Articles for category: Delivery

2 months ago

Greg Thomas

The Loss of Design

If we stopped designing houses, the results would be interesting. We design for a reason, we take that moment to breathe and think about a problem, stare out a window in silence, and contemplate what is happening. Any problem that has been solved well took this time to figure out the design. Yes, it’s not always fast, but it should always be good.

Thinking is Good

We’re being reconditioned right now to not think because it takes too long. I spent 3 hours this past weekend thinking about how to approach some problems, what would and would not work, and running through the scenarios in my head. I’m still running through a coding problem with AI that is encroaching on 5 hours, trying to get something to work that should just work (based on the APIs). Thinking is good, thinking is important, you should plan time to think about solutions to problems and not only barrel through them as fast as you can.    

2 months ago

Greg Thomas

Why You don’t like Sprints

Because they force you to do something you’re not (at least not all of us are accustomed to). Focus and get to the finish line. When you are running, that’s the goal, the finish line – whether the road is bad, it’s raining, there are lots of people – the goal is the same: get to the finish line. And the only way to do that is to focus and move your feet. That focus and that goal are what many sprints are missing when they fail. Get to the goal.

Check Engine Light

Your software needs a “Check Engine” light – it’s getting harder to maintain, it doesn’t work as well as it used to, or it’s hard to understand. Better yet, have an engine light for your team. Does everyone know how our software works and what we do? Do we have a scalable group, and are they growing? It should flash red in your eyes, and like your car make you pause. The problem is we don’t, so we keep driving along until it blows up.  

3 months ago

Greg Thomas

Two Things at Once

You can’t do two things at once, but we’re getting close. I can put Claude to work on a problem while I switch to something else. I’m still not doing two things at once, but it now looks like. But if I have to spend time reviewing and integrating what Claude wrote on a greater scale later on, have I done two things at once? Whether it’s minimal time or not? Are we doing two things at once or are we queuing up future work to review?