Wrapping up a release isn’t easy

For nearly the past two weeks, I’ve been alternating between two pretty major tasks. The first is the next release of Moon River Milestones.

It’s pretty amazing that in just a few short months, I designed and implemented a product that people are already using, and finding to be quite useful. On one hand, it would be great if I could get a huge amount of publicity for the product. On the other hand, there are quite a few features that are simply lacking that exist in the products being sold by my competitors. Getting all of that publicity on a version that doesn’t quite compete with the big boys really would seem like quite a waste.

My list started at roughly 50 issues to be resolved for this version, and after pushing some of them off to the future, I was left with 37. Of those 37, I have 11 left and half of those are very tightly coupled together into what is essentially one huge piece of functionality. The others are somewhat smaller tasks, amounting to no more than 2-4 hours of work each. Unfortunately, the huge piece of functionality is well… huge. It’s very time consuming to do, fairly complicated, and involves a lot of ‘what if the customer does x’ scenarios. Those are the most complicated problems: determining what the user is going to want to see.

Whenever I get stuck, I alternate back to my other major task, which is finishing the design of my next product code named Aurora. The high level design is mostly completed, and I just finished speaking with a potential client about whether the current design will meet his needs or not. That’s an important step that a lot of people tend to overlook when designing software. Software is only as useful as the problems it solves. In speaking with him, I uncovered a few areas where it could certainly have used some minor modifications, and uncovered some features that were never in the original specification that needed to be.

I have another potential customer whom I’ll be speaking with hopefully later this week to get his feedback. Once that is done, I’ll start filling in some of the more mundane details of the design specification. Once the next version of Milestones is released, work can begin in earnest on Aurora. More on that later.