I'm reminded of an exchange I had with a company that was doing the whole your data lives in our cloud and that lets you share it and I'm over here like most of the time I don't want to share that, but when I do it's called a file and I can attach it to an email or stick it on shared storage or put it on a thumb stick and hand it to someone or make it available on my own web site or... Plus I can use it when your company fails or my Internet is down or there's a DDOS. Files are great. Use them.
Had a small detour for technical support. A growing category of issues that I'm seeing people get frustrated by fundamentally boils down to people not knowing where their files are (because the apps they're using make it too hard to know that). As soon as the file is found outside of the app that hid it, the thing they need to do becomes trivial.
Updated the shop's web site to include the latest arrivals. https://wilsonscoffee.com/
@ekaitz_zarraga I mostly use Kate for programming which does have syntax highlighting, but the code for letting people buy stuff on my shop's site was done in nano over ssh. It worked fine. My own personal preference is to not have the editor take forever to load and not have to fight against auto-anything. Otherwise it doesn't really matter to me and I'll use whatever.
Not sure what changed but I've been noticing more problems with unintended duplicated key presses lately while typing. Pretty sure it's a hardware problem but I've jacked up the repeat delay setting and that seems to have mostly fixed it. Still, it's probably getting closer to time to replace the laptop.
Slower or faster and operator anxiety shoots way up and with that comes a reduction in control accuracy and an increase in mistakes resulting in failure to hit product spec. Faster incoming data rates can still be useful because you can apply well known statistical methods and get cleaner data at human friendly rates ($1 microcontrollers have enough compute these days to do that) but most of what's out there just doesn't really bother.
Based on actual user/usability testing, software for monitoring what's going on in a coffee roaster should be presenting updated data somewhere in the range of 2 to 10 times per second, with 4-5 updates per second being an especially good sweet spot for maximizing operator joy. This is something that computers from 3 decades ago would not have struggled with.
The third is a competing free option that was just never really designed for larger scale use and is missing a ton of features that home roasters don't need but anybody trying to operate with batch to batch consistency should have.
Another is a commercial sponsor of the event that I get to have a little chuckle about every few years when their marking brags about a feature they could have mined from my software more than a decade prior.
One is software from the roaster manufacturer which prompted me to contact them about certain design decisions that can cause safety problems but that also was so slow that I pulled open the electrical box to write down part numbers and look up spec sheets to find that they could be driving their whole UI a lot faster for better user experience.
@fribbledom Can it be done by (date)? Only if you let me get to work instead of redirecting me to other stuff.
Yesterday's coffee delivery includes the return of the coffee with the megalomaniacal frog on the bag, a new coffee with a squirrel on the bag, one with mountains and coffee plant parts, and one labeled for delivery to Ukraine which I hope is just a case of someone stenciled all the bags that way and not a case of prior contract cancelled due to war.
Author of Typica software for coffee roasters.