It's snowing in Reno today. A false spring came in early April, bringing
blossoms to our cherry, peach and apple trees. The unmistakable signs
of life were everywhere...lizards, baby doves, young wild rabbits, families
of quail...one of the warmest early springs on record. But today it decided
to snow in earnest, which probably means that much of the new life that
volunteered to begin this spring won't make it. That's the way it is with
a false spring...
We got home from JavaOne late on
the Friday of that week. It was an awesome event...one of the best ever
by my reckoning. There weren't nearly the 20,000 geeks that had filled
the Moscone center to its capacity in prior years, due, I'm sure, to
a slumping global economy as well as a nod to Java's maturity -- but
that made it better. Less is more, and that can hold true with technical
conferences as well. The sessions I attended were terrific, the
vendors displaying wares on the pavilion floor were all high quality,
real technology players, and those attending were as committed as ever.
This year, the number of attendees and the number of vendors were both
at comfortable levels if personal space is any measure.
Liz and I drove to S.F. and back, enjoying the drive as much as the
week at the St. Francis, one of the coolest hotels in that part of
town. We managed to stay in a room in the old part of the building...
a wing that made it through the earthquake of 1906; we could live there.
Once we decided to buy an espresso machine for the room in order to
save money on what would otherwise be a pretty shocking daily Starbuck's fix,
the room became our new home, one that we could easily abide long term
if ever the opportunity arose. Stay at the St. Francis if you're
ever in S.F. Ask for a room in the old part of the building.
You'll be pleased.
I'm feeling vindicated already. IBM's bombshell and
Micro$oft's tacit approval (two monopolies!
what a coincidence!) has recently made noise about patent enforcement
on a communications protocol level. SOAP, you see, and ostensibly other
pieces of ebXML, are owned and constrained by patents held by one or both of
these firms. "Wait," you protest, "didn't both firms agree to not charge
for patent rights?" Not really...
Did they agree to a royalty free (RF) license when
offering these protocols to an otherwise unsuspecting industry?
No. Reasonable and non-discriminatory (RAND) licensing was what
they both argued to maintain....just in case, some day, they decide to
charge for use of their inventions? No...not really...more important by far
is the influence that simply having such a big stick engenders.
RAND vs. RF has been one of those contentious issues recently in vogue in
the Networked Distributed Computing industry. For years the Sun approach
of "agree on standards, compete on implementations," has been the rule of
thumb for the
W3C, which to date has set communication
protocol standards for the Internet. Since its inception, the Internet
has been made possible by widely agreed-upon standards (like TCP/IP)
for communication between systems. Hyperlinks, file transfers, message
passing...fundamentally all communication on the Internet has been
made possible by RF licensed protocols. But no more...
RAND protocol contribututions from a monopoly practically guarantee
undue influence over protocols otherwise RF. The threat
of a RAND license from either of these two firms tends to squealch
freedom of expression (i.e.; innovation). The interesting aspect
about IBM and ebXML is the influence over the standard that
IBM now owns. In order to proceed with the standard, IBM must
be satisfied, either via royalty payments or agreement upon "that which is
standardized," meaning that while IBM cannot ensure the success of
ebXML, it can now most certainly guarantee its failure.
It's an interesting strategy: enjoin in order to defeat.
The RAND approach allows a firm to control the industry. Imagine
Micro$oft as the "owner" of the basic protocols that drive communication.
That means that everyone...every vendor, every developer, every consumer...
everyone who would deign to use the public network would be beholding
to Micro$oft. The Redmond giant could charge whatever they choose
for access. And you'd have only one choice: either pay them, or you
don't play. Considering that the growth curve for eCommerce is
still pretty dramatic, despite the dot-com debacle, for one firm...
or a handful of firms...to be positioned as "Internet toll booths"
seems a little suspect.
Which brings us back to the Web Services nonsense I wrote about in
my last entry. The protocols in question
are enablers of Web Services. To the extent that M$ and IBM are
Web Services pioneers (defining the very notion of Web Services via
WSDL, UDDI and SOAP [WUS]), it should be clear that both firms fully
intended to embrace, extend and own the net via this misguided bucket of
WUS crap which simply doesn't stand up to critical analysis.
My prediction: either Web Services will fall, or IBM/M$ will utilize
it to bury the competition without delivering anything near the
claims made regarding its potential. I give it two years or less.
And Sun? Sun seems to be following the Pied Piper too, which is very
disconcerting to this old hack. There are still many at Sun who
are dubious about the WUS death march, still many who still don't
understand why on earth we're not all yet embracing JavaSpaces and Project Jxta
in earnest. As viable organic frameworks, perhaps the only ones on the horizon
with the potential to even begin to deal with the complexities to come,
one would think another approach might be obvious....alas. There are
still many at Sun who share this view...but we're in the minority, it seems.
JavaOne '02 was great because of the stealth Jini track and the Jxta
sessions and the BOFs. Most attendees chose WUS sessions, which in all
fairness were made better because of Java/XML interfaces and the JCP that
argues them. But I must say, in all honesty, even with the promise of Java
and the community of open-minded developers who suport it, to me Web Services
is very much like this false spring; the hope of new life crushed by the
very temptation to blossom early. And after such a harsh dot-com-debacle
winter too. Could be an unseasonable year...
If
there's hope for humanity it's in software