How random should things be?

Jun. 3, 2023



I’ve managed to extract my music library, including all the albums I’ve added to streaming, as a CSV file and write a routine in R to select an album at random. The plan is to write about that album for the blog in roughly the time that it takes to listen to that album all the way through. I already did this yesterday for Goldfrapp’s Black Cherry.

But I have to level with you. That was not the first album that was picked. Nor is the album selection process entirely unbiased. I got to a working situation and stopped there. Essentially, I smoosh together artist and album title, so that albums like compilations, or albums where the lead artist collaborates with multiple other artists, are slightly over-represented. I say slightly as the pool was already pretty large with just unique albums. I might do what I did yesterday and ignore singles or compilations, especially as the code takes almost zero time to give me another idea.

I’ve experimented with this sort of idea before. I have a big list of about thirty or so ’things to write about’. (It’s actually thirty one, because I once was such an optimist that I thought I could write something every day about something totally different.) Selecting from that list at random does bring up repetitions pretty often, it’s a fair while before you get through all of your list. Some simulations are probably in order, I’ll do that when I get some time.

Then I thought about how interesting blog posts are often about the mix and match of ideas, making unusual crossovers. I investigated picking two of my topics so that the writing suggestion would be “write about (main topic) from the point of view of (secondary topic)”. This often produces interesting results, but some are non-starters. You definitely wouldn’t want to be bound by that approach. However, it would be quite interesting to pick two albums at random from my library now and then, the enforced comparison (even if it takes place briefly in my head) might spark a post about a different comparison that’s more valid, or even given me a way in to how to write about one of the albums.

I don’t think I can write about a random album every day forever but I might try it for a bit. My fear is that I have a limited way of writing about music and often fall back too much on how personal memories generate my response to the music. In some sense, that’s what my understated classics sequence is for, but made for very specific choices. However, I do want to write more, and this is a pretty low stakes way of writing more often. It’s also a good way to keep my touch typing skills sharp.

This post is part of me addressing items number eight and number fifteen from the list I offered up earlier this year. Even just a few months later, I am not as certain about that list as I was. I wrote it more as an idealistic way of breaking the funk I’ve found myself in. At least this random idea is a way of trying to do that.

As for the benefits of randomness, I think it is a way of shrugging off bias (or at least trying to) and pointing myself toward the lesser visited spots in my library. And it is my library at least, so I should in theory like what is there. And if I don’t, at least that might make for an interesting post too. I know I have loads of stuff (especially since streaming) that I have barely listened to or not at all. As with books, I sometimes dip in a little and never get the chance to come back later. At least this approach might increase those chances.

I’ve also tried an app that generates notifications at random from a list. The list was meant to be prompts me to write various posts but I have never managed to write a single post prompted by the app. This is likely because I only gave it five rather abstruse suggestions. It also always seems to catch me at the wrong time. I will have to think about how that could be made to work more effectively. If you’re interested, the app is called Yapp and I recommend it because even though I didn’t act on its suggestions, the app itself did everything I wanted to do and it was free (and ad-free!).

Back to home page