Here is a compelling reason not use pie charts when visualizing data:
One of your stakeholders has an MBA and has read exactly one book on data visualization (you know which one). He thinks pie charts are oh so terrible, and if your reports and dashboards contain pie charts, instead of focusing on the data being communicated, he will complain about the pie charts. Because your goal as somebody making charts is to faciliate understanding, avoiding pie charts altogether can actually be a smart strategy. For all their merits, you never scrictly "need" pies (or any one chart type). If chart's form would be a distraction from the data, it is maybe not the the right chart.
Fuck that guy.
mynewfavoritesong.bluesky.social
When I signed up for Val Town, I had a first project in mind. While Spotify's algorithms do a serviceable job notifying me of new music by artists I like, it's not really possible to "follow" a songwriter. Helpfully, Spotify maintains Written By... playlists for many songwriters. Here's one for Jonnali Mikaela "Noonie Bao" Parmenius, who has co-writing credits on two of my absolute favorite pop songs of the last 15 years.
But these first-party playlists aren't in chronological order, and I don't know of a good way being notified when songs get added. So it's quite hard to make sure I don't miss out on the latest credit from Sabrina Carpenter's go-to Amy Allen.
So the plan was to build a cron job val that used the Spotify API to poll a playlist on a regular basis, and check whether anything had changed since the last time, and then create some sort of alert upon detection of a change. It seemed buildable, even with my modest dev skills. I was able to use some tools that Steve from Val Town built. I was ready to request a playlist for the first time, at which point I learned that Spotify somewhat recently enshittified their web API such that first-party playlists are no longer available.
So, I'm still looking for a good way to keep up with songwriters.
But I was able to take the same basic idea and turn in into My New Favorite Song Bot (here's the val with the code). It's the same thing really: a cron val looking for new songs added to a playlist. When it catches something new, it posts to Bluesky. I think it's going to break when I add too many songs. I also haven't really thought about what happens if I try to added multiple songs at the same time.
Because another Val Town user named Jonas Merlin had already made a Bluesky bot template, the whole process was quite tractable. It's been working so far! Every time I've added a song, the Bluesky post has dutifully showed up.
I'm a little ambivalent about what having a Favorite Songs bot will impact my relationship to music. And I haven't really decided how good a song needs to be for me to add it to the playlist. Or what consitutes "new". There are seven songs so far; I seem to be on a pace of about one song a week. Also Spotify is maybe evil, and even if it isn't, it is still probably bad for the world and I shouldn't use it and you shouldn't either. But I do, and you can listen to the full playlist here.
The object of "Is It A Sandwich?" is to just sort of have a conversation.
Decide who goes first or just start talking.
A player proposes a candidate. Every player begins discussing whether or not that candidate is a sandwich. After some debate, players make a judgment that Yes the candidate is a sandwich or No, it is not a sandwich. Players are allowed to change their mind as debate unfolds. The round ends when a player proposes a new candidate to debate.
(Note: although the round ends when someone proposes a new candidate, players may revisit the candidates from previous rounds at any point to reopen debate or change their judgment.)
The game ends when the food arrives. There is no winner. There are no losers.
"I should blog more" is a thought I have pretty regularly. Having ideas worth sharing and crafting in them into something that others might find interesting has always been a very rewarding experience.
But the thing is: Writing is really hard.
And I'm not feeling too optimistic that it's about to get any easier for me. Or even that I have ideas worth sharing. Which means posts here will a combination of infrequent and of poor quality. But I'm trying again. Here is my space for blogging.
You should probably blog more, too! Writing time is time well spent.
(This site was built in Val Town. Thanks to Steve from Val Town for getting me started.)