Development Model for Services

Today we are very happy to participate in the announcement of the Service Component Architecture (SCA).
I among many others have written on the subject of contract first development for services, and the importance of being able to design first and deploy later. SCA defines a programming and deployment model to directly address this requirement.
Our interest in this is related to the language independent nature of SCA (we provide Java and C++ implementations of our runtime), in the ability of our customers to obtain the “right” kind and size of infrastructure for SOA, and in the ability of tools to map easily and correctly to multiple SOA runtimes.
IONA has a unique, industry-leading microkernel-based, plugabble runtime specifically designed for SOA based on Web services. WSDL is the design metaphor we support, and through configuration changes we support WSDL mapping to multiple communication transport protocols and data formats.
We are developing an open source version, and leading a tools project at Eclipse. So we are very pleased to be part of SCA, since it’s annotation-based approach promises to help bridge the independent tooling and heterogenous runtime worlds.
Until now, other vendors have tended to focus on adapting existing runtimes, such as J2EE and EAI architectures, rather than developing new infrastructure from the ground up for SOA. Now this is changing, and the good news is that it is changing through industry cooperation.
I expect significant improvements for developers and SOA customers since SCA is aimed at making it easier to concentrate on the service definition rather than on the coding it often takes to map a service description to executable code.
To put this another way, I see a lot of criticism today of WSDL and Web services tools because they tend to be too tightly tied to existing object models and languages. SCA takes us in a better direction, providing designs, metadata, and configuration based deployment specifically for SOA infrastructure.

Robert Wyatt Review

Entertainment Weekly posted a review of the new Robert Wyatt album. They are right that this for Robert Wyatt fans, but if you are then it is great. They rightly call Rock Bottom his masterpiece, and I can’t believe they actually performed some of the songs live, since they seem so thoroughly studio tracks. Robert fills in for some of the studio sounds using his voice.
Taken together with the Alan Moore interview October 28 this explains a lot about why I just renewed my subscription. They cover a broad range of movies, TV, CDs, DVDs, and books, and their reviews are well done and reasonably impartial for a magazine with advertising.
Unfortunately their Web site is restricted if you are not a subscriber, but it is great if, like me, you are a subscriber, since extended Web site access is free.
This is unfortunately in contrast to the Consumer’s Union Website, where they ask you to in effect doubly subscribe, once to Consumer Reports print and again to the extended content on the Website.
As much as I’m a fan of Entertainment Weekly, I am a real Consumer Reports junkie. Because you just never know when you might have to buy something … I read every issue pretty much cover to cover, and have continuously subscribed for 15 years or so.
I know they don’t take any advertising, and you get real unbiased product evaluations (no entertainment reviews though), but it strikes me that they don’t really get this Web thing – if you subscribe to their content it should be one price for both, not one payment for online and another for print. I think they would actually encourage more subscribers that way, and gain rather than lose revenue (I imagine a concern over losing revenue must be the reason they do things this way).

On TV (well, SYS-CON TV)

With traditional print advertising revenues being challenged more and more by the Web, media companies are choosing between cost cutting and investing in better content. Under the leadership of group publisher Jeremy Geelan, Sys-Con is, despite their annoying pop-ups, one of the traditional print companies investing in new ways of attracting “readers.”
Last year while cutting through the Web Services Edge East exhibit hall on the way to deliver my session on SOA best practices (based on this article), I ran into Jeremy, who grabbed me for a quick, unrehearsed interview on his new TV show. I met Jeremy at the first Web Services Edge East conference back in 2002, and had appeared previously on Sys-Con “Radio” a couple of times (broadcast to the exhibit floor) but this was something new.
Can you tell that they told me to look at the camera and not at Jeremy? I was so intent on it that I forgot to look at Jeremy at all. I hope I didn’t make the same mistake last Friday, while taping two panels and a one-on-one for Sys-Con TV‘s new series out of the Reuters Studio in Times Square. You’ll have to let me know, after they come out, of course (I’ll post again when they become available).
The first panel was called “Thinking in Services” – which Jeremy said was inspired by my blog entry on the subject. It’s important to think about the design first, and the technology later. After all, SOA is more about a change in thinking than technology.

The “Thinking in Services” Panel Afterward – David Kershaw – Professional Services Manager, Professional & Education Services, Altova, Aaron White – Software Architect MindReef, me, and Paul Lipton – Senior Architect, Web Services and Application Management Team, CA.
Jeremy posed questions about services and SOA, and we ended up with a good, lively debate among the five of us. At one point Jeremy referred to me as a kind of “grandfather” of software. I think this was because of Aaron, who is 24 – close to my son’s age in fact. One thing clear is that the current generation of Web services tools are not very well geared to the new thinking. Products like Artix however provide a good foundation for an SOA since they focus on creating the standards based services layer you need to get started.
(Paul got the award for looking the best on TV, by the way.)
Before the taping we took turns getting made up during the live broadcast of the App Server Shootout, the big event of the day. This is certainly a tough and very competitive area of the market right now. One interesting thing this time was the introduction of price as a comparison factor.
At the end of the day I decided to wear my makeup all the way home to see if my wife would notice. She didn’t . (But that is, at least what they told me, the mark of a good makeup job.)

The Makup Artist
After thinking in servces Jeremy invited me to join the taping of a panel called “Enterprise Open Source” since IONA is in the enterprise open source market with Celtix. We spent a good bit of time talking about the relationship of open source to commercial products, with the conclusion that the world is a “blended” one for now. Which is great for us, since unlike JBoss or EnterpriseDB, we can comprehensively support both the open source and commercial versions of our ESB.

The Enterprise Open Source Panel Getting Ready for Taping
Andy Astor, CEO and co-founder of Enterprise DB is on the left and Shaun Connolly, Vice President of Product Management, JBoss Inc. is on the right. Yakov Fain, a Sys-Con columnist, joined later. Jeremy is in front of the studio desk. The guy standing to the left is one of the studio technicians. Through the screens in back of the desk you can just make out the Times Square view.
And then it was time for the one-on-one. Jeremy kindly let me go first since he knew I had a plan to catch. A momentary delay while the makeup artist was summoned to apply some spray to some disoriented hair…
I hope this turns out to be the right kind of content for Sys-Con to invest in. Meaning that it is helpful, informative, and useful to the readers and viewers. All I know is that by the end of the one-on-one with Jeremy I could not believe that 7 minutes had flown by so quickly. And then I was in Times Square with my rollaboard, fighting the crowds for a taxi at rush hour amid the flash and hustle of one of the world’s busiest corners.

Successful WS-TX TC Launch

Meeting hosts Hitachi Software ensured that the initial WS-TX TC meeting was kicked off in style, with the Taiko drummers from San Jose playing at the reception last Wednesday night.

Taiko Drummers of San Jose
And a great buffet at the evening reception after the first day. This has definitely set the bar high for TC hosts. Of course this was a special occasion since it was an inaugural event, but few, if any of us could remember the last time a TC was kicked off in such style. I think we voted to thank Hitachi at least three or four times (and of course we would like all votes to be so easy .

One of the Food Tables at the Reception Buffet
We also got through the agenda in a day and a half, which was about a half day faster than Ian and I had anticipated. We recorded 47 participants, who heard presentations on WS-C, WS-AT, and WS-BA to provide background on the contributed works. We also had a presentation by Alistair Green of Choreology for background on the issues they’ve submitted.
Overall the specifications appear to be in good shape. With any luck we should be able to achieve our goal of progressing them to OASIS standards within the coming year. If the initial impression is to be trusted, this may be one of the few areas of WS-* in which few contentious issues exist. Then again, we are talking about an area of the industry that hasn’t changed fundamentally during the past 20 years.
Paul Knight did a great job on the meeting minutes, and Dug Davis’ chat tool was very popular.
One of the hardest tasks at a meeting like this is to gain consensus on a regular meeting time (since so many participate in multiple TCs and WGs) and on dates for upcoming face to face meetings.
Using a great “two phase” process, Ian first polled the group about which times and days the members “could live with” and then which times and days the members “could not live with.”
The winning times and days produced by this method are alternate Thursdays at 11:30 eastern (starting Dec. 1), March 14-15 at Raleigh (hosted by IBM) and May 30- June 1 at Dublin, Ireland (hosted by IONA).

WS-TX TC Underway

Hello from sunny Cupertino (it’s raining in Boston), where Hitachi is hosting the initial meeting of the OASIS WS-TX technical committee.
Ian and I were approved as co-chairs this morning, and at the moment we are going through the background on the submitted specifications.
The charter, which we reviewed this morning with minor clarification, has us completing an OASIS committee specification draft of WS-Coordination within six months, and committee drafts of WS-AT and WS-BA within a year.
The meeting is going well so far, and we have about 28 people in the room and several more on the phone.
I’m really looking forward to progressing these specifications to OASIS standards during the next year. As I have said before, this represents the convergence of my interests in TP and Web services.
Last night I was talking with some of the TC members and I realized that I’ve been working in TP for 27 years now…starting in 1978 with a system that used an HP 3000 for the back end COBOL “handlers” that accessed the database and an HP 9000 to drive the Burroughs poll select block mode terminal protocol.
In those days we were very excited about transforming batch systems to online systems – some of the younger folks commented that they are just as happy not to have been around then.
But this was what got me started – seeing the benefits of immediately querying the database after an update rather than waiting overnight to get the results, or to find out that an error occured.
In the Web services area my main interest is in software standardization – the benefits to be achieved by turning what is essentially still a craft industry into a more scientific or industrial industry.
I was talking with someone about this on the plane out here last night, someone not in the computer industry. And to the question of how did we figure out that we’ve built the right products with the right features at the right price in comparison to what other vendors do…basically today that tends to be a judgement call. How do you evaluate the quality of a piece of pottery, quilt, or sweater at a craft show?
Today with software it’s the same question, but we need to move toward standards that will allow us to better compare products against each other.

Microsoft Memos

Well, everyone else is talking about this today – might as well chime in!
I just finished the Ozzie memo. I had read the Gates email yesterday. I thought the email was right on target. It is a clear recognition of a significant new industry trend and a very important recognition of the need to change and adapt yet again. But I wonder whether it will be possible this time. They have spent the past few years institutionalizing their very successful culture, and this change represents much more than a change in technology direction – it is a fundamental business model and culture change.
I do not see evidence in Ozzie’s memo of any specific plan of attack. The description of the problem seems very good, but there’s nothing about a solution, or how exactly he expects Microsoft folks to go about implementing the proposed changes. He’s asking for a study, for each division to respond with a proposal to solve the problems he describes, and he has created a process within which he will be assigning people to work on solving the problems. But again, nothing concrete about how it’s going to happen, or what it will mean in terms of products and technologies. We all know the problem – what we are looking for is what Microsoft will do about it, and there’s nothing here about that.
Of course, the email and memo were very likely designed to be released to the public, which is pretty clever if so since a large part of the change they’re talking about involves innovation “in the open” as Google has been doing, for example. They have been releasing their works in progress publicly, as does the open source community. Traditionally, Microsoft’s innovation is performed behind closed doors, and even if you sign an NDA you may still not get access to everything they’re doing or thinking. So “leaking” the memo and email would be a clever move in this direction, even though they don’t really contain any specific details about their plans.
Unlike the earlier efforts they cite as precedent, such as the famous Internet memo and the “bet the company” on .NET and XML/Web services, there’s no concrete action here other than a call to work on proposals and processes for solving the problem. No solutions are proposed.
Furthermore this comes at a time where Microsoft is heads down working to fulfill its commitments and promises around Vista, Office 12, Communications Framework, Workflow Foundation, new SQL Server, new Visual Studio, etc. These are all scheduled to ship around this time next year, and I can’t imagine everything is ready. I have more of a picture of the folks in Redmond sweating out the final days of testing, bug fixing, etc. in a big rush toward the release date, as always happens in large software projects, rather than sitting around contemplating solutions to new problems.
Bill Gates says he wants quick action but Ozzie says this initiative is intended to start after this next generation of products ships and people will start to free up. But this means effectively nearly a year before the company can really start working on this initiative.
Beyond the technical challenges, which amount to turning around in mid-stream, this also represents a cultural challenge since Microsoft has built up its entire business around the shrinkwrapped license. A recent Gartner survey showed that Office 2003’s biggest competitor is Office 2000. The article about upgrading to Office 2003 shows that majority of users are still on Office 2000. In fact a higher percentage are on Office ’97 than on Office 2003. And now we are talking about introducing Office 12, and working on “services” at the same time, or after the release of Office 12 (when staff will be freed up), when customers are still not moving to Office 2003.
In other words, Microsoft’s fundamental business model, that drives all the cash flow, is under threat because people are not upgrading – people do not seem to need new features in Office or Windows, preferring what they have – while at the same time they are trying to change the business model of how they make money on software by turning to advertising. They do not have the proposed solution for that – just the problem statement – and they are still struggling to execute their current plans.
Finally, I think the recent reorg into divisions will work against this proposed change to “services.” Ozzie is asking each of the divisions to come up with plans and proposals to address the problems he outlines but they are as likely to compete with each other as cooperate, as he points out they are already doing in several areas.

WWW2006 Server Fire

An unfortunate and serious fire has more or less destroyed the computer lab at the University of Southampton, UK, which was hosting the www2006.org website and mailing lists.
The conference organizers have set up temporary page to allow papers to be submitted.
The deadline for submitting papers has been postponed till November 11.
The conference organizers have asked me to post the following message: