Todd Biske: Outside the Box

I was listening to David Linthicum’s latest podcast in my car, and it just struck me the wrong way. I literally was yelling, “no, no, no” and wanted to beat my head against the steering wheel. This particular podcast was related to this post by Dave. In the podcast, he also made plenty of references to this post from Dion Hinchcliffe. The podcast, these blogs posts, and others by Dana Gardner and Joe McKendrick, are all about this notion of WOA- Web Oriented Architecture. The reason that it hit me the wrong way is that it just screamed out to me as another case of an over-hyped technology driven approach that tries to give the impression that it will make everything better. I just don’t feel that’s the case. Now, since Dave reads this blog, I should point out that he does have a track record of also railing against enterprises and vendors just throwing technology at problems, so don’t hold any of this against him. This is simply how I perceived the podcast, and just as people have perceived my posts in ways that I hadn’t intended, I’m sure my reaction to this podcast may be a bit of a surprise to Dave.

For the record, I don’t have any issues with SaaS, REST, WOA, Web 2.0, or anything else. What I do have issue with, however, is thinking that the use of any of these technologies (and I may as well include SOAP, WS-*, and all of those in the mix) will heal all that is wrong with large enterprise IT. If an organization chooses to Salesforce.com CRM on demand instead of SAP or Oracle CRM deployed within their firewall, does anything really change? Yes, there’s no doubt that there are potential benefits as far as getting CRM up and running goes, but then what? In the enterprises I’ve worked with, the bulk of the projects were not about implementing some new vendor package. Some of that was always occurring, but there was plenty more that was about integration, enhancements, and other development activities. As we’ve already seen in the SOA space, defining SOA as strictly a technical thing limits it to just another distributed computing technology that provides incremental improvements, at best.

If SOA is instead looked upon as a new way of organizing the assets and the organization, now we’re talking about dramatic change. We’re talking about breaking down silos and managing dependencies on boundaries that make business sense rather than boundaries that were arbitrarily defined by projects at hand. When IT can’t get something done in the time the business wants, it’s usually due to poor processes and politics, not due to their use of WS-*, REST, or SaaS. If you choose to go down the SaaS route, or any other offsite approach, it may get that portion done faster, but it also takes a whole bunch of things and makes them someone else’s problem, rather than solving those problems. At that point, you’re now in line with all the rest of their customers, so that thing had better be something that you don’t see as a competitive advantage. This doesn’t mean SaaS or WOA is bad, but what it means is that you need to have IT working properly to make good decisions on how to use SaaS or WOA appropriately. I stated this in a post back in December of 2006. When an organization views itself as a collection of services, it is now in a position to pick those services and determine where a service provider, whether in the traditional sense or in the SaaS/WOA sense, can be leveraged optimally. There are plenty of things that should be someone else’s problem. There are plenty of things that shouldn’t. Making IT work better is a problem that the organization must address, and simply throwing new technologies at it has a proven track record of not fixing it. At the same time, these problems are not going to be fixed in the timeframes that the media would like, so instead, we’re going to have to deal with naysayers when we don’t have the dramatic success stories after a year, two years, or now in the case of SOA, five or more years. It is a difficult proposition, and honestly also one that may actually be a low priority for many organizations. If things are going well, why not treat it as a slow, incremental improvement that we only notice after we look back to where we were 10 years ago? If things are going poorly, however, something needs to change. Make sure you take the time to change the right thing, rather than just putting more lipstick on the technology pig.

5 Responses to “It’s not about the technology!”