RE: [ebxml-dev] New WSCI spec competes with BPSS....

[Adam Sroka <AdamS@rewardsplus.com>]

Title: RE: [ebxml-dev] New WSCI spec competes with BPSS....

I think that this discussion is great. I like conspiracy theories, and I find that there is almost always some truth to them. On the other hand, I think that they usually take a little bit of truth and go way too far with it. I don't think that the reason that the community has failed, thus far, to center around a single standard, be it ebXML, BPMI, or anything else, is because the vendors are out there looking for proprietary lock downs. It is certainly something that happens. We've all seen Microsoft do business. But, I don't think it is a foregone conclusion in this instance.

First, if you take a sincere look at the business strategies of companies like IBM and Sun you'll see that they are more likely to be hurt by lockdown in this area than to make any real profit off of it, even if they are the sole owners of the proprietary technologies that form the "lockdown." The whole idea of interoperability is that different businesses with different ways of doing business are able to do business together. For many modern companies this absolutely mandates open standards, and this trend appears to be growing.

Hopefully, we all remember the thread that went on here a few weeks ago about IBM's IPR claims to some of the technologies. In a world where companies like Microsoft are allowed to do business, that kind of move can be scary. On the other hand, if you take a quick look at IBM's licensing strategy you'll see that they take a strong position on open standards, support open source initiatives, and only take a hard line on protecting their IPR because it is the legally sane thing to do in this day and age. In fact, IBM releases much of their code under a special open source license that differs from the LGPL mostly in this regard (And is in many regards a more "bulletproof" open source license because it protects the author's rights without interfering with the community process.)

When you get down to it, I think that there are a couple of real good reasons that the industry is failing to unite behind ebXML:

1. The technology is new and unproven. It doesn't matter how good it is, because nobody has really done it successfully yet. That is enough to scare away any MBA on the planet. Until somebody takes the initiative and makes it work the industry won't want to unite behind it. Unfortunately, since the standards are about interoperability, one person taking the initiative isn't enough. That person has to interoperate with someone else. And that someone else is hard to find before the technology is proven. This is what we call a "Catch 22". We're going to have to find a way around it, but it will take time.

2. The standard is big. Very big. That means there is additional risk to taking it on, because you've got to do a lot of work up front to adopt it. The developers (You guys) envision that implementers will build software that makes it easier for the less technical (Or rich) companies to adopt it, but until that happens it narrows the potential candidates even further.

So, why then don't the big companies build implementations that they can sell? Open standards meet proprietary software and everybody is happy. The problem is that it is extremely easy for a big company like Sun or BEA to send an employee or two, full time, to participate in a standards body and come up with a new way to solve the problem. However, when it comes to actually implementing that solution that takes a lot of money, time, and other resources. So, if I am Sun I have to make a decision: Am I ready to put my efforts behind implementing ebXML? Or do I send a couple of engineers off to a meeting to discuss an easier way to solve the problem (Whether that is what they actually come up with or not?)

The answer: until there is compelling evidence that my efforts to back a standard will reward me with real dollars, I am going to continue to explore other avenues regardless of the technical merits. If geeks like you and me ruled the world it would be a very different place (I'm not quite arrogant enough to say better ;-) Until then, this is the way the people who make decisions think.

Thanks,
Adam

 -----Original Message-----
From:   Andrzej Jan Taramina [mailto:andrzej@chaeron.com]
Sent:   13 June, 2002 16:26
To:     bhaugen
Cc:     ebxml-dev@lists.ebxml.org
Subject:        Re: [ebxml-dev] New WSCI spec competes with BPSS....

Bob asks

> How is it an "ebxml killer"?

The contact has since clarified that to be "bpss killer"...which is still a bit strange
given the FAQ entry on the relationship between WSCI and ebXML BPSS that you
quoted.

> Why do they want to kill ebXML?

Don't know.

> Why do you think they want to kill ebXML?

Because ebXML is a global standard that is not owned nor proprietary to any single
vendor.  Take a look at what is happening in the Web Services world.  Sure, all the
vendors have agreed to the core WS standards (eg. XML, SOAP, WSDL and
possibly UDDI as well)...but higher up in the protocol stack we have all sorts of
proprietary candidates.  For Process/Workflow/Choreography we have the W3C's
WSCL, IBM's WSFL, Microsoft's XLANG, BPMI's BPML and now the
Sun/BEA/Intalio/SAP consortium's WSCI along with the OASIS ebXML BPSS. 
Security includes many proposals from the likes of IBM/MS (WS-Security) versus
SAML. And the list goes on.

The vendors are trying their darnedest to do proprietary lock in at the higher
levels....many of these "proposed" standards come with Patent/Copyright gotchas
that could be leveraged by various vendors into ongoing revenue streams (mind you,
I wouldn'd mind 1 cent for each B2B/A2A transaction passed myself <grins>). We
had a recent example of that on this list not too long ago (fortunately the public outcry
"convinced" the vendor to do the "right thing" in that particula instance and
circumstances). I am not very confident that this will be the case on many of these
"new" proposals.

> Yeah, why would Sun want to kill ebXML?

Good question.....

> Hmmm. Maybe that's why they want to kill ebXML - it's not proprietary.

See above comments.
 
> Thanks for the alert,

de nada!

...Andrzej

Chaeron Corporation
http://www.chaeron.com

----------------------------------------------------------------
The ebxml-dev list is sponsored by OASIS.
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.ebxml.org/ob/adm.pl>