SOA Governance

I'm trying to research to understand what SOA governance is. Everyone writes about it as if everyone else already understands what it is. So I went to a vendor to try to understand.

At first glance, the word governance speaks to rules and control. Some of the writing I've seen led me to believe that in the SOA world, it wasn't rules and control from a human perspective, but from some type of technology implementation.

The more I read though, I learn that it is a mix of both. I can't seem to find anything definitive on what technology tools there are or what they do, but there is more indication of the human side of the governance process.

Yes, it appears someone has to set some standards, make appropriate decisions about what to implement and what not to implement. But there may be ways to monitor and/or control how it it is implemented on a network.

I guess I'll keep reading.

http://www-306.ibm.com/software/solutions/soa/gov/

Still no Focus on the “A” of SOA

Dave Linthicum went to the SOA World Conference recently. He laments that everyone seems focused on the technologies that support SOA. But yet the vendors can't explain how to understand how to leverage their technology to meet your needs.



It seems that everyone has forgotten that the A in SOA stands for architecture. I could see how that could easily become an issue. Things like this are easier for everyone to understand if you can "buy" something. According to Dave, though, you don't buy an SOA. It is something you do.



Makes sense to me.



http://weblog.infoworld.com/realworldsoa/archives/2007/06/still_no_focus.html

Where have all the SOA Standards Gone?

So Dave Linthicom is commenting on how few standards or discussions about standards there are concering SOA right now. I have to agree with his statements at the end. Standards help and are important, but they have to come together on the implementation of technology. They can't just be about marketing. According to Dave, that appears to have been the trend in the beginning.

You have to wonder who is writing the standards, and who is buying into them. What bodies approve/finalize them? Is this a pick and choose process? Or has the process of developing standards become confused and lost its focus?

In my mind there shouldn't be 60+ standards for SOA. I doubt seriously there are 60+ technologies involved. Is there a right number of standards? My answer would be one or two on each technology. Is that a pipedream? Is it unrealistic? I just know that it needs to be useful rather than overwhelming or conflicting.

http://weblog.infoworld.com/realworldsoa/archives/2007/07/where_have_all.html

IT EMIS 17

IT EMIS 17

SOA business case: smaller is better

Joe McKendrick's blog makes a case for starting small in order to learn how to monitor benefits. This way you can see your business becoming more flexible and adaptable. Doing smaller projects that directly target specific business problems may be the best way to go in order to build the case for SOA across the enterprise.

This approach makes sense for a number of reasons. Smaller projects cost less and may fit better within constrained IT budgets. They usually have more achievable goals that can be reached faster. It won't take three years to deploy an empty system that delivers potential but has no data yet to realize that potential. You can work out implementation kinks and business process issues on a much smaller scale. And the lessons you learn will be much less painful than bigger projects if you don't succeed. It should be easier to understand where you went wrong. Big projects have a tendency to evolve in bad ways rather than simply having a single, major obvious mistake.

As Joe points out, the really good news about this approach is that it is win-win: small-scale technologies to fix a business problem while providing scalable, secure and resilient platforms that can form the basis of a tranformation.

One could almost assume that this is intuitively obvious. But I can see where companies could get sucked into the concept of fixing it all at one time.

http://blogs.zdnet.com/service-oriented/?p=903

IT EMIS 17

IT EMIS 17

IT EMIS 17

SOA approaches pondered

Okay, so I'm finally reading an article that is describing SOA more than most. According to the author, SOA is an enterprise architecture design methodology and is not about technology. It started out that way, but then it drifted...into Web 2.0.

Rob Levy, BEA CTO, says the next wave of SOA will bring Web 2.0-like capabilities into the enterprise. So now we are combining topics. According to Levy, the next generation of SOA applications needs to cater to human nature and collaboration. Because we are, essentially, social creatures with a need to collaborate.

In general I believe we are doing more to collaborate in the workplace, but is it enough? And can security needs support collaboration without stifling it? Time will tell, but I believe it is going to come down to risk management. We are going to have to manage more risks than we ever have before.

That is what it is going to take to succeed in business tomorrow. Maybe not today, but definitely in the future. Everything and everyone are going to have to integrate to some degree to move forward, and not be left behind.

Good luck world, we're gonna need it!

http://weblog.infoworld.com/techwatch/archives/012744.html

SOA, RIA and the Human Factor

So I finally found an article that tells it like it is. I am shocked at such tactless honesty! Who would have thought anyone would say such a thing out loud in public!!

"One reason an organization might pursue Service-Oriented Architectures is to help them get promoted. "

I would NEVER have guessed that anyone would try to push for such a change for so selfish a reason as career goals. NOT!

So he goes on to describe how hard or easy it might be depending on where you are in the organization. I love the sarcasm in this article. But he is just pointing out realities. This stuff really happens. I have seen my own organization do it, though not with SOA.

Ultimately he provides a realistic view of how to approach the process of making a decision to transform to SOA or not. An interesting read.

http://soa.sys-con.com/read/396343.htm

New Adventures in Spin

Open XML seems to be moving toward a standard. Extremists are getting less and less extreme.

Issues that are left include ones that just need clarification in the text. Otherwise known as "wordsmithing". Some things, however, are going to involve sytax additions.

For the article I read, the blogger was asked to brief the Bureau of Indian Standards. He indicated it was a very interesting opportunity. Lots of good questions.

One of the problems in this standardization process is trying to build a standard for pre-existing technologies. This standard does not want to transform the spec.

"The best-practice for a fast-tracked standard is that the spec clearly describes the technology under consideration warts and all, avoids unnecessary platform dependencies (which is not to say that it need not describe platform-specific features) and that it does so in a way that allows maintenance and subsequent improvements."

The writer didn't feel that the reviews would yield "yes" or "no" votes. Instead they will get "yes with comments". Should be fun. I hate that part.

Service-Oriented Architectures and Business

Well, maybe they don't mix after all. Or if they do, the business world isn't getting it.

How soon will SOA be a part of the way we do business? According to IBM and the SOA Consortium, we'll see a strong rise in the top companies in the next three years. Saugatuck Technology and Jack van Hoof say it will take at least 10 years before we are really there.

The answer could lie somewhere in the middle. I believe it will depend on how much change is needed in current architectures to meet this as a goal. And more important, should SOA be a goal of business? Will business be able to compute and demonstrate the ROI for SOA? The risks for IT projects where the intent is contribute to SOA will be an inability to prove ROI. And it will be extremely difficult for IT departments to define their business case to support decisions in the direction of SOA.

Ultimately the question will come down to: SOA - savior or black hole?

Introduction - VoIP

I have only just figured out that I need to do a blog. I thought was supposed to add to the RSS feed multiple times each week.

For my first attempt, I am combining two classes and discussing Voice over Internet Protocol (VoIP). An old "friend", Deutche Telecom, is investing in a company that provides phone services over the Internet. The arrangement with Jajah will give Deutche Telecom a test bed that provides them with the opportunity to check out new telecom revenue generators. Jajah gets the backing of one of the largest telecom companies in the world.

Jajah has a tough road ahead of it. It is still small, and it is facing lots of competition. Deutche Telecom has to find ways to maintain revenue since so many clients are moving away from traditional phone services.

Good luck to both. Hopefully Deutche Telecom is doing better with its reputation than it was when I was living in Germany!