I spent Wednesday and Thursday at Chris Sells’ fifth XML Devcon. This is a high-level gathering of the .NET XML (and thus Web-Services) community. It’s being blogged to the max (nicely aggregated on the conference site), and there’s an eWeek person here journalizing in realtime. It’s been fun and educational.
Content · Here’s how I’d summarize: First, you have the talks from the heavy Microsoft insiders (Box, Anderson, Purdy, Rajpal), then wild-cards like me and Sam Ruby arm-waving about generalities, then application stories from the real world.
The Microsoft people’s content was pretty well all superb, although some tried way too hard to be funny. The application stories were all over the map; some were excellent, for example Patrick Cauldwell and Scott Hanselman on the serious magic they’re doing at Corillian; and I’ve already plugged Jeff Barr’s piece. A couple were, frankly, lame, and would never have made it onto the program at the Big XML Conference. But on average the content was good; and if you need to be on top of what Microsoft thinks about Web Services, you need to be there.
Hey, the computers here aren’t all Macs.
Attitude Problems · There were three amusingly distinct sets of attitudes on display. The people who are way down in the infrastructure trenches—like Don Box, Sam Ruby, and me—tended to be fairly withering in their scorn for the technologies that are part of the problem not part of the solution. The end-users, like the Corillian guys, who are actually out interfacing with the end-users building things, were a bit scarred but saying “see, we can make it work”. Then there were one or two representatives from the marketing community, the people that would be in the vast majority at a normal conference. For example, Rebecca Dias, while charming and a good listener, was really pretty hard-core: “Everything’s basically just fine and I can show you hundreds of success stories!”
Warm Glow · Damn, was it ever friendly. At the kinds of conferences where I normally hang out—XML, Java—there’s an air of competitive tension in the air, and when there’s pain out there in the economy, you can feel it.
At this event, there was approximately zero visible organizational stress. One reason is that the hospitality (food, drink, logistics) was lavish, and that Chris Sells, our host, is a charming, welcoming, amusing guy, and a good chair. Dropping into cynical mode, I might observe that being a monopoly makes for less of that nasty competitive strain, and also means there’s more money to throw around on showing the loyal retainers a good time.
Chris Sells and Don Box (photo by Chris’ son).
But it was a good time, there was even a musical performance: Give SOAP a Chance, a twisted retake on Give Peace a Chance with witty raps between the verses.
And I have to say I was touched, they made a big fuss over me. My helping-build-XML phase was so long ago that it rarely crosses my mind now, but they organized a funny, tasteful little salute where all the Microsoft guys put on hats in the style I’m often seen wearing. Some of them looked quite OK that way. Thanks, guys.
SOAP · Speaking of SOAP, I have tended to be pretty negative because it shared a niche in my mind with the teetering tower of WS-fantasy. I’d always kind of known that, in theory at least, SOAP didn’t depend on that stuff, but I’m now on my way to being convinced that you might be able to get some good practical use out of it without drinking the WS-Kool-aid. Hmmm.
XSD, WSDL, Blecch · That EWeek article linked above highlights Don Box’s comment that if the WS-* project dies, he’d expect to find the fingerprints of XML Schema and WSDL on the murder weapon.
I had said unkind things about W3C XML Schema in my keynote, with some trepidation because many aspects of WS-* seem to be pretty tightly coupled to an XSD-centric world-view. Imagine my surprise when I discovered that with a couple of exceptions, the gang at the conference shared my distaste. Those exceptions aside, there were two general opinions about XSD. One was “It’s a bleeding sore and we should fix it!”, the other “OK, there are problems but we can work around them and it’s too late to go back now.”
WSDL is even more tragic.
XSD’s ugliness mostly hides in the plumbing where civilians don’t have to go
near it, but WSDL is in the customer’s face; as Nelson Minar of Google once
told me, “The WSDL should be the .h file for my Web-services
application.”
Indeed, but WSDL as it stands is not something that wins friends and
influences people.
Even if it’s too late for WS-* to untwine itself from XSD, maybe there’s
still a hope for finding a better way to publish service interfaces?
