BizTalk is not dead

The reports of my death are greatly exaggerated. ~Mark Twain (or BizTalk Server)

For the past 5 years or so, various technologists have been predicting or even proclaiming the untimely demise of Microsoft BizTalk Server. Usually these predictions come in the form of blog posts espousing the new hotness whether it was a new product from Microsoft or a competing product claiming it was a better middleware solution than BizTalk. Oddly enough, the chatter ramps up the loudest shortly after Microsoft releases a new CTP of technology X that may be tangentially related to an existing product like BizTalk.

The problem with all of the proclamations is that they never seem to come from the Microsoft Product Development groups or BizTalk MVPs. I am not an MVP, but I do consider myself to be plugged in to what is happening with BizTalk. With that in mind, I would like to briefly take a look at the current state of BizTalk’s life and put an end to the urban legend.

It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so. ~Mark Twain

Many organizations base their decisions of whether or not to use a particular software product upon the chosen vendor’s ability to support the product ongoing. For the people claiming BizTalk’s demise, I would suggest reviewing what we know about the BizTalk Server Support Lifecycle directly from Microsoft’s support site:

Note: Mainstream Support Remaining calculated based on current date of 7/15/2011.

I’ve summarized the tables quite a bit and I’ve added in the results for Windows Server 2008 to highlight a few dates. Do you see what I see? The operating system on which BizTalk is running will fall out of mainstream support well before BizTalk falls out of support. Obviously, Microsoft has been known to adjust the support timelines. The more important point I’m trying to make is 3 years is a very long time regarding technology. What did Windows Azure look like 3 years ago? I had not even heard of it. Windows Server had just been released and many of us were still getting used to using IIS 7. Things change rapidly and part of our job is to keep up and advise those around us based on the current state of the industry as a whole.

Just this past week, Tony Meleg, Senior Technical Product Manager at Microsoft, gave a presentation at the Microsoft Worldwide Partner Conference (#WPC11) entitled ‘The Future of Middleware and BizTalk Roadmap’ http://t.co/Fz1yKbH It was a great presentation. I’m not going to recap all of the highlights, but I will point you to the blog posts of Saravana Kumar http://blogs.digitaldeposit.net/saravana/post/2011/07/15/The-Future-of-Middleware-and-the-BizTalk-Roadmap-Recap.aspx and Richard Seroter http://seroter.wordpress.com/2011/07/14/is-biztalk-server-going-away-at-some-point-yes-dead-nope that do a very nice job of summarizing the presentation. One of the most important takeaways is that Microsoft is taking a ‘cloud first’ delivery model for AppFabric. Windows Azure will see 2 to 3 releases per year and only after several features are baked in will we see them in the on-premise Windows Server AppFabric version. The eyebrow raising thing is that Microsoft is only planning on building one thing: build once, run everywhere. It should be interesting to watch. What is important to recognize is that the on-premise delivery will lag behind the cloud delivery and may take 5 or more years to truly materialize. In that timeframe, we may not see any truly innovative features introduced in subsequent releases of BizTalk, but it is clear that BizTalk must continue to be supported because Microsoft will not have anything to replace it for quite some time.

As Tony showed in his slides, the first version of BizTalk was released on May 2, 2001 – over a decade ago. Since then, BizTalk has single-handedly spearheaded Microsoft’s middleware strategy. Ponder this for just a moment: in the world of technology, what has changed in a decade? Everything. .NET was a young pup, iPhones didn’t exist and many developers were trying to figure out what the heck XML, SOAP and JavaScript were all about. As good as BizTalk has been at being everything to everyone, there does come a time when it makes sense to change how system integration is done and align business needs with the right technology. This is precisely what Microsoft is attempting to achieve and they will ultimately succeed. Over the past year or so, I have had several consulting engagements where I have recommended against using BizTalk because Windows Workflow or a couple of custom WCF Services would suffice. Why? Because BizTalk did not fit and the cost of the licenses could not be justified based on the integration requirements. On the flip side, I have had just as many instances where BizTalk was the only product that could do what the customer needed. I had a suite of technologies within the Microsoft stack to choose from and I was able to mix and match to create the most cost effective solution for my client. That is called progress.

Based on Tony Meleg’s presentation (and I consider him to be a pretty reliable source), it is clear that Microsoft has a very long term vision on delivering Windows Azure and Windows Server AppFabric. So, I think we can safely say that BizTalk is not dead, but change is definitely coming. Isn’t it always? For those of you who may have heavy investments in learning BizTalk and feel like you are already behind the curve, here are a few terms I would like you to read and consider:

Orchestration / Workflow
Service Bus
Publish / Subscribe
Messaging
Services
Routing
Persistence
Hosting

Do those terms sound familiar? They should. Anyone who has ever worked with BizTalk has utilized one or more of the terms listed above. Now for the ‘gotcha’ moment: I actually took the list of words from the AppFabric documentation on the MSDN website. While the developer experience to work with those technical areas may look different in the future, the amount of time that has been spent learning BizTalk is already providing a bridge to the new paradigm. Even if you have not looked at what Windows Azure or Windows Server AppFabric have to offer, you are familiar with the framework based on patterns and practices that you already know.

BizTalk is not going away anytime soon. Don’t do anything hasty like throw away BizTalk books or get that ‘I Love BizTalk’ tattoo taken off just yet. If you are using BizTalk and are not on the 2009 or 2010 versions, I would suggest coming up with a strategy to upgrade to 2010 to maintain product support through 2016. Microsoft has given us a lot of lead time to start looking at the future of their middleware and integration strategies. Take the time now to learn the new terminology, tools and methodologies. Seriously, you have a couple of years, but I would not wait until the zero hour to try it out. As Tony mentioned at WPC, the AppFabric Team at Microsoft is working with the Patterns and Practices Group to develop guidance and best practices in order to help the developers at large hit the ground running without having to figure it all out for themselves. There are many BizTalk MVPs cranking out blog posts, user group presentations and code samples about integrating AppFabric WITH BizTalk. Notice I said ‘WITH’. This is key. Windows Server AppFabric has a long way to go and the delivery dates are slated to be years in the future before it is considered fully baked enough to replace something as powerful as BizTalk in an on-premise flavor.

BizTalk is not dead. Long live BizTalk. ~Mike Diiorio