jack

I spent a bunch of time upgrading my resource fields. I'm going to be doing more of that. But then I was finally able to get grey and purple science being produced.

I had a set-up like a single looooong line of assemblers with three belts of basic resources, and one belt for "everything else", so I could just add extras on the end (with blueprint to copy-and-paste).

There's a lot of lessons in scaling here. When I'm adding 5-10 new assemblers, it's easy to copy the assemblers and belts, and then fix up the connections between manually. If I was doing a much bigger base, it'd be more important to be able to drop in another identical row without manually fiddling.

Similarly, it was a mistake to mix everything-else assembling and science-research assembling. I had enough assemblers the proportions of the intermediates got all out of whack. I fixed it for a few things by adding "if this item is already on the belt, don't produce more logic", but that's not really sustainable. If I go further, I should probably have a science section which produces only intermediates consumed by the science, and a separate "stuff for me and for building" section which produces only-so-much and then stops.

But I was able to fix this up by manually adding a lot of the inputs that aren't *very* quick to assemble, but I already had lots of in storage, and fixing up anything that was under- or over- producing. And I need to remember that's often true -- some things need to be 100% automated because you need to use them thousands of times without thinking about the details; other things, 90% automated is fine because you can just fix up the details a bit and get the result you need with less work than further making the automation "perfect".

So I got a steady stream of blue, purple, and grey science going to my labs. For a while, I even had 20-ish labs going at once. And I started researching some of the post-blue technologies. I'm not sure what's most useful to have at this point -- the ones I most want are personal fusion reactor for the exoskeleton, requester chests, and rocket silos, but I think those all need yellow which I'll set up next. In the meantime I'm filling the technology out with anything that may be useful: incremental mining and logistic improvements, nuclear power, rocketry, etc.

Personal

I love having something I'm able to get really involved in. Ideally those would be hobbies which produce something useful, not only feel good but don't achieve anything, but it's ok to do that sometimes. But it feels good to have some concentration binges.

I'm really cautious about becoming "oops, spent 8 hours playing computer games", partly because I sense I might be prone to that, but partly, I think, out of unnecessary worry.

But there are positives too. Getting really wrapped up in something does make me feel satisfied -- I think I sleep better and feel less need to snack when I'm not hungry.

And I've had a couple of late nights, but it's good to test my resilience a bit. I used to have late nights all the time, more through, "oh god, I don't want tomorrow, I won't get anything done" than "woo, today". Now I'm sleeping a lot earlier and better (having someone to sleep with is v much better). But as with other things, when I feel like I've established the habit, and then cautiously relax my vigilance, I discover I'm developed habits that seemed normal for many people: of being up late *occasionally*, but being able to absorb it as long as it's not too often.

And as I said, it's produced quite a lot of useful introspection about software engineering :)