Dev Notes 16: Polishing the IMDb Movie Data and Visuals

Dev Notes (DN) document progress towards some larger article. Discussion is preliminary

Today's notes are once again fairly journal-y. While in theory I should have been moving on to other projects today, I spent most of it polishing Thursday's article on the 9-gap in movie ratings. For the first time I attempt to do a bit of forward planning in these dev notes.

Changes to IMDb Movie Article

I flushed out the story a little clearer. This was a lot of editing. I generally think it reads pretty well, though there is a long cycle of small things that would ideally like to do. I also added more examples as well as a page showing the top 100 list.

I created a scatter plot version of the data.

We can start to see the pattern where the extreme ends won't really have 9-gap but there is a pretty large mix of 9-gap and non-9-gap films in the 6 to 8 range.

The scatter was a little hard to wrangle. I went through a few versions of this (a plotly version, an svg version, then the current Altair version). All are a little buggy in their own way. The pitch of Gemini 3 now fully vibecoding things wasn't quite there in this case. I would have to sit down and build it out more manually to get it really clean probably, but I think workable enough currently.

I also worked on the link and cover images for the post. The idea was to try to communicate the idea of the 9-gap in a small image. This was another case where I played with Gemini 3 to try to get it to generate these images. It ended up making something with hallucinated numbers that could work. Still impressive, but still would need more prompt tuning to get something as nice as some online demos that have been making rounds this week.

November Planning

It is now the 22nd, meaning my daily November publishing is nearing the end. Thus I am left to start planning out the next few days of posts.

• 23rd Sunday: Solicit feedback on the IMDb article online and try poke at the wikidata study to see how I can call that done.
• 24th Monday: Something unrelated to the AI/Tech. Also some other work and activities.
• 25th Tuesday: Likely a continuation of that.
• 26th Wednesday: Start on an autonomous vehicle article
  • I want to discuss how autonomous vehicles influence public transit policy. I started a version of this last year. Some aspects of this change with the Trump Administration, but the general thesis is around being careful on starting decade long transit projects that can't be amplified by AVs, as well as to explore programs that offered subsidies for carpooling. This will have a more rhetorical angle which will be a nice change of pace. So far my articles this month have mostly been exploratory.
• 27th Thursday and 28th Friday are mostly taken up by holidays.
  • My loose plan is to try to make one of these a ~2hr post, possibly on nuclear energy, or continuing the AV post.
  • The other can be Part 2 of the IMDb Rating analysis
• 29th Saturday: AV Post Finish
• 30th Sunday: ? Unclear. I guess this can be a slack day.

That's the plan though... My initial loose November plan was slightly frazzled, but hopefully perhaps this one can be a bit better. At the same time, part of the goal of the month was to take a flexible view and get to work on a bunch of projects, so not going to be rigid on planning here.

Conclusions

Once again these notes were a combination of tweaks on old posts, but also for the first time a bit of forward planning.