Wednesday, September 12, 2007

More obvious misgivings about Microsoft and SOA

InfoWorld blogger and MuleSource CEO Dave Rosenberg has some thoughts on Microsoft and SOA in light of, and in advance of, the recent spate of BizTalk announcements and partnerships. Like myself, he takes exception to Microsoft's claims of SOA support and affinity.

Here are some excerpts from an interview Rosenberg did with ITBusinessEdge:
Question: So. You're among those who feel Microsoft doesn't "get" SOA? Why do you say so? Do you think they don't get it -- or don't want to get it?

Rosenberg: I would say that not only does Microsoft not get SOA, they purposely are trying to usurp the whole concept for their nefarious doings. Microsoft offers nothing in the way of architectural development tools or infrastructure that supports the ideas behind an SOA. The Microsoft architecture of .NET is not designed to be service-enabled and when you try to service-enabled it yourself -- when you try to build the infrastructure so you can take advantage of reuse, you find yourself in a very customized system having defeated the purpose.

If you look at Microsoft infrastructure, it's all about trying to lock people into their hegemony and SOA is all about giving control to the user.

Question: Did you see Microsoft's 100-something page thesis on SOA, titled "SOA in the Real World?"

Rosenberg: I can't tell you I've sit there and poured over every word. It's not unlike their "Get the Facts" anti-Linux campaign -- it's simultaneously interesting and full of crap at the same time.

What's interesting to me about Microsoft's approach is the obvious thing to do with SOA is to say, "Of course we have a strategy -- here's what you do now and here's what we'll do in the future." What should've been very easy for them to say, "Yes, we'll be a part of this and we want to start to think about our way of doing [SOA]" -- that would be acceptable. Instead they take this bizarre approach.

There's no clear answer from Microsoft on what their vision for SOA is or how their products or things you'd buy from them would participate in a SOA. For that matter, there's nothing from Microsoft that would say to someone, "I should use these products for my SOA."

All of the sudden .NET went from being a language, to an application framework, to being a "Windows platform" again.

Developers are quick to shun things that they don't trust and I think Microsoft sets the tone for how their development community thinks about larger scale concepts and so far they've succeeded in making SOA confusing.

Question: So, is Microsoft's talk about SOA a barrier to its acceptance among developers?

Rosenberg: From what I can tell, they're not doing themselves any favors. The whole sort of wait-and-see approach is not great. What's interesting is developers do eat that Microsoft dogfood pretty fiercely. They wait for Microsoft before they make choices.

It's a challenge for architects in terms of being flexible and having agility in their work. If you look at IBM or BEA, it's very clear what their vision of SOA is. With Microsoft it's just not clear.

I think some of that is related to the fact that .NET was not built to be a SOA or a services platform. They don't want a heterogeneous environment, they want the entire Microsoft suite everywhere. For instance, a couple of months ago they were calling it service-oriented infrastructure. What is the rationale with not going with the industry standard terms?

Before I started this company [MuleSource], I was the CIO of financial services firm. Initially we had an all LAMP and Java infrastructure. Prior to my getting there, they outsourced the development of a .NET application. We went from having a highly scalable, flexible infrastructure that was able to consume data and services across the enterprise and added a .NET application into the infrastructure -- it was a complete and utter silo.

It was built around this framework that was not meant to go in a services direction, and we basically had to adjust everything else in the enterprise for this one application that couldn't align itself with the rest of the business.

And that is the unfortunate side of what .NET has done to development. Conceptually, Java guys are more likely to understand the SOA model than the .NET guys are -- developers are not trained to think about other applications in .NET. They aren't trained to think about wanting to consume or expose a service to another application. That has so far been a barrier to using .NET as a platform to do SOA.

For better or worse, this development style is what .NET developers live and breathe. If Microsoft doesn't have a component, it doesn't exist. But in the real world, you still need to solve that problem.

How can a company that in-the-know be so clueless about as an important concept as this? This is the world's largest technology company and it's proven to be completely impotent and useless with SOA so far.

Question: So, do you think this is a sort of marketing ploy by Microsoft or just because they didn't anticipate that SOA would take off?

Rosenberg: I think it's a bit of both. What I'd tell you is that they should embrace SOA. Open source, you can see why they don't embrace it -- but SOA, they should say it's awesome and encourage it. It doesn't make a lot of sense.

In the real world, we see people who have similar situations to the one I described, with one or two .NET applications that they want to get on the [enterprise service bus (ESB)]. Does Microsoft give you an obvious answer? Kind of, but not really.
My take is that inside of Microsoft its aggressor A-types are all about dissing SOA and promoting .NET ad nauseam. At the same time the Microserfs and developers must understand the inevitability of SOA for at last a portion of the most advanced and innovative enterprises' and service providers' architectures.

And so, as the world turns toward SOA, Microsoft will fight quietly inside of itself about what it really is as a company -- a partner to its customers, or a parasite on the hide of productivity.

Ultimately the marketplace will determine Microsoft's end-game role. If there's an advantage to SOA for those that embrace it broadly and effectively -- and I believe without question that there is -- then there will be a penalty for those that do not embrace SOA principles. This will become apparent first as SaaS and hosted, on-demand applications providers.

For those IT shops that throw their infrastructure fates to Microsoft's software development and business development competencies (where the latter is the strength), they may encounter cost and agility disadvantages.

If I and Dave Rosenberg are wrong on SOA benefits and n Microsoft's lack of general support for SOA, then the more purely .NET shops should demonstrate market strength via lower TCO and greater business agility over time. The .NET-based businesses should play better amid complex, global business ecologies, and be able to take advantage of mixed sourcing across a waterfront of available services.

Under the dictates of comparative advantage, .NET and its Microsoft-oriented progeny should not create greater value and higher productivity for its customers than more general alternatives, such as open SOA. A choice of the best service for the job will dominate a choice over only the .NET service available.

I'll be waiting and watching, though my risk is an observer is much lower than the architects placing their bets in the coming years. The ante, incidentally, is the very survival of their companies as we enter an "flattened world" era of increased globalization, lower barriers to entry, open trade, and lightening fast market disrupters.

For those with an inclination to hedging bets -- banking on general SOA while supporting service-enabled .NET might be more secure and auspicious than banking on .NET while trying to support SOA objectives from a comparative disadvantage.

Tuesday, September 11, 2007

BEA, Adobe flex their muscles in pushing toward RIAs on the enterprise desktop

BEA Systems Inc. and Adobe Systems Inc. are teaming up to give enterprise customers a design environment for rich Internet application (RIA) development. The two companies announced Tuesday that BEA will bundle Adobe's Flex Builder 2 software with BEA Workshop Studio. This move will let developers build cross-platform RIAs that integrate with services oriented architecture (SOA) and Web 2.0 infrastructures for enterprise mashups.

Under Tuesday's agreement, announced at BEAWorld in San Francisco, the BEA-Adobe bundle will include Flex Builder 2, as well as the Adobe Flex SDK, which operates under the Mozilla Public License.

In April, I predicted that Flex -- especially operating under open source -- could become an industry standard for RIAs, and this latest announcement might be an indication that prediction is coming true. What I said back then was:

"With Adobe bringing its Flex and possibly Flash into open source — and perhaps creating unassailable de facto global standards as well — then the Web 2.0 red shift to RIAs and away from other models could be complete."

Using the Adobe Flash Player, which is nearly ubiquitous in home user applications, developers will be able to build interactive data dashboards, customer and employee self-service applications, and B2B applications that will be agnostic as far as platforms or operating systems.

The applications created with Flex can then be integrated with other BEA products -- including those from the WebLogic and AquaLogic families -- and to deploy the applications with Adobe's Intergated Runtime, a cross-operating system app runtime that allows developers to extend RIAs to the desktop.

As part of the agreement, Adobe will distribute an evaluation license of BEA's WebLogic Server with Adobe's LiveCycle Enterprise Suite (ES) software, giving customers a turnkey infrastructure to deploy LiveCycle applications using WebLogic characteristics.

In other news from BEAWorld, BEA released the AquaLogic Registry Repository 3.90, the first product component of WorkSpace 360º, which in turn is a building block of BEA's newly announced dynamic business application platform initiative -- code-named Genesis.

Genesis is expected to converge SOA, business process management (BPM), social computing, and other technologies to better manage the service creation lifecycle across the enterprise. Tony Baer has some good thoughts on Genesis and BEA's actions this week.

The goal of Genesis is to allow businesses to adapt to changing market conditions without the constraint of old IT models, allowing the enterprise to assemble, change, and deploy dynamic business applications. Look for BEA to unveil the roadmap for Genesis at BEAWorld in Shanghai in December.

Registry Repository 3.0, which was announced Tuesday is designed to improve the management and governance of SOA deployments, combining the governance capabilities of BEA's AquaLogic Enterprise Repository and Service Registry, providing governance throughout the SOA lifecycle.

Among the features of the new repository are:

  • A central repository for sharing and managing metadata to enable a more seamless flow of information across the stages of the lifecycle.
  • Embedded workflow to provide a structured process for managing assets as they are developed.
  • Unified tooling to create a seamless navigation.
  • Open metadata interoperability framework for integration by third-party technologies.
  • Embedded governance and control throughout the lifecycle to help ensure that dynamic business applications are defined, designed, built and run in alignment with business goals and objectives.

Monday, September 10, 2007

Analysts 'debate' to focus on SOA management issues and outlook

I'm really looking forward to this Friday, Sept. 14, in Boston when I join up on the podium at the Harvard Club with Jason Bloomberg, senior analyst with ZapThink, to dig into the current and future landscape of SOA management.

Jason and I will be examining such questions as:
  • What should management look like in the world of service-based applications?
  • Can traditional monitoring tools be effective at managing tomorrow’s SOA deployments, or is something new needed?
  • What standards are needed to make SOA adoption a success, or do they already exist? What role do they play, and what are the alternatives?
  • As enterprises increasingly deploy SOA-based applications and infrastructure, the lingering question remains: how do current plans and approaches to application management stack up to managing these mission-critical deployments?
Tidal Software is hosting the lunch gathering -- they're calling it debate -- and then sponsoring the production of a BriefingsDirect podcast from the discussion. Tidal CTO Martin Milani will be the moderator. We'll point to the podcast here when it's available.

Tidal, incidentally, provides IT managers visibility into, and offers control over, how newer SOA-based composite solutions perform. This means management for both packaged applications and custom components in Java and .NET. I'm very interested in how SOA management interfaces (or not) with other forms of business management. That is, what are the next steps?

Can we go from SOA management to larger-abstraction management with a language and interfaces that appeal to both business managers and IT planners/operators? Is there, after all, a larger architecture of management in the offing, one in which SOA management acts as a catalyst toward business management and operational change management? Also, are we in need of standards, or a commercial best-of-breed approach, to begin this process?

I don't think that IT as a corporate resource can continue to avoid the huge opportunity for providing value to the corporate leadership by being the automated and integrated means to execute on the outcomes from various disjointed facets of business management. Some day, perhaps, to manage the SOA is to invoke change and agility across IT resources comprehensively, top-down and bottom-up.

And perhaps SOA management will therefore be a valued steering wheel on the dashboard that actually runs and changes a business. CEOs and COOs, I'm quite sure, would rather use technology as the rudder to business agility, rather than consider it a high-cost handicap to change.

The launch into this topic, by the way, began with some seemingly opposing quotes from Jason and myself in a article earlier in the summer.

So, if you're in Back Bay and care to dive in with us on all things SOA management, come on by. The luncheon event is at 11:30 a.m. on Sept. 14 at the Harvard Club of Boston, Estabrooks Room, Main Clubhouse, 374 Commonwealth Ave., Boston.

You can RSVP at 650-475-4645, or via See you there.

Saturday, September 8, 2007

BriefingsDirect SOA Insights Analysts on RIAs, Microsoft Silverlight and Enterprise 2.0 trends

Read a full transcript of the discussion. Listen to the podcast.

The still-maturing technology around Rich Internet Applications (RIAs) and rich media interfaces and video players was jolted last spring when Microsoft's Silverlight offering was unveiled. Already a Linux version is in the offing called Moonlight. The RIA news hit about the same time as the JavaOne and Web 2.0 Expo events were in full swing as well.

We used the timing to bring some IT analysts and experts together for a podcast discussion to examine the role of RIAs and rich media with SOA and the impact on the Enterprise 2.0 space. Join noted IT industry analysts Joe McKendrick, enterprise architect Todd Biske, and independent blogger Barb Darrow for our discussion, hosted and moderated by your's truly.

Here are some excerpts:
We seem to be moving beyond just the notion of an RIA into specific platforms, and/or approaches for doing this. We now have a slate of new products and approaches from Microsoft around the Silverlight brand. We also have news from Adobe about open-sourcing the Flex toolset that helps create content that’s supported on the ubiquitous Flash seamless download client through browsers. And, we've also seen Sun Microsystems pony up with the JavaFX scripting language, also designed for RIAs.

Are RIAs are more than a sideline, and are they becoming a mainstream way of bringing content, data, and applications to users?

It’s a nice step up from the browser interface that we've all been accustomed to for the past decade now and very competitive with the fat-client concept of Microsoft Windows, which is still Microsoft’s bread and butter. To a large degree, they should feel threatened by this.

Two or three years ago, I was invited to a Microsoft technology summit, and they collected about 40 of us in Redmond. It was just a general discussion around some of the things that they were doing, but it was a series of diehard Java advocates, diehard Flash advocates and diehard Linux advocates. It was an interesting exercise just to listen to what they had to say. Microsoft was really trying to hear what would make Microsoft more attractive.

I don’t know whether coming out with a direct competitor to Flash is going to make it any more attractive in the eyes of the Flash developers, but I think certainly playing to their strengths in the existing Microsoft development community, and bringing in some of the best practices of the other development communities, is a smart move on their part. That’s why I think Silverlight will continue to play a role.

We went from green screens all the way to these rich desktop environments. Then, we went back to just the HTML forms, and you knew that pendulum was going to start to shift back towards the center again, that users were going to need higher levels of interaction and capabilities on that Web-based platform. I think this is just indicative of that trend.

In enterprise circles, it’s much more about AJAX than it necessarily is about Flash. Then, you have Silverlight, and now JavaFX Script, which I think are more in the same category as Adobe Flash, than targeting the AJAX world. I've yet to see an enterprise application focused on Flash development. It seems to have much more of a place either in content distribution or the general Internet space. Still, it’s gaining at least mind share, and so we’ll have to see whether this begins to make a push more to the corporate enterprise world.

This whole move to rich clients is interesting. I cover IBM software, and I've got to give them credit. IBM has been talking about this for a while. There’s this kind of contention between new kids coming up, who are used to downloading everything they want and doing mashups -- they grew up this way. And then there's this traditional IT environment that constrains from above what you can do. In that spot, IBM has a little bit of credibility. IBM is trying hard to adopt this mashup/social networking thing going forward, but I'm just wondering. Are they a player here?

IBM does things in a big way, and I've seen them doing a lot of work in this area, in terms of Wikis and blogging. They're even getting involved in the whole second-life scenario. They have a way of moving into these markets in a very big way, and I don’t see them ignoring the whole Web 2.0. Like everyone else, they're piloting things, seeing how it fits in with the enterprise.

IBM probably needs to have some activity in this place soon, because, on one hand, we can look at Microsoft and Google, and they are both application providers outside of the Lotus space. IBM is not an application provider in the same sense. So, some of the things that you see Microsoft doing with Microsoft Live and the Google applications, I wouldn’t expect to see any big push from IBM.

Another thing that struck me at JavaOne this year was the dearth of announcements from other major Java-oriented vendors, and I'm thinking of IBM, BEA, and Oracle. It was really a silence, and what I think has happened is that Sun waited for so long to declare its intention for Java, and then to open-source it under GPL Version 2, that they lost the community. Now, the community is off doing things under Eclipse, Apache, SourceForge, OSSI, or whatever. So, the momentum of the community and the ecology for Java was lost, as Sun basically sat on the fence, trying to figure out how to make more money from Java. I don’t think it’s something they're going to recover from.

This comes back to the whole notion of the client side of this. Will Java, as a development platform, have a role in the development of the client side? It’s well established on the server side, and that’s not going to change any time soon, but what is the future of the client platform, and will it be a case of these RIAs coming down into the enterprise?

Or, will we continue to see a separation of "Here are things down in the content-heavy world of the Internet" and "Here is the corporate world?" Even in the corporate world, either you’re building Microsoft applications, because that’s what’s on everybody’s desktop, or you're building Web-based applications. More and more of the presentation technologies are going towards AJAX, rather than anything you're doing in Java JSP.

Maybe there's a third way on this, and that would be that you go for the minimalist, when you are dealing with data, transactions, and workflow issues, but there is a whole other side of enterprise productivity around collaboration, learning, discovery, and knowledge transfer. These videos and rich media, be it text, audio, or video, whichever you choose, or all three, could be very powerful. We could see instances where we are going to get both. We are going to get a lot of minimalist widgets, but we’ll also get lots of rich, movie-grade video, when it comes to the other side of the equation, which is not dealing with machines and data, but dealing with people.
Read the full transcript for more IT analysis and SOA insights. Listen to the podcast here.

Produced as a courtesy of Interarbor Solutions: analysis, consulting and rich new-media content production.

Thursday, September 6, 2007

Nexaweb and Kapow team up to bring Web-based mashups to enterprise desktops

Enterprise 2.0 vendors Nexaweb Technologies and Kapow Technologies have teamed up to smooth the way for bringing internal and external data into enterprise mashups and composite and other rich Internet applications.

By marrying Nexaweb's Web 2.0 Platform with Kapow's Mashup Server, the partnership will extend the reach of enterprises to take any application component with a Web interface and repurpose it into an application involving Web-based content, data, or business logic. This real-time access to wide-ranging data sources -- internal or external -- will give businesses the agility they need to become and remain competitive in an increasingly data-driven environment.

David McFarlane, Nexaweb COO, says the combined technologies will allow companies to peek out from behind the firewall. "External data, such as market quotes, politics, world events, weather, traffic, and third-party analysis and commentary can have significant impact on your business," McFarlane said. "They should be available to your workforce to help them make decisions faster."

Nexaweb's Web 2.0 Platform, a standards-based application development and deployment solution, allows organizations to tap into legacy, SOA (services oriented architecture), and third-party data to deploy Ajax-based business applications over the Web.

The Kapow Mashup Server, a Java-based solution, uses the Web front-end as the interface to integrate Web-based content, data and applications. Kapow uses the ability to access the web interface to create mashups at the user interface level (UI), at the application logic level, and at the data level. The mashup server also includes a highly-scalable, robust enterprise-class deployment framework and a visual design environment.

Tony Baer at Computer Business Review Online sees a wide range of opportunities from the partnership:

"That means that, besides data from database sources, you could now treat wikis, blogs, or other web document-centric content as first class data in for the data driven mashups that are more suited for enterprise use than presentation-oriented mashups that literally placed this style of development on the (Google) map."

Nexaweb and Kapow will be seeking customers for deals involving bundled capabilities, and are offering a discount for companies that sign up in the first few weeks of the offer.

TIBCO revs up enterprise Ajax with General Interface 3.5

TIBCO Software gave a boost to Ajax in the enterprise last week when it released Version 3.5 of its open-source Ajax toolkit, General Interface (GI).

The new version of the award-winning package includes improvements to performance and tooling and offers such other enhancements as an integrated context-sensitive help system -- which TIBCO claims is "nifty" -- and reworked vector drawing APIs. When a developer creates lines, shapes and fills with the API, GI will render them as VML for Internet Explorer and SVG for Firefox.

Performance enhancements include faster data display and rendering times, as well as a boost in speed to make IE6 rendering nearly at parity with more up-to-date browsers.

As far as tooling enhancements, a new benchmarks toolbar gives file size, load time, render time, and HTML size, which allows developers to optimize early in the development process. A drop-in debug build can also alert authors of the need to optimize.

GI was voted Best Ajax Toolkit by Infoworld in 2006. Either the Professional Open Source Edition or the Enterprise Edition can be downloaded from TIBCO.

Wednesday, September 5, 2007

MuleSource fires up MuleForge collaboration site; Apache Tuscany team announces 0.99 SCA release

Collaboration is at the heart of open source, and in that spirit, MuleSource, which provides open-source infrastructure and integration software, this week launched MuleForge a Web platform that will allow developers to explore, download, test, and contribute to re-usable mule extensions.

With over 45 projects already on the site, MuleForge is designed as a development base for the Mule community, and offers tips, documentation, and other resources. It also automates building and compiling code, and collecting dependencies, which should save developers large amounts of time.

Among the projects already on the MuleForge site are a plug-in for AS/400 queuing support, a JavaSpaces integration package, and a session initiation protocol (SIP) connector for integrating data and services in VoIP and other telco applications. There is also a connector that allows real-time communications between and internal data sources.

Additional features include:

  • Source code control
  • Continuous builds
  • Issue tracking
  • Documentation wiki
  • Project statistics

MuleForge is also designed to provide community coverage for individual projects and gives access to hundreds of developers to further develop and test code. To sweeten the pot for potential contributors, Mule is offering individuals prestige -- a gift denoting "elite status" -- and the all important cash. Once a quarter, the developer who submits the best project will receive $500 and a spot on the home page.

In other SOA open-source news, the Apache Software Foundation's Tuscany team has announced the 0.99-incubating release of the Java SCA project. SCA is a set of specifications aimed at simplifying SOA application development.

The current release builds on the stability and modularity of the previous releases and includes more complete implementation of SCA specifications, support for distributed SCA domains, SCA policy, OSGi implementations, and pub/sub support. With numerous bug fixes, the 0.99 release is expected to be the last point release before the 1.0 version.

The release and further information can be downloaded from the Apache Tuscany team's Web site.

Friday, August 31, 2007

IBM's Cell processor looks like a candidate to power numerous infrastructure appliances

News that IBM has announced an upgraded version of its blade server based on the Cell processor this week has me wondering about the versatile and powerful "systems on a chip"'s use in appliances, too.

We're heard some hinting and "that's a logical outcome" statements from IBM officials in the past few months, so the pairing of Cell and appliances would not come as a complete surprise. When the details emerge, the price-performance and ease of deployment benefits of these high-powered, multi-core appliances could be very impressive.

When IBM announced plans to buy Telelogic, the deal made sense to me more when the use of a specialized Cell-fired appliances was made part of a possible future portfolio. When we had Jim Ricotta on a recent BriefingsDirect SOA Insights Edition roundtable podcast, the IBM general manager of appliances indicated more specialized appliance to come from IBM, though he did not finger Cell specifically.

Appliances from IBM should be expected in more componentized infrastructure roles in the coming months, for sure. They make a great deal of sense for data and content optimization and balancing, for SOA-support functions such as ESB, registry/repository, and as discrete services support stacks in a box (a business service appliance).

Those specifying services or functions will not need to consider the underlying platforms or inherent low-level integration issues, just focus on the largely standards-based interoperability characteristics for these functional units. Appliances allow greater exploitation of open source efficiencies by the vendor, with less complexity for the end user, and a better margin for the seller (more than just service and support).

Indeed, we may see some sort of a face-off in terms of total cost and performance between virtualization approaches and appliances approaches. Why not use both? I expect that appliances may very well be filling up larger list of new requirements for enterprise architects over the next two years.

The multi-core attributes of Cell, plus the proprietary 'Synergistic Processing Elements' (SPEs) for the chips provide the means to exploit parallelism and finely tune each box for the specific functionality at hand. The fact that these specialized and closed functional components (hardware, software, integration, optimization) require much less set-up and life-time support, appeals to architects (if not integrators). They may also help on energy use and heat-dispersion issues as well.

The ability to scale by virtue of adding (or subtracting) boxes, plus the ease of swapping and redundancy -- all bode well for more appliances-driven architecture (ADA [... sorry]) for SOA and high-performance yet specialized computing at the best total cost. These attributes will be of interest to hosters, service providers and telecommunications providers.

IBM's Ricotta told our analysts that appliances can cut costs by half, compared to traditional deployment approaches. When you take such economic common sense and toss in the technological secret sauce of optimized and specialized Cell chip-sets ... the balance of Power could well shift toward appliances in the most competitive datacenters.