Re: [boost] New library in Vault: Msm (Meta State Machine)

Subject: Re: [boost] New library in Vault: Msm (Meta State Machine)
From: Andreas Huber (ahd6974-spamboostorgtrap_at_[hidden])
Date: 2008-10-16 18:26:15

> > What do t & D stand for (they do not appear in Figure 2)? They don't 
> > seem to be
> > necessary to implement the chart?
> You should also read the mini doc I included.

Right, sorry I missed that one.

> I added them to make the problem a bit more interesting and show how to 
> solve it
> with flags.

Ok.

>> 1) Is it possible to implement the transitions exactly as shown in the 
>> chart?
> It is not, again as explained in the mini doc. I see this as an UML goto
> and decided to implement superior tools instead.

So IIUC then you think that transitions crossing state boundaries are bad 
practice and should be avoided at all costs? In general, I agree that you 
should avoid them when you can, e.g. in the example chart the transition 
triggered by y should rather be a self-transition of C, which is 
semantically equivalent. OTOH, the transition triggered by x can only be 
made non-state-boundary-crossing by introducing a guard, which is IMO less 
concise than allowing for the transition as shown in the chart.

Moreover, things become a bit more messy if you want to allow for multiple 
ways of entering a composite state. For example consider Figure 8 in

<http://www.wisdom.weizmann.ac.il/~dharel/SCANNED.PAPERS/Statecharts.pdf>

Here the "alarms-beep" outer state exists primarily because you don't want 
to duplicate the "any button pressed" and "30 sec in alarms-beep" 
transitions for all three different alarm beep states. The other diagrams in 
the document show more uses of state-boundary-crossing transitions. Granted, 
Harel sometimes uses them excessively, in places where he could simply have 
transitioned to the outer state instead and let the default transition take 
care of going to the inner initial state. But in a few places I don't see 
any equally concise substitute for such transitions.

As an aside, I don't see much of a connection of state-boundary crossing 
transitions with goto. Given your code in the C() function, any transition 
could/would be implemented with goto?

> Of course I could implement these entry and exit goto-like pseudo states 
> if users want
> them.I just strongly doubt they are needed.

Given the examples in the Harel paper and the fact that the UML standard 
expressly allows them, I would encourage you to support them. BTW, for the 
same reasons I don't detect value in seeing a composite state as a fully 
fledged state machine (as your framework seems to do). I know that the UML 
has so-called submachines but they are really just simple composite states 
with entry and exit points thrown in.

Just my 2c.

Regards,

-- 
Andreas Huber
When replying by private email, please remove the words spam and trap
from the address shown in the header.