CNET también está disponible en español.

Ir a español

Don't show this again

Tech Industry

Breaking the logjam on Web services

IBM's Bob Sutor says remaining squabbles about royalties and intellectual property don't threaten the progress of Web services. Still, disagreements with standards bodies linger.

When Bob Sutor hears about squabbles over definitions for key Web services standards, he dismisses them as little more than passing headlines. "These things will play out," says Sutor, who directs IBM's Web services strategy and is a key figure in the Web services standards process.

There may be more than a hint of optimism in his upbeat pronouncements. Indeed, disagreements still remain, and IBM and Microsoft have variously clashed with other companies over their standards submissions.

But Sutor is nonetheless convinced that Web services are no longer in danger of getting sidetracked by the internecine struggles that seem to be a concomitant part of the standards setting process.

Web services refers to a technology approach in which Web-based network applications from one department or company can interact with Internet applications from another department or company via the use of open standards. These connections are seen as more flexible and less expensive than older electronic data interchange networks, which could only handle the exchange of specific transaction data.

Sutor recently spoke with CNET News.com about the standards battle and the lingering obstacles to greater corporate adoption of Web services.

Q: Within the World Wide Web Consortium (W3C), it sounds as if some members are in favor of collecting royalties while others are not. Is there danger of a stalemate?
A: These things will play out. I think that the talking up around the whole royalty-free aspect has been a red herring for many people. Most of these things are royalty-free already. We say it's royalty-free and people keep saying: "And it should be royalty-free." So it's a topic of discussion that receives a lot of attention.

Has it been overblown?
Yes, it has been overblown in the sense that the specs we've been putting out like WS Security or SOAP (Simple Object Access Protocol) are royalty-free, so what more do you want? Do you want us to pay you money beyond it being royalty-free? There's a limit there. So I think it's become part of the political side of standards rather than a real issue.

Whenever you hear of more than one standard in any one particular area, people are screaming, "Why can't these people figure out what they're doing?" It takes time.
Is the real question this: That the smaller--and thus less influential--members want the process to be entirely vendor-neutral while it's virtually a given that bigger companies will have more influence?
In some sense that's always been true. It was true with the Web itself. Part of it certainly reflects experience. If you talk about enterprise software, for instance, we've been in it for decades. Odds are that we're going to have fairly strong opinions about what we think is the right way of doing things. The odds are that if you ask a WebMethods, they're going to have strong opinions about this as well. The same goes for an Iona.

The fact is that everything gets into these standards groups eventually. SOAP, which has been in the W3C for two years, has been shifted a little left and a little right, and I don't think anyone's going to claim that we've had more influence than anyone else. We chaired the thing, that's true. But if you talk to the folks, they would say we've been pretty darned neutral about this.

But wouldn't it be true that IBM has more say about the future of SOAP than software developer Dave Winer even though he had a big hand in shepherding the original spec?
In terms of where it gets standardized?

Yes.
We had to do the submission to the W3C but Dave went along. If he had raised a stink and said, "Hell, no"--he was one of the stakeholders and that would have stopped it. I think people don't necessarily get the idea that what we're doing with Web services is part of one big program. It's not 4,000 individual random specs that keep showing up. There's a program of work here; we're not just randomly picking odds and ends here. There's a road map for what we have to do. So we end up with a set of standards that work well together well and functionally handle anything we want them to do. It's not just this strange, open-ended process

Are intellectual property issues likely to keep cropping up?
There have always been IP issues since people began doing standardization. It's much easier if you and I are sitting down at the same table and we can decide what the intellectual property rules are. But what if someone down the block who we've never heard of then says there are intellectual property issues--that's why a lot of this isn't so cut and dried. The people who are at the table can come up with the decisions and decide they're going to do it royalty-free and that's great, but it's these hidden types of things that are out there. But we keep chugging along.

Your feeling is that things are nonetheless progressing?
I think they are. There's a lifecycle to doing standards. Something gets announced and certain people jump up and down--typically the authors--saying this is the greatest thing ever and the world is saved. But then certain other people take offense.

We're almost at year three in the standardization process, and I think it's going to take another two years to finish up a lot of the other things.
Because squabbling is an unavoidable part of the process?
For whatever reason--they don't like it technically or they don't like it because those other people are getting the attention or just because they are feeling cranky that particular day. Whatever it is. So you go through this period where there's this positioning back and forth. Typically what happens is there's a lull for a few months and then, lo and behold, something gets started in the standards organization including those people who were complaining on day one.

What are the chances that a squabble over the formulation of future standards could short-circuit the progress of Web services in the corporate world?
I think there's too much momentum now. Whenever you hear of more than one standard in any one particular area, people are screaming: "Why can't these people figure out what they're doing?" It takes time.

But is there any one hot-button issue that people can't agree on?
The one spot where people are still facing off is on choreography and that really goes to the politics of the standards organizations themselves. But the momentum for standards for Web services has clearly shifted to OASIS (Organization for the Advancement of Structured Information Standards). The W3C has decided that it wants to start a group around Web services choreography; IBM doesn't necessarily think that's the right choice or right venue, but we'll see how it works out.

Will IBM abide by the decision?
In what sense?

If they decide on something that's not to IBM's liking, will you be good soldiers and abide by the decision of the group?
Who's the group though? The W3C staff or the 500 member companies? That's where it gets little difficult because if you look at the models of the W3C and OASIS, they are very different. The W3C has a very strong centralized model of control; OASIS is far more decentralized so it doesn't necessarily make any difference what the OASIS staff feels about standards. They are more focused on providing the environment for people to get together. So you have this tension between them.

Is Web services adoption proceeding at the same pace?
We're seeing a lot in Europe, though just the beginning in Asia. So far, most of what you've seen are big vendor pushes rather than a broad community-based movement in Asia. Europe is right up there. They shifted to XML (Extensible Markup Language) a little later, but they learned a lot faster and picked these things up.

Web services conferences have been around for at least two years, but why haven't chief information officers more widely embraced the technology?
We're almost at year three in the standardization process, and I think it's going to take another two years to finish up a lot of the other things. But we're now talking with people in the mainstream, not the early adopters, and they're asking how they can talk about the business value of this to their boss. That's important because while we've done good job of talking to developers, the one big audience we haven't gotten to really well is the CIO.

If you can't quickly explain the return on investment, you might as well save the conversation. Can you today make that return-on-investment argument for Web services?
We can but not to that level of specificity. We can go into any industry and say this is what your friends and competitors are doing, but it's still very difficult to measure specific ROI.

OK, but in this kind of economic environment, that won't really make it, does it?
It's not concrete enough, but the problem is that you're talking about such a broad integration technology that you've got to go in and analyze particular companies and their particular set-ups. Are they merely using Web services for something they've done before or is it something brand new?