The White Heat of Marketing

August 15, 2001

Edd Dumbill

One of the wonders of our dot-com age is the degree to which subjects previously too esoteric to be mentioned outside of geek circles are now of general concern. Hearing my parents talk about how much RAM and megahertz a computer has seems novel and somewhat incongruous. Austere and venerable broadcasters, who once sneered at so much as a "w-w-w-dot", now spew URLs and urge us to share our views online. In a similar way, upcoming Internet technologies used to be the sole preserve of IETF mailing lists and conference conversations. Now we find that when a developer does so much as write ten lines of code, there's immediately a press release about this startling advance.

It's in this white heat of public fanfare and one-upmanship that the fabric of the Internet and XML are being developed today. Not so long ago "embrace and extend" was the Microsoft mantra. Nowadays every company is so busy embracing that the XML scene has started to look more like a perpetual group hug than anything else. One feels that if IBM or Microsoft were to propose something as ludicrous as a SOAP-enabled bathtub protocol, there'd be 20 endorsements and a new W3C XML Bathing Activity before anyone realized quite how daft the whole thing was.

At the moment few areas of XML interest exemplify this trend more fully than "web services." From SOAP's controversial start as a Microsoft brainchild, IBM and eventually Sun joined the party, and now all the big companies are busy "yapping" -- a new usage of a familiar verb I'm rather fond of, it means to create and promote "Yet Another Paradigm". It would have taken a miserable individual indeed to fail to rally to the call, as numerous XML conference keynotes promised us the automobile capable of rearranging opera tickets in response to traffic jams.

Following the standard YAP pattern, acronyms and consortia followed: DISCO, WSDL, UDDI. Suddenly we're in the middle of a new wave of technology, another set of features, and nobody really seems to have asked, to say nothing of answered sensibly, the question: "Uh, what do I need this for?" Perhaps some of the uneasiness is due to the fact that web services proponents have never demonstrated how to do any more than retrieve a stock quote. Hang that; I want to see the code for the opera-loving car!

Yet critics largely remain unheard. Much reasonable criticism was made about SOAP's reuse of port 80 to get through firewalls and, likewise, about the overloading of HTTP rather than the use of existing extension mechanisms. However, the real driving force isn't really technical excellence or even utility, but a mixture of commercial imperative and the breathless excitement of invention for its own sake.

On the other hand, we can't just wave our hands at web services and dismiss them. Believe me, I've tried. There are reasons for the popularity of the idea, some of which are deeply connected with the popularity of the Web itself. Everyone enjoys creating a home page, communicating and sharing with others. As software developers we have a deep need to see our software being used, communicating with others' programs. In whatever guise it takes, distributed computing over the Web is an important future direction.

Other groups have realized this too. Even though the initial technical solutions of web services may stick in the throat, it now stands for something larger than SOAP. For instance, the rather academic and esoteric RDF can be seen as part of web services; the implementation details may be markedly different, but the ends are the same: computation and information services distributed over the Web. So the area that is web services grows by accretion. I've even come across a proposal for WSUI, a Web Services User Interface language: surely a contradiction in terms.

What's being challenged here are our ideals about how XML should be driven forward. XML was borne out of an industry consortium, and there has been distinct emphasis on continuing its development in that way. We've seen many benefits of this approach, not least the new understanding that embracing open standards is not only non-harmful to a business but beneficial. However, the other understanding arising out of three years of consortium-led development is that inspired technologies rarely originate in committees. It's not news that the W3C's most successful XML technologies had a strong element of personal leadership and vision.

Today's W3C committees have much more of the appearance of political band-aid than leadership. The W3C's glory days as the site of arbitration leading to reconciliation are definitely over. The new model seems to be that the largest companies propose something, and everybody else rushes to agree by joining the relevant Working Group. Things are getting messy again. Now it's not just geeky mailing lists and conference corridors: XML technologies are well and truly out of the closet and a powerful marketing tool to boot. The word "standard" has just about lost its formal meaning.

Perhaps out of the intense heat and pressure a few diamonds will result to form the next round of true standards. More likely we'll forget about this year's current excitement as we're well into the buzzwords of next year. And if, as I suspect with much of the current web services trend, there's no "there" there, that might not be so bad.

And maybe one day, my car will love opera.

Fragments

I'm spending this week at the GCA's Extreme Markup conference in Montreal. An extremely technical conference, Extreme is packed out with veterans of SGML and XML. One of the pleasures of this particular group of people is their wit, humor, and downright eccentricity, not to mention a deep affinity for good restaurants. This week's fragments take the form of tidbits from the conference.

Topic Maps and RDF buffs have come in for a healthy dose of leg-pulling and teasing. The opening keynote characterized these technologies among the "cloud dragons" to which we pray. But analyzing these technologies more carefully, Jonathan Robie presented an excellent talk showing that the upcoming XML Query language was capable of serving as a query implementation for both RDF and Topic Maps.

The ad-hoc posters have proved an unexpected delight. My personal favorites include a ditty by Syd Bauman about the supremacy of Topic Maps over RDF, and a selection of extracts from the W3C XSL-FO specification rendered as beautiful calligraphs. The extracts themselves were examples of the XSL-FO spec's brutal contempt for the English language and drawn by Liam Quin, the W3C's new XML tsar. Quin accompanied his illustrations with a strong statement that specifications should be written in plain language. Given his new post at the W3C, we might dare to raise our hopes.