than the "modelling" and "simulation". And the article was about the latter.
I couldn't find one place where the author talked about something along the lines of MVC. While MVC is a common paradigm for OO GUI's, it is not the only one, or the most widely used (this happens to be: GBOCCBLWP, or "great blob of code confusing business logic with presentation")). The examples about graphical elements were just there to make the reader feel familiar before he went into less familiar territory.
Actually, Simula is even quite nice for simulations today, it has some very nice libraries to help you, and it supports coroutines in the language. A more modern language (i.e: no call-by-name, easier-to-understand coroutines) called Beta is there to support it's legacy today (with some other nifty features, such as a unification of objects and functions (called patterns), and a module system allowing separate compilation of nested patterns).
There is a very good book available as a PDF-file from Mjolner Informatics. It more or less captures the Simula view of OO (Kristen Nygaard is a co-author). It is interesting to see how much OO has drifted from OOA and OOM to OOP after it became popular outside scandinavia. Not that that necessarily is a bad thing, but possibly important to keep in mind.
|