I’m looking into software architecture design. How to start the design of a system from scratch that will have all the trimmings (input validation, error tracking, log files, updates, etc) that won’t shoot itself in the foot with bloat and bugs? So far I have only seen these problems solve retroactively: assume it’s not bloated or buggy until it is… then address the problem. I suppose as a program manager I’m a born worrier and I assume the sky is going to fall.
Interesting quote off the following web site:
http://www.lukehohmann.com/books/bizswarch/
“What is especially troubling is … the idea that technical issues can be somehow separated from business issues, or that business issues can be somehow separated from technical issues. At best this is simplistically wrong; at worst it can be a recipe for disaster.”
This is a great talking point, because the case can be argued as follows:
(From the “technology can do anything” school): We have really great examples in the market today of products that do things that computers can do already. They crunch numbers and process tags. The best way to make your product stand out is to consider a market need, then work to fill that need with the technology that fits. This will set you apart from the incremental-progress competition.
(From the “how do you expect us to do that” school): Sure, everyone likes a technical challenge. But a market need does not mean that it is best solved with technology, or even should be solved with technology. If it’s really new, you have to write all new tools and create a whole new platform before even beginning your product development. Often a staffed 800 number will solve the problem just fine. Let’s be realistic.
My current position is somewhere in the middle. I think great things can come from tinkering and exploring the limits of the tools available. Also, we should respect good ideas that appear out of nowhere and have no technical basis. Everyone should carry around a palette of these raw materials with them, and match them up wherever possible.
11:56:15 AM 
|
