Todd Biske: Outside the Box

A recent conversation brought up another real world governance example that I think is an excellent illustration of where we need to get to in our SOA (and IT) governance efforts. The town I live in borders the Mississippi River. In early 2007, the city council signed a master development agreement with an area developer to develop the “bottoms” area, essentially farm land that butts up against the river levees. Shortly after that, elections changed the makeup of that council, and now there is a majority that is against this development and has been doing everything they can to keep things from moving forward, so much so, that the developer has a suit against the city. That’s not integral to this post, however. What has recently been the topic of discussion is the levee system. The Army Corps
of Engineers is going to decertify the levees and require the local municipalities to pay for upgrades. So, the question at hand is, “Who’s going to pay for this?” The master development agreement calls for the developer to pay for the improvements as there’s a vested interest there, since the levees would be protecting the businesses in the development. If that agreement falls apart with the current issues, some alternative method of funding would need to be found. The city and/or county really only have taxes at their disposal. Either sales taxes or property taxes would need to be increased.

So what does this have to do with SOA and IT Governance? Well, look at it this way. The master development agreement is like an IT project in many companies these days. The primary focus is on delivering some new businesses to the area, but as part of that, there’s some infrastructure work that needs to happen. The levee upgrade is a clear infrastructure need, but it’s extremely difficult to find funding for it without bundling it in with some other more tangible effort that is closer to the end user, in this case the residents of the surrounding area.

Now, if this effort were the typical IT project, What would happen? The developer would build their shopping centers, distribution centers, warehouses, whatever, but they would put in only the infrastructure required for that. Imagine seeing a levee that surrounds their development, but as soon as you leave their property, we go back to some crumbling infrastructure which may actually now be worse, because the added strength of the new section puts more pressure on the aging infrastructure. Now, this won’t actually happen, because the Army Corps of Engineers have regulations that will ensure that the levee upgrades are built properly.

What we need to happen in the IT world is to have the same types of regulations so that when projects go and request funding, the scope is such that funding includes building services and infrastructure in the right way, rather than how it is today where the scope is set solely to the immediate needs of the project. The problem is that we don’t have the regulations that can actually set the scope. We instead have regulations that begin with something like, “If you build a shared service…” which immediately gives the project managers or proposers a way out, because they’ll claim that their services aren’t going to be shared, when in fact, they probably simply haven’t looked into whether there is a shared need, simply because of the risk of expanded scope. I think some form of business domain model as I’ve discussed in the past (here and here) that classifies domains of capabilities from the perspective of “share-ability” is a step in the right direction.