Thursday, November 13, 2008

Gartner says that initial SOA adoption rates are slowing. I think they are wrong.

I just read a Government Technology article stating that Gartner says there is a dramatic decrease of organizations planning first-time SOA projects. Here's the money quote.

Since the beginning of 2008, there has been a dramatic fall in the number of organizations that are planning to adopt SOA for the first time. In 2008, this was cut by more than one-half, down to 25 percent from 53 percent in 2007, while the number of organizations with no plans to adopt SOA more than doubled from 6 percent in 2007 to 16 percent in 2008.

Unfortunately, I no longer have direct access to the Gartner research (One of the things about leaving my former employer that I will miss the most.) so I have to comment only on the secondary source, the above mentioned article. That's too bad because I can't look at the methodology used to gather the research. Knowing Gartner from the past, however, I can make some guesses.

I'm guessing they talked to IT departments and asked, “So, what are your spending plans for SOA?” and probably, “Have you already adopted SOA within your organization?” Those aren't bad questions, but I think we could all have predicted the results. Given today's economic uncertainties, Big IT isn't going to spend lots of money on big IT projects that don't have an established track record for success. Unfortunately, the success of SOA in the enterprise, at least Big SOA, is not a given.

But these results don't even come close to telling the full story. I personally have worked on three projects over the past few months that were absolutely based on a services architecture. However, if you were to ask the IT departments of these organizations about the projects, they would not have tagged them as SOA.

Why not?

Because these weren't 'SOA projects,' they were business initiatives whose solutions happened to make use of SOA. In all three cases we didn't buy expensive middleware to run the software. We didn't embark on an orgy of service writing to SOA-enable myriad legacy systems. In two of the three cases it's probable that I was the only person who knew the underlying architecture of the solution was services oriented. These were true Guerrilla SOA projects.

So while I completely believe that Big IT departments are slowing down in their implementation of Big SOA projects, I don't believe for a minute that these same organizations aren't expanding their use of SOA. It's just that IT doesn't know. And what they don't know, they can't blab to Gartner.

So take what Gartner says with a grain of salt. Sure some of the Big IT SOA projects may be on hold, but don't assume that Big IT owns all the action. SOA is happening all over, without IT knowing anything about it.

Tuesday, November 4, 2008

Hello again, everyone. (Or all two of you)

Loyal readers may remember that I stopped writing this blog when my boss instituted a policy requiring all blogs be approved by PR before publication. Since I wouldn't submit my blog to PR, I stopped writing it, at least under my own name.

I'm no longer with my prior company, so it is time to start up this blog again. I've got a backlog of issues to write about. So many, in fact, that you can think of this as my Q4 2008 editorial calendar.

  • Finally, after all my pleading, all my whining and all my cajolery, I get to test drive JackBe Presto. We've gone a round or two in the past, with JackBe refusing me access, and me swearing that I would give them a fair review. You can read my requests here, here and here. I'd like to say that my razor sharp reasoning finaly convinced JackBe that free and open access to their technology is the best marketing strategy, but it wouldn't be true. It's just that I now no longer work for a company JackBe perceives as a competitor.

    So far it has been a bit of a trial. Since I had to give back my corporate computer, I'm starting from scratch with a laptop having only, god help me, Vista, installed. Before I can install the Presto suite I have to install the JDK and Eclipse, which reminded me of all the myriad libraries and jar files I already had installed on my old machine. I suppose I'll figure out what's missing when I write code and don't have the appropriate jars...

    So I haven't quite gotten Presto installed yet, but hopefully some time today I'll be able to dig in.

  • The financial services industry has been an early adopter of business-driven initiatives using guerrilla SOA and mashups. Needless to say, this market has other things on its mind just now. The question we are all asking is, "What's in store for us given the meltdown?" I've been building mashups for a handful of global banks for the past nine months, and I've got some opinions about what we can expect for the next nine months.

  • Interneer looks like another promising application builder tool, along the same lines as Coghead and ActiveGrid (WaveMaker). Once I finish with Presto, I'll take Interneer through its paces and let you know how it stacks up against its rivals.

  • Despite our best efforts, the hype surrounding SOA is dimming. This is actually a good thing, not a bad thing. SOA is no longer newsworthy because it is passe. Mainstream. Old school. Dull. As SOA moves to the mainstream, raging debates about SOAP versus REST, stateless web services, and whether Event-Driven Architecture (EDA) is an extension of SOA, seem to be naval gazing. I'll give my take on the state of SOA, where it is, and where I think it will be moving in the next year.

  • The industry is lousy (Nasty word. Read the etymology here.) with WS* standards. Along with the product reviews, I'll start to explain some of these specifications and how they fit in to the evolving SOA 'standards.'

That should take me through Q4 2008, along with random rants from the field. Feel free to write with suggestions for topics, or just to say, "Hello."

It's good to be back.