RE: The synchronous/asynchronous definition (was RE: Snapshot of Web Services Glossary on Response types)

I can't answer the question of whether it is worth this much effort.
Here is what I think the situation is:

1 - There are definitions of synchronous and asynchonous currently in
the Glossary.  They are not good and definitely need to be replaced by
something -- or eliminated entirely, I suppose.  I think (hope)
everybody agrees with this.

2 - The terms are certainly used in the Usage Scenarios document, and I
think that they are probably going to be in the Architecture document
itself, right?

3 - This thread makes it VERY clear (to me at least) that not everybody
has the same thing in mind when they use the terms.  It seems to me that
this is a good reason to say that the terms really need to be in the
Glossary -- and then some discipline needs to be exerted to remind
people to use the terms in whatever that sense might be.

4 - We have had at least one other lengthy thread, or series of threads,
on this subject a number of months ago.  These threads died away and
much of the material in the current threads is similar (although this
one is more detailed).  One option would be to let this thread die out
without resolution.  In that case, it seems likely that there will be
yet another similar thread several months from now that essentially
treads the same ground YET AGAIN.

My personal opinions:

A - People with different backgrounds approach the terms in different
ways.  That is, there is a synchronous(J, J=1,..,N) where N seems to me
to be about three or four.

B - For each J the questions are not particularly unanswerable and there
may be some nits but they are not big deals.

C - I'd like to see us, as quickly as possible, choose one value of J,
put a definition on paper, and get on with it.

D - If C is not possible, I'd like to see us choose more than one value
of J, associate adverbs modifying the terms with those values of J, put
the definitions on paper, require that the terms only be used with a
modifier in the documents -- and get on with it.

E - It seems to me that there has been ENOUGH discussion.  It would seem
to me useful for the participants in the discussion to field their
candidates for C and D -- and to GET ON WITH IT.

-----Original Message-----
From: Champion, Mike [mailto:Mike.Champion@SoftwareAG-USA.com] 
Sent: Tuesday, February 25, 2003 11:12 AM
To: www-ws-arch@w3.org
Subject: The synchronous/asynchronous definition (was RE: Snapshot of
Web Services Glossary on Response types)



Maybe I'm missing something, but this seems like either a nit that we're
spending too much time on or an unanswerable question that better minds
than ours have failed to resolve over the last 20 years or so.

It's important to maintain focus on the cases that we can actually add
value to, and ignore the ones that will get better by themselves or will
die anyway. (Beating the "triage" metaphor into the ground). I'm
*personally* (not wearing chair hat) not convinced that this is a good
use of our time. Could someone one the WG explain why we consider the
definition of synch/asynch worth this much effort on the mailing list?
Are we in striking distance of an acceptable definition?

Thanks!

Received on Tuesday, 25 February 2003 12:55:45 UTC