I’ve been thinking about the current state of things, the good, bad, and the ugly of this new space, and I figured I would share some thoughts:

The Good

What’s cool about SOA is that you can actually make the business significantly better. The notion of reuse and agility is a real benefit, and those who take steps to implement the notion of a SOA correctly will gain great benefit from it. I love that aspect of SOA, and feel that it will both heal some enterprises, and optimize others.

The Bad

We are suffering from the same mistakes we made in the client/server and distributed object days. Lack of planning, and over simplification of the technology is leading to some quick failures. Those implementing this approach and technology need to understand that it’s a journey, not a project, and it’s not just an application, it’s a way of doing things.

The Ugly

The confusion. The “ESB” thing, the “Registry” thing is ugly stuff. Not that the technologies are bad, but consumers of buzzwords are confused about what TLAs (three letter acronyms) relate to their issues, and then finding out that there is no standard for TLAs. Insert my last rant here…

All and all, things are okay on the planet. However, they can always be better.

Posted by davel in | Permalink | Comments (0)

SOA Expert Podcast Show 33...My Interview with ARCast a Crosscast

Listen to the latest SOA Expert Podcast

Posted by davel in | Permalink | Comments (0)

Ron Jacobs, Host of ARCast was nice enough to have me on his ARCast Podcast show from Microsoft SPARK. This guys has way better Podcast equipment than me, I have a lot to learn.

Anyway, you can find the Podcast here.

Posted by davel in | Permalink | Comments (0)

Listen to the latest SOA Expert Podcast -- EAC Keynote 12 Steps to SOA.

Slides are here:

Download file

Posted by davel in | Permalink | Comments (0)

Posted by davel in | Permalink | Comments (0)

Or, go to www.soaexpertpodcast.com.

Posted by davel in | Permalink | Comments (0)

Lately I’ve been looking at a trend within larger software companies. They have two roles now, the CTO and the Technology Evangelist. I’m not sure I get that.

I’ve been a CTO at my last 3 companies, and it was my job to evangelize the technology, including presentations, articles, books, and more. Strategy was easy if you were the communicator as well. Why? Because of the data points you got out there in the real world which feed directly into your understanding of your own technology strategy, and the value of that technology strategy in the marketplace.

One aspect of the job, speaking with the press, standards bodies, and end users, lead to better information, thus better product and technology strategy. Thus, I’m not sure I could just sit in some office somewhere and create the strategy without attempting to influence others around the acceptance of the technology.

However, a few companies are locking their CTO in the back room and hiring professional evangelists to communicate with the outside world. You really need a CTO that can do both roles, else you’re going to find they can do neither.

Posted by davel in | Permalink | Comments (1)

I’m off to many speaking events over the next few weeks, including the Info World SOA Executive Forum. This is the 3rd time speaking at this event, and I was just thinking about what topics are going to be hot this time out. They are always different.

If any of the e-mails I’ve been receiving are an indication, I think that the productivity angle of SOA, and just getting a project going, are going to be the hot topics.

I think we’ve analyzed to death the notion of a SOA, debated topics such as ESB, swore on Web services standards, and designated some people in the organization as SOA Gods, now it’s time to get to work. Who’s with me?

So, the focus should be on first or second steps, including getting ones house in order, preparing for the layer of new stuff. This includes a semantic understand of your enterprise universe, and an accounting for the services already there.

For most enterprises, this is a year long effort, and a lot of busy work. Many will be tempted to jump directly to the technology analysis phase, but will find that doing so too early in the process will kill you before you get the SOA project out of the gate.

Moreover, we need to keep in mind that SOA is a journey, not a project. So, you have to think holistically, not in the terms of a single project life cycle.

That said, let’s get started.

Posted by davel in | Permalink | Comments (0)

Just got out of the shower and I was thinking about some of the ways to avoid SOA failure, based on some of the things I’ve been seeing in the world of the SOA technology end user. Here are a few thoughts:

First, never ever, ever, never, ever, let a vendor drive a project for you. Not that vendors are evil, but they are not independent thinkers when it comes to technology. I’ve seen 4 companies make this mistake in the last 3 months, and are correcting now at huge expense.

Second, don’t’ manage by magazine. What worked for that cool company in that magazine is a great. However, your problem domain is unique, and needs unique analysis and solutions. One size does not fit all.

Finally, A-B-C…Always Be Consider…alternative technologies, the skills of those building your SOA, the driving forces behind the SOA, and this includes ROI. Yes, we need to consider ROI people.

Don’t forget, I’m speaking at the Info World SOA Executive Forum next week in San Francisco. I will be wearing cloths then, so feel free to stop by and say hi.

Posted by davel in | Permalink | Comments (0)

The notion of ESB has been getting some knocking around lately. Why, because of what the vendor community is doing to that term, it’s confusing everyone I know that’s seeking SOA solutions right now. Indeed it’s getting a lot of negative blog-perspective lately, including this, this, and this.

This does not mean that ESBs are bad technology; I’m sure many of the ESB products out there work well. I’m just pointing out the fact that ESB, the term, has three major drawbacks…it’s overused, over hyped, and confusing. Again, generally speaking.

The fact of the matter is that no two ESB products are alike. Some are messaging systems with Web services interfaces, others are renamed message brokers, and other are orchestration layers that no longer want to be orchestration layers now that BPEL has a few black eyes. It’s a marketing push that’s leading many vendors to jump on the ESB bandwagon, but it’s actually just making SOA, which is very complex already, even more confusing, which is not good for this emerging space.

So, what do you do? Just say no to ESB. Focus on your own requirements, and what the product does specifically to meet them. It matters not if it’s ESB, EAI, B2B, SaaS, or EIEIO. Look past the marketing at what the product does, and draw your own conclusions. If you do that, the whole ESB thing will just stop, and ESB, the term not the technology, will no longer be interesting but the solutions will.

Posted by davel in | Permalink | Comments (4)

You can subscribe or download to the podcast here.

Posted by davel in | Permalink | Comments (0)