It's March Madness. Every TV at Shakes has a different game on. Everyone at work is talking about their brackets. In the spirit, Carfax hosted a quick tournament today on a Pop A Shot style arcade game that they mysteriously put in the main dining area a week or two back.
Once they announced a tournament, I knew I'd be using some of my exercise time to practice during breaks. All this week I was experimenting with different throwing / shooting techniques during breaks.
When I first joined Carfax 2 years ago, they had a free throw competition on the real court outside. I practiced hard for that one but ultimately didn't win or even come close. I practiced free throws regularly for a year after that in preparation for the next, but for whatever reason there was no free throw competition last year. So when this basketball related competition came around, I was hoping to not go home empty handed.
And luckily I succeeded this time!
All in all, I faced 5-6 challengers on my way to winning the small tournament and each battle basically came down to me being more poised than my competitors. I trusted my technique and didn't psych myself out.
The prize? A $50 gift card to Amazon.com. And I splurged and bought a Roomba - a robot vacuum. Obviously it was more than the prize money but I had been investigating these products here and there because I don't want to sweep the floor regularly.
I'm driving to Vienna once again because I'm interviewing two people down there.
Also, regarding PAPIO, I gave it a facelift recently. Now it doesn't look like craigslist.
And I'm redesigning the backend for this app to use DynamoDB more efficiently. Basically the previous structure I had going just doesn't make sense if I want to store new types of data: clips.
I've been spending a lot of time learning the ins-and-outs of DynamoDB. It's a powerful document store that is basically free at my current scale but comes at a cost of upfront design. Basically, you have to really know what you're doing to design one of these things. Document databases are more about focusing on particular access patterns and optimizing around that whereas SQL databases are about having normalized tables of data that get linked together at query run time. While SQL tables provide a much more flexible way of accessing the data, it comes at a cost: it's expensive, hard(er) to integrate with serverless architectures, and does not scale well.
Scale doesn't apply to me currently but the other two are annoying so I'm putting in the time to learn better DynamoDB patterns. I've always used it as a simple key-value like database, but it requires more knowledge to design a table with more relational data.
We had two mixing sessions with Wil Reeves this past week. We've been getting more done in those sessions that we anticipated so we're rolling along and damn near finished. Next Friday should be our final day mixing and mastering for this first record. The other bandmates have been coming up with album designs and researching copyrights. I'm eternally thankful for them doing that because it's not my area of interest or expertise.
With the warmer weather I'll probably start riding my bike to work soon. Going to Rho Engine Room has been pretty good for me but it's kind of out of my way and makes me spend more time in my car than I'd like. If I can get my exercise through simply biking to work and shooting hoops during some breaks, I'll probably do that for a while.
That's all for now folks.