Bill Gates, Edd Dumbill, and the Semantic Web

		Print
		Email weblog link
		Discuss
		Blog this

Tim O'Reilly
Mar. 03, 2004 11:43 AM
Permalink

I recently sent around to my editors Steve Lohr's account of Bill Gates' recent stump speech, the NY Times article Microsoft, Amid Dwindling Interest, Talks Up Computing as a Career. (An archive copy is also available on Dave Farber's IP site.) For once, I found myself 100% in agreement with Gates. According to the article:

"In an effort to counter the trend [of declining computer science enrollments at universities], Mr. Gates, who personifies technological optimism and the potential payoff, sought to reassure students that their futures were no less bright in an era of outsourcing. The effect of computer technology, he told them, is just beginning and opportunity abounds....
Mr. Gates said electronic commerce had not yet even begun, and that huge gains in communication, convenience and productivity are on the near horizon. He acknowledged that there were challenges to be overcome in areas like privacy and computer security, skipping lightly over the fact that security flaws have bedeviled many Microsoft products. But even the headaches, he said, are merely intriguing problems for smart computer people to conquer, and profit from.

I am so there. As Jeff Bezos likes to say when exhorting his staff to innovate, "It's still day one." We're in the stone age of computing.

The NY Times article and Gates' comments are certainly worth a read. But what really triggered this blog entry was Edd Dumbill's response to my posting on the editors list. He quoted the last part of the article:

Mr. Gates scoffed at the notion, advanced by some, that the computer industry was a mature business of waning opportunity. In one question-and-answer session, a student asked if there could ever be another technology company as successful as Microsoft.
"If you invent a breakthrough in artificial intelligence, so machines can learn," Mr. Gates responded, "that is worth 10 Microsofts."

Edd (who is the editor of xml.com") used this quote as a jumping off point to champion the Semantic Web, and the vision of data interoperability that underlies it. I thought Edd's comments were too good to be read only on internal O'Reilly lists, so I'm sharing them here:

I find this an interesting thing to say. The traditional stance w.r.t AI is to scoff. The corollary of course is that as soon as something from AI becomes reality it ceases to be AI and that boundary is pushed back by definition.
I am fascinated in general by the attitude of software companies to AI-like topics, and in particular the semantic web. I won't attribute to cunning what can reasonably be attributed to narrow-minded stupidity, but there is a marked reluctance among big players to support the semantic web work at the W3C.
I'm finding it increasingly difficult to believe that it's because they think it's all going nowhere. I'm starting to think that it's the usual reaction against open standards: if we start exposing semantics on the web for all to see, then vendors lose another route to lock-in.
I agree with Gates that the opportunities in computing are only just beginning. The software we have now is so bone-achingly primitive and frustrating. I want to throw my computer out of the window. The number one requirement has to be a radical improvement in inter-application data interoperability: no more silos! The semantic web technologies meet this need directly, which is why I like them. On top of that can be built the user interfaces that meet my needs.
My hope for the future lies in making the data smarter. Vendors rely on making their software smarter. This seems to be the essential conflict.

This idea of making the data smarter is absolutely central. I have been speaking about this myself for some time. As we move to a network-based software platform, where applications don't live on the local machine but are distributed between rich client front ends and huge database back ends, "open source" alone won't really solve our problems. It's open data we're going to be fighting about.

Tim O'Reilly is the founder and CEO of O'Reilly Media, Inc., thought by many to be the best computer book publisher in the world, and an activist for open standards. O'Reilly Media also publishes online through the O'Reilly Network and hosts conferences on technology topics, including the O'Reilly Open Source Convention, the O'Reilly Emerging Technology Conference, and the Web 2.0 Conference. Tim's blog, the O'Reilly Radar "watches the alpha geeks" to determine emerging technology trends, and serves as a platform for advocacy about issues of importance to the technical community. For everything Tim, see tim.oreilly.com.

Comment on this weblog
You must be logged in to the O'Reilly Network to post a comment.

Showing messages 1 through 3 of 3.

Is it difficult to create/maintain/use ontologies?
2005-06-23 13:28:42 apsheth [Reply | View]

mchampion shared the concerns that earlier Peter Norvig also expressed in his posting at AlwaysOn.

Ontologies represent agreement. It become highly practical to develop useful ontologies
(a) depending the scope of the agreement (in increasingly more focused and targeted sphere, and hence more managable and feasible way: the whole world, common sense, the Web, discourse in a language, arbitrary topic that could occur in news, specific domain, specific task/application), and
(b) whether the agreement is capture as forma, semi-formal, and informal language/model (a viewpoint Tom Gruber drew my attention to).

Several highly succesful domain ontologies are already in extensive use, and for enterprise customers, more targeted populated ontologies are routinly developed (in our work at Semangix, to date, that has ranged from a million to 14 million instances). Those who still hold the view that it is not practicle to develop and maintain ontologies are either out of touch with avilable technologies and success stories in creating/applying ontologies or are tying to solve too big of a problem or a wrong problem for an ontology-driven solution. Ontology is not a pancea but for the right problem, it does wonders.

I have offered more extensive rebuttal to this point of view at:
http://lsdis.cs.uga.edu/~amit/blog/
Semantic Web: A different perspective on what works and what doesn't
. I would be happy to point anyone interested to several large open source ontologies.

Amit Sheth

No Latin & less Greek...
2004-03-06 11:32:15 laventurier [Reply | View]

Semantic standardization in the form of schema design/modeling isn't that hard, especially in key technical, medical and scientific fields with generally rich standardized and extensible ontologies; promoting the acceptance and integration of such standards may prove much more challenging. The idea that we're wallowing in a primordial soup of bits and that AI techniques hold the key to progress isn't quite right, though knowledge engineering/expert systems were a pragmatic approach to situated AI knowledge representation, and semantic web/document engineering/web services methodologies represent a similar, but standardized, data centric platform for the same kind of modeling, plus, crucially, networking.

What's needed is a deeper model or metaphor according to which semantic integration can proceed, not just more technical voodoo. One idea is that such generic emerging problems as identity/location management, medical/genetic privacy, vendor selection and quality control, JIT education would benefit from a central web service functioning as the back office for some sort of consumers' union or buyers' club enforcing the interests of its' members. (My dull stab at this is the not yet half-baked Utilitus Information Utility www.utilitus.org - *real* work along these lines may well be afoot - if anyone knows of any, please post!).

The idea that we are in the stone age of computing, while obvious in terms of applied materials science and the statistical truths a fully wired world will reveal, shouldn't divert anyones' attention or efforts from the ethical and political potentials of the moment. Let's give this principle the posh name 'teleocom'.

How do you turn "metacrap" into "smart data"?
2004-03-03 14:59:06 mchampion [Reply | View]

How does this smart data get created? Definitely not by Joe User and Betty Blogger, unless Cory Doctrow is a million miles off base. The typical answer is good meta-metadata created by text indexing/mining software and lots of hand-crafted, application-specific taxonomies that help the text mining software do its work. The problem is that It will be very hard work to build useful taxonomies / ontologies even after the Semantic Web standards and tools are mature.

Also, my guess is that the big companies are not being either cunning or stupid in their non-enthusiasm for the Semantic Web; they're saying that the standards for smart data are the least of the problems here. Smart data won't come out of the box or out of Word or Blogger, it will be created by software that we don't really know how to build yet. Lots of thought and money are going into this problem, by Google, IBM, Microsoft (I'm sure), and a zillion startups.

Showing messages 1 through 3 of 3.

Return to weblogs.oreilly.com.

Weblog authors are solely responsible for the content and accuracy of their weblogs, including opinions they express, and O'Reilly Media, Inc., disclaims any and all liabililty for that content, its accuracy, and opinions it may contain.

This work is licensed under a Creative Commons License.