These are a little messy looking because they just got weeded: carrots are pretty picky about having enough space: you can pretty easily lose 20+ percent yield (and/or take that much longer) if they’re too weedy. We have such nice sandy loam here: even after 16 years it’s still fun to be able to grow root vegetables and have them just work. In Pennsylvania we were on heavy clay and… yeah, forget it. A year or two before we moved we had one small patch that we dug ridiculous amounts of sand and compost into and got some good carrots, but it took a lot of work.
I also spent a bunch of years here thinking carrots weren’t worth it as a market crop, because you don’t get much per pound, but then we tried it and it turns out if we do really well we can get several hundred pounds from a 60-foot by 30-inch bed. And they store well. So it turns out to actually be a great crop.
Today I was reading the 2006 paper Purposes, concepts, misfits, and a redesign of git. It feels like a mostly pretty common-sense method but I don’t think I’ve seen all these pieces put together in such a clear approachable way. It’s only about 20 pages long. It has me thinking that I should try to be more intentional about my design and actually write things down to try and make things more… accountable? Verifiable? Testable, maybe. I have often informally looked through forums and things to try and see if I’m solving real problems. But keeping clearer records feels like a good thing for a lot of reasons. Showing other people my reasoning and the evidence that I’m not just making things up in a vacuum, and so on. I haven’t read the author’s other paper Towards a Theory of Conceptual Design for Software yet. I feel like I got the basic idea from the Git paper. But the theory paper is even shorter than the Git one, and I do plan to read it, maybe tomorrow. I expect there’ll be more interesting details of the method in there.