From: Scott Woods (scottw_at_[hidden])
Date: 2003-06-08 23:19:21
> for a telecom product. It was great if what you were doing fit well into
> but plugging in say an std::list was a nightmare -- basically undoable (C
> wrappers and such). He hated every minute.
Me too :-)))
> 4) I think you are right that your submission can cover most of the
> ground of both. That said, I for one, think that coverage of UML is
> the more important of the two. A few developers I've met actually
> understand something about UML state diagrams. Barely any have
> any idea about SDL and I've worked on various telecom projects
> for at least 8 years...
Agreed. Andreas has it covered.
Perceived competition between UML and SDL is a bit of a fizzer in that
they have the same roots and huge overlap in goals? Didnt see it like this
myself until recent catch-up. UML wins on familiarity and comfort?
Your point about "few developers" is spot-on. Consistent with a recent
CUJ article that tried to sell the value of state-machines. Used the example
of a VB implementation of calc.exe to highlight the fact that we (developers
at large) tend to avoid proper handling of async events.
Its everywhere (GUIs, devices, networks...) and yet a low percentage
of software solutions are formalized with the likes of UML. Maybe boost::fsm
is going to be sexy enough to lure a few more developers?
8 years of telecoms without SDL? Almost jealous ;-)
> Unsubscribe & other changes:
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk