john devadoss' blog

I have had the opportunity to work very closely with the patterns & practices team over the years. We have collaborated on the Composite Application Block (which was incubated in my team many years back), on Web Services patterms, on the Smart Client Architecture Guide (a blast from the past :-)) and more... I have had a great deal of respect for the team, the people, and their ability to take on challenging problems, across diverse domains and deliver.

so, it was a privilege and a pleasure to be asked to take on the role of leading the team.

There are some amazing people in the team, the passion is contagious and something to be seen to be believed - and there few places if any inside Microsoft, or any other company for that matter, that give you the opportunities to work in this area.

I am very excited. And looking forward to this journey.

ps Glenn I did eventually blog this :-)

There is a recurring pattern that I am seeing in the large enterprise space that I think of as the 'my cloud' pattern.

Essentially, large enterprise organizations that are interested in the business and technology model advantages of moving to a software + services model, but are asking their IT organization to become a service provider, as opposed to out sourcing their IT portfolio to the sky.

It is often (implicitly) assumed by many that BPM comes after SOA.

The challenge, even with little BPM, is that the focus is primarily on optimization, not on innovation. It is almost always about the 'boxed-in process' - and about drilling down - less about stepping back and out. Paving the cow-paths has its pros and cons as you can imagine - but there are few if any that would argue that paving the cow-paths is the path to innovation...

anyway, this thread was spurred by some discussions recently on what is the next big thing...

Is it 'management' when you sell to the technical decision maker - and 'governance' when you sell to the business decision maker?

I know I am being cynical here. But I see too many web services management vendors that now call themselves soa governance vendors.

I absolutely understand the pain and the need for governance. But it has always been and it is still about the people and processes 

Vendors cant sell governance - nor can customers buy governance.

Governance is hard work. It requires driving cultural change across the organization. There is no short cut to success here.

Joe's column is one I have been following religiously for a while - a must read if you at are interested in SOA.

His write-up on how we think about SOA really nails the core of our philosophy. Just do it!

well, I am out today with a virus and am coughing my way to the doctor - but that is a different story.

We talk about 'discovery' in SOA, cough, cough. I have been working with customers on SOA for over 7 years now - and I must confess that I haven't seen enough of discovery to beat the drum on this.

Do you have real-world evidence/case studies to highlight this in your work? Please drop me a note if you do.

Following up on my post from last year http://blogs.msdn.com/jdevados/archive/2006/08/02/687269.aspx - I think there is a similar distiction that we can make for BPM.

Alex Moulas nails it in his column http://www.bpm.com/FeatureRO.asp - "A large, progressive company wants to be a more processed-focused enterprise. Best practice dictates combining business analysts with consultants from a major system integration firm to detail architecture and interview employees to capture the real business process. Plans are developed, and with methodologies in hand, significant initial progress is made. As weeks and months progress, more time is spent managing documents, drawing and re-drawing process maps, reviewing, correcting, checking interdependencies and trying to ensure consistency and uniformity. After 11 months, the combined teams are disbanded and no actions or conclusions are reached. Budgets exceed planned expenditures and frustration prevails. Ultimately, the entire project is put on the back burner or …

This is a typical “best practice labeled” engagement that will more often than not fail. Not because of poor planning, resource allocation or budget constraints, but because of a typical and traditional “top-down,” macro-view analysis" "

This is big BPM.

What he calls Business Process Discovery is what I think of as the real-world, emergent approach - what we could think of as little BPM. Driven by the community - "capturing the business process as it exists and is exercised daily in the enterprise, by users on IT systems. " - very well said Alex.

recently, I was in an email thread where we had someone ask about dependency on XYZ.

I think this is an area, where much like the conservation of energy, you can shift the dependencies around - language, tool, framework, operating system, vendor, systems integrator, hoster etc - but at the end of the day you still have dependencies.

You are better off when you are aware of the dependencies - the unknown dependencies are the ones that are going to be expensive to deal with. I would recommend investing on a real options analysis if you are looking to add some rigor to your dependency analysis - just be aware that there your solution always has dependencies.

Harry has been asking me when I am going to start blogging again - thanks for pushing me Harry.

He talks about my big SOA, little SOA post from a long time back - it seems that the nomenclature has taken on a life of its own now - I saw Joe McKendrick refer to this recently as well.

Yeah - how do you eat an elephant again? one bit at a time - yes, we are on the same page.

We have been through this before - http://www.microsoft.com/presspass/features/2006/oct06/10-04SOA.mspx

About 18 months we at Microsoft first started presenting our ideas on bridging SOA and Web 2.0 - way ahead of the industry analysts, and other influentials. http://www.eweek.com/article2/0,1895,1918120,00.asp is an article that Darryl Taft at eWeek wrote based on my keynote at VS Live in 2006.

Recently, I had the privilege of once again meeting with Darryl to discuss the state of this convergence. You can find his excellent article here.