Stephane Bortzmeyer wrote:
On Thu, Sep 14, 2006 at 05:00:20PM +0100,
Stewart Bryant <stbryant(_at_)cisco(_dot_)com> wrote
a message of 56 lines which said:
Isn't there a suitable text based state description language
published by the CCITT that we can use
Pointers are welcome but you probably mean SDL, aka Z.100
(http://www.itu.int/ITU-T/studygroups/com17/languages/Z100.pdf and
http://www.sdl-forum.org/).
Yes - SDL / Z.100 was the one that I was thinking of.
There are already existing languages for state machines and all those
I know are displayed at http://www.cosmogol.fr/related.html.
rather than invent our own from scratch?
There are several reasons why none of them seemed useful for the IETF,
in the specific context of state machine description in the RFCs.
* several are not published as a stable standard (such as Graphviz or
SMC), so they cannot be normative references,
* those who are published as a standard are not always available
(Z.100 is an ITU standard and they do not publish everything freely,
the SDL forum publishes a non-authoritative version and even tutorials
are not freely available, see
http://www.iec.org/acrobat.asp?filecode=125).
Perhaps we can fix this with a liaison to the ITU saying that in the
interests
of co-operation and a desire to use a common state machine language
to ensure the interoperation and uniform interpretation of our protocols
would they make Z.100 available to all who are designing and implementing
IETF protocols.
* some are extremely complex, intended for a much more general use
(such as UML and SDL).
Do we mean more complex or more complete?
Ask yourself why no state machine in the RFC is described with these
languages. And why RFC 2360 does not mention them.
I have no idea.
- Stewart
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf