Software Development: One Step Forward, Two Steps Back..
Tuesday, July 22nd, 2008At the office, we, the Web Services team form a team of four people who handle the core data storage and communications centre for the companies main application. It’s by no means the most complex (it’s extremely simple in fact) or even the most important, but none the less, it forms the glue between each of the client applications and hosts the file/data store for all shared information used throughout the system.
As part of the build process we’ve done a recent refactor which improves performance and security and aims to generally make our service easier to consume by every device that uses it. This process has taken the better part of six weeks next to full time work by four of us and at least two weeks work from a fifth person that left recently. I don’t want to say I’ve poured my heart and soul into it, that would be lieing, but its definitely been something that we’ve become immensely proud of.

- The Software Development Life Cycle
Anyway, today the decision was made to drop all the work completed toward the new version, and continue where we forked off on the previous release, which really is a huge shame. Now I feel like we’ve taken a step backward, erasing a bunch of work that would have ultimately made the project a lot more stable and laid a few of the foundations for future-proofivity, if thats even a word. I shouldn’t get too aread of myself though, in reality it was months away from the final goal, but at least there was move towards that goal.
I shouldn’t be surprised though, there seems to be many examples of this and far worse (mal)practices in the IT industry around the web, and it seems to me to often be the way the things work, especially in smaller Software Development situations. I’ve even seen it before at previous jobs, and its never pretty.

The thing that gets me though, is the absolute waste of resources and man-hours that goes into a mistake such as this. Assuming we worked two thirds of the week each, I’d put the time spent at:
4.5 staff x 27 hours x 6 weeks = 729 man hours
By my crude calculations this is just about enough man hours to build a small skyscraper, or the Titanic if you will, but… That’s Life, after all!
Who knows, in another month, the project will probaly take another path altogether requiring a complete rewrite, refactor or perhaps disbanded altogether!