[ebxml-msg] RE: Public usage scenario documents

[Martin W Sachs <mwsachs@us.ibm.com>]

David,

In addition to baked beans, some people would like to do financial
transactions where messaging reliability does matter.  Even with the baked
beans, a system failure could mean that you receive two container-loads of
beans instead of the one you thought you ordered because you  re-sent the
order, thinking that the first one was not delivered when it was delivered.

We do not have a note in the msg spec which says that reliability is not
guaranteed in the face of system failures. In general, unless we lay out
the rule for determining whether a use case is in the 95% that work or the
5% that don't work, people will meet disasters.

"The requirements were not clear" is one example of under-specification.
No doubt, the Hubble Space Telescope initial disaster was also a case of
under-specification of test cases if nothing else.

Regards,
Marty

*************************************************************************************

Martin W. Sachs
IBM T. J. Watson Research Center
P. O. B. 704
Yorktown Hts, NY 10598
914-784-7287;  IBM tie line 863-7287
Notes address:  Martin W Sachs/Watson/IBM
Internet address:  mwsachs @ us.ibm.com
*************************************************************************************


                                                                                                                             
                      David RR Webber -                                                                                      
                      XMLGlobal                To:       Martin W Sachs/Watson/IBM@IBMUS                                     
                      <Gnosis_@compuser        cc:       "Cutler, Roger (RogerCutler)" <RogerCutler@chevrontexaco.com>,      
                      ve.com>                   Randy Clark <Randy.Clark@bakerhughes.com>, "'bhaugen'"                       
                                                <linkage@interaccess.com>, ebxml-msg@lists.oasis-open.org, eBTWG List        
                      05/28/2002 09:01          <ebtwg@lists.ebtwg.org>, "'Duane Nickull'" <duane@xmlglobal.com>,            
                      AM                        Christopher Ferris <chris.ferris@sun.com>                                    
                                               Subject:  RE: Public usage scenario documents                                 
                                                                                                                             
                                                                                                                             
                                                                                                                             



Message text written by "Martin W Sachs"
>
- Except that in this case, there is a small matter of interoperability
between arms-length implementations, so merely relying on implementers to
know what to do won't get us there unless the specification is precise and
complete.
<<<<<<<<<<<<<<<<<<

Marty,

I'm looking for a balance here - we have to understand that we are NOT
trying to put orbiters around far flung planets, but instead trying to put
a better cheaper tins of baked beans on the shelves at Walmart by truck
and railroad.

So while I understand the desires to get things right - we need to separate
out goals - and say - hey for right now - for the implementations being
fielded
we have enough - we have technical notes in the spec's giving clear
marks where the extended functionality is underconstruction.

And we should certainly not be saying the spec' does not work or is broken,
when for 95+% of use cases its more than good enough.

Now when it comes to the W3C - they have a happy habit of getting the
requirements totally mismashed, IMHO.  So if we can do anything for the
W3C it is getting their requires straight before hand.

Strangely enough - the rockets failed not because of the technical bit and
bytes - but because the requirements were not clear...

a) All measurements will be in metric.

b) Re-use of old components from earlier sub-systems must be crosschecked
     and approved.

If one half of the team is building a cruise-ship and the other half a
ferry, you're
in big trouble!

So I'm not against extended work here - we just need to understand why
and for what audience / use case needs.

Cheers, DW.