As a general rule, technical people are fascinated by form and structure – whether in technology or elsewhere. Architecture piques the interest whether it's system architecture or building architecture. Today I'm going to talk about two different approaches to solving an architecture problem, why I think one fails and the other succeeds, and the implications that it has for architectural governance.
An interesting architectural challenge facing modern city dwellers is managing to live decently in the tiny spaces that result from modern pressures on housing. This is enough of a problem that many cities are now encouraging the development of sub-300 square feet apartments.
There were a couple of interesting articles outlining different approaches to this problem in the last quarter of 2014:
The first approach is the “Manhattan Micro” loft in New York City at 425 square feet. It's located in a brownstone on the Upper West Side, and the approach has been to build upwards. The bed is on a shelf above the entrance area; the cabinet-size bathroom is under the stairs and the kitchen is under the bed shelf, by the entrance.
All well and good.
The second approach is the “All I Own” house in Madrid. This apartment uses a different approach to the problem; there’s a small space by the louvre windows, but the kitchen, the work area and all storage units are on huge moving shelves hanging from the ceiling. If you want to cook, you move all the shelves to the left side of the stack. When it's time to sleep, you push half the shelves to the left, half to the right, and the bed folds down.
Now, this second approach certainly uses more out-of-the-box thinking. It's an elegant use of space. Would I want to use it? No, and the reason is summarized in one of the comments on the video; “Man. It would really suck if you woke up in the middle of the night and wanted to go to the kitchen and grab a snack.”
Yep, that's it. The issue is that almost every change in use requires pushing around shelves that are up to a ton in weight. It's like having to log out of your laptop and log back in to switch from writing a design to looking up some facts online.
So let's talk about what this can tell us for IT architecture, and in particular, the design of architecture review mechanisms.
Any amount of architectural governance requires a level of overhead compared with zero governance. In my experience, most architects and IT managers still see the value of such an effort. The problem comes when the review process adds so much effort that it becomes unworkable.
The defining issue is; how do you ensure that the overhead is as unobtrusive as possible? Rules of thumb would be:
- minimize the number of interactions
- align the process with any existing working practices
I know that I'd probably be ready to move after two weeks if I lived in the “All I Own” house. The same is true of an architecture practice that added an extra hour onto every busy day with layers of review.
Conclusion: Adding a noticeable amount of extra overhead that you have to perform constantly is a recipe for frustration.