When my role changed from developing to architecting I put a fair amount of thought into what I thought would make a good architect. My goal was instead of saying something was impossible to explain to the business folk what it would cost them in terms of money, people, and trade-offs. The philosophy is that technology should not be a barrier to the products and experiences that the business wants to create.
I’ve now come to a point where either I am completely incapable of articulating costs, or the business folks just don’t care or believe me. It is not a fun position to be in to explain to someone “doing X means horrible thing Y occurs” and they choose X anyway. If Y has no direct impact on the individual’s goals then it is understandable why they might want to proceed anyway. The typical case for this is when the technology is not designed to deliver the experience the product needs, which leaves me in a tough spot. We either:
- pay the (often prohibitive) cost of doing it correctly
- cram the square peg in the round hole
- come to a compromise
- don’t do it
The first two are (by the philosophy) successes, while options 2 and 3 are the more popular choices. Option 2 has costs as well, but they are the external types of costs I described above: the code becomes unmaintainable; the ability to deliver features will slow to a crawl; the complexity will eat us alive.
On the other hand, is there a practical difference between saying “that’s impossible” and proposing consequences of proceeding so dire that no reasonable person would proceed? It seems like the former would be significantly less frustrating and yield better results, but shameful at the same time.
This post is brought to you by late night rambling and sleep deprivation