Business Value of SOA

Last week I had the honor of being the invited speaker to a large corporation’s quarterly architecture board meeting, sponsored by the CIO.
I was asked to provide some perspective on why SOA is so important for business.
The trends toward service orientation and service oriented architecture are motivated largely because of their promises to solve important IT problems, such as developer productivity, sharing data across functions and organizations, and streamlining processes and procedures.
Because the software industry is typically so full of hype and exaggeration, I often get a comment along the lines of “well, Web services and SOA are just the latest fad – in another five or ten years something else will come along to replace them.”
To me this is like saying the World Wide Web – from which Web services gets half its name – is going to go away any day now since it was the latest fad ten years ago. I see URLs all over the place, on TV ads, billboards, magazine ads, and even on products such as shower heads, tools, and building supplies.
If Web services can fulfill their promise, and bring Web-levels of interoperability and platform transparency to the IT environment, and if companies invest in an SOA based upon them, they will not go away. They will become as permanent a part of IT infrastructure as the Web is of publishing and consumer culture.
What threatens this promise has nothing to do with technology, but rather with intellectual property. The success of an SOA based on Web services depends upon freely available standards, unencumbered by patent claims and licensing fees. As I think I have mentioned before, Tim Berners-Lee did not seek patents for the inventions that created the Web.
Over time the applications based on the SOA may change, but once firmly established the SOA will be a key part of all future IT solutions, since it enables the production of new applications with greater effeciency (i.e. higher reuse) and predictability (i.e. standard interfaces to features and functions).

This Week – XML at W3C TP

Unusually this week I didn’t travel anywhere. Although to be fair I did drive in to the Hyatt at Logan Airport for the W3C Technical Plenary day. It might have felt more like going somewhere except that the trip was so familiar.
At the TP (this always confuses me since I spent more than 20 years in transaction processing, but then I remember that this is not the provenance of the W3C) the hot item was a panel discussion on the future of XML.
These events are sometimes more memorable for the people you meet outside the formal sessions. For example, just after lunch I caught up with my old friend Jim Melton, who was on the XML panel representing of the XML Query working group. It must have been 10 or 12 years…
The controversy over XML seems to boil down to whether or not the language is done. Standards committees have a tendency to look for additional work when their original charters expire, whether it’s really needed or not. This point was well made during the session.
Someone even got up during the Q&A to propose recinding XML 1.1, since it hasn’t caught on widely (see Noah’s talk for some details). The problem seems to be that XML 1.1 isn’t compatible enough with XML 1.0 – XML 1.0 parsers can’t handle XML 1.1 without significant modification.
Another interesting point of debate was binary XML. The main question seems to be whether binary XML should be a straightforward transformation of text XML, or whether a binary XML format should include structural information within it (and therefore be “parsable” in its own right). Everyone seemed to agree that a WG and a standard are necessary, since a dozen or more proprietary implementations already exist, but what that standard should look like remains hotly debated.
I was very glad to hear that our metadata proposal (as amended by Microsoft) passed in the WS-Addressing WG face to face earlier in the week. Everyone has talked about Web services supporting multiple transports from the beginning of SOAP, and in fact this is one of the main justifications for why WS-Addressing is necessary in the first place (i.e. Web services shouldn’t have to rely on any transport-specific addressing mechanism if they are to be truly transport neutral). But until this change was approved, the Addressing specification did not explicitly support the multi-protocol feature. My colleage Steve Vinoski has written extensively and well about this in his blog, and in fact led the effort.
With the metadata extension to WS-Addressing endpoint references in place, it will be possible to publish the same service over multiple transports, and provide failover and availability qualities for a service.
What I’m worried about now is getting the WS-Policy framework submitted to W3C. We need to progress policy in an open forum, the way WS-Addressing is being progressed, since metadata definition and management are so important. No one seemed to know for sure what the holdup is.
As a sad footnote to the week, I learned that my old friend Frank Willison passed away nearly four years ago. Ironically, I learned it in re-establishing contact with another old friend, Jeff Mandell. I spent the rest of the week filled with memories of those great days at Digital, from 1984 to 1990 or so when we were all just kind of starting out and Digital was doing great. We shared a lot of laughs, music, and (embarrassing as it is to admit) foreign cigarets in the hall… Like the tributes all say, Frank was a wonderful human being, and a great writer. We used to tell him to quit the day job and give Dave Barry a run for his money. Check out this excerpt from the Best of Frank:
“Partway through Elliotte Rusty Harold’s talk about namespaces, I realized where this relentless drive toward abstraction was taking us. Every new level of abstraction draws the computer-based world closer to the concepts we talk about in the real world. We’ve moved from waves to bits to data to information to infosets to application objects. As this process continues, some ambitious Comp Sci graduate student will realize that somebody already created the tree structure mapping the highest level of reality. That person was, of course, G. W. F. Hegel. Hegel’s dialectic led him to create a map of reality that, at the top of the tree structure, divided everything into either the material or the spiritual realm. That dichotomy was resolved in God, and, my friends, that’s about as far as you can go.
“That ambitious Comp Sci grad student, eager to get his Ph.D. and begin making real money, will create The Two Final Infosets: MatterML and SpiritML. Then, late one night, as rain falls in torrents and lightning flashes outside his laboratory windows, he’ll run XSLT to transform the material world to the spiritual world. We’ll be gone. The last material object on earth will be that graduate student’s open copy of XML in a Nutshell. It makes an editor in chief proud, in a perverse kind of way.”
It seems clear from this that the Semantic Web folks are barking up the wrong tree…
I can’t believe I lost touch so completely that I didn’t even know he was gone.
The lesson is to keep up better with friends. And so was it great to see Jim again after all these years.