Taste testing yesterday's production test batches. These were roasted against plans adjusted in CRUCS (I copied in the milestones from the initial exploratory roast and then made adjustments intended to get something easier to do with a larger batch size and based on the initial sensory observations) and the results are quite nice.
Or at least that's what I would be working on, but the cat says no so I guess I'll have to finish that up later. A nice side effect is that the settings fields will start out displaying their current values instead of whatever the browser feels like inappropriately auto-filling (Firefox wants to do crazy things when I change the forms).
Working on adding settings persistence to CRUCS using Window.localStorage. For example, I have graph defaults set to 20 minutes by 500F which is a nice safe default for most of the roasting machines still running, but there are also a lot of faster machines being operated by people who don't need a duration that long or people who prefer °C. They can change their preferred settings once and CRUCS will keep the updated settings. Settings (or anything else) are never sent back to the server.
Looks like I might be seeing a trend on to ship order size again. Prior to taking orders online people would tend to place larger orders so the shipping cost wasn't too bad (I charge whatever USPS charges me). I got online ordering set up a bit before COVID started and average order size went way down (number of orders increased). Over the past month average order size is back up to what it used to be. Haven't looked into this any deeper to gain insight into why, just thought it was interesting.
Room reservations for the event I'm going to have been a lot more hassle than it should be. Resort web site and front desk chat were completely useless. It looks like I might have to make arrangements to stay somewhere else on my last night, but I'll see if the email contact I was given comes up with something useful. Car rental reservation also had a going sideways moment.
That seems like a pattern in my work. I don't have the time, budget, or energy to do everything so I try to get the ideas to people who are in a position to make things better, that doesn''t happen, and eventually I get sufficiently frustrated at the state of things that I roll up my sleeves and build the thing myself. Sometimes there's some uptake after that, sometimes not, but at least I end up with better tools for myself.
Anyway, very excited about this. It's already a lot nicer to use than anything I've seen so far (at least for my particular needs) and there's a lot of space for some seriously powerful features that I've been advocating for a long time without much success at convincing others to write this so I didn't have to.
Last night I got through a little more server config and got a good chunk of a user guide written for CRUCS. There's a little more setup that needs to happen around being able to take money from anybody who wants to pay for it and a bit of work on the main application to finish up the initial feature set, and then I want to put together some tutorials covering different use cases to walk people through that. Of course, I also want people to be able to just play with it and figure it out.
Author of Typica software for coffee roasters.