A Little Reason, Please
I believe the exodus from siloed systems to a web services type environment will inexorably occur. As applications come online, they will advertise their standard output using UDDI or something like it, and their services will be consumed. Some agnostic protocol will be the new HTML, except now systems will consume the content, and then present it to users, rather than browsers consuming markup language and presenting it to users. Eventually all systems will have to understand that protocol.
Groups like OASIS are working to make this a reality. There are a lot of obstacles that need to be dealt with, such as security concerns, evolving XML standards and the like, but I believe it will eventually happen. Probably sooner than later.
How does this apply to the state enterprise? There is talk about our state enterprise architecture. This stuff is the architecture of the future. I believe that we will be building systems based on the principles that I mentioned above. Either we move in that direction now, or we scramble later. Of course, you can argue that it is too bleeding edge right now. You would have a point. But my point is that if we don't think about these things and do what is prudent now, we create more work for ourselves later. We could at least try to implement some REST-ian principles in what we do now.
A few more words on obstacles: State government is a very unique environment. It is not motivated by the market, only by its constituency. Everyone has their own constituency that they serve. Our constituency at ITS seems to be an unnatural combination of our actual customers (the agencies) and legislative regulations. It is not a free market environment, in fact, it can't even be ran as if it were a free market environment, because rates are tied to a bureucratic process rather than to the market. How do we implement what is the beginning of the future when we are subject to this environment which beats down change and innovation?
Here's another issue with constituencies: They expect results. You have to show that you are doing something, or you are out and someone else is in. Well, because results aren't quantified with dollar signs, everything gets out of whack. Results are quantified with the number of programs and applications and initiatives and committees and plans. If we measure stuff like that, is there any question that we get the kind of service that we do?
Sorry for rambling on. Here's the bottom line, in my opinion: Those of like minds need to work together and move the cause of sanity and responsibility forward. Whether it is implementing a new web technology, or chosing contractors for a construction project, Return on Investment should be King.
1:49:49 PM 
|
