Finishing up Findka's grand scheme
Future of Discovery
3 August 2021 – Jacob O’Bryant
Over the weekend we visited LA for my brother-in-law’s wedding reception. First time flying with the baby. She’s seven months old now. Glad to be home.
I mentioned last week that I was planning to focus on marketing/growth for The Sample now. I’ve spent a bit of time on that, but I’ve also started to think that there are at least a few low-effort-high-impact things we could do to improve the product. In particular, I thought of a different way that the recommendation algorithm could handle users’ interest keywords (the checkboxes that people select when they sign up). Planning to have that running in an A/B test sometime today. In the mean time, Marshall is working on some kind of recurrent neural network thingy that we’ll run in another test once it’s ready.
In general, perhaps it would be best to stay focused on product quality now that we have a decently sized user base—enlarge the interest rate rather than the principal.
Over the past several weeks I’ve been laying out my grand scheme for Findka (see the start of last week’s post for the links). There are a couple parts left, however I’ve already written about them in other places and I don’t have much to add now. If you’re interested, read user-centric data, make the web more interoperable and build a career path for software inventors. At some point maybe I’ll write up a post explaining how those things tie in with Findka. Or maybe we’ll just do it and then people will find out that way.
Every time I write this newsletter I think about how my workflow for reading, thinking, and publishing sucks. I think tools are the problem. This has been a big motivation for me to start working on that user-centric data thing lately—I want to make it easier to write little throwaway apps that all work together seamlessly, so I can iterate towards my perfect workflow instead of being stuck with all these monolithic apps that have been optimized for their own business objectives instead of my productivity. (Not trying to throw shade on anyone; I’m doing the same with my business).
Tools for thought and open protocols for connecting ideas across knowledge graphs, independent of tool. Who is working on this? Is it too soon? I want to believe commoditization makes this inevitable, but not sure it does.
@svoisen I’ve experimented in the area. not tools for thought specifically. . current plan is to make something like Solid project but more practical, with Crux DB (graph database)
We need user-centric data
Maybe I should mingle more in the tools-for-thought crowd; however that term usually makes me think of note-taking which I think is overrated. Eh. Probably not doing it justice.
Anyway I’ve decided to stop publishing essays weekly until I get my workflow sorted out. Instead I’ll just write about the stuff I’m working on directly, i.e. “here’s what I did last week”/”here’s what I’m doing next week.” That sounds a little boring, so I’ll probably switch to sending this out twice a month instead of every week.
See ya!
Discuss this on Discord.