Lets take a moment to talk about the many faces of Activity Pub (this network)
Having written that down, it's actually a little bit surprising how many new machine designs fail on key criteria. Sometimes these deficiencies mean exercises that should be doable in minutes take hours instead. Others mean your very expensive machine is likely to become impossible to repair within a few decades.
Drafted about six pages on how to choose a good machine for learning to roast coffee. It should also be a good read for someone considering opening a roasting company who doesn't know what to look for in a first production roaster. It briefly covers things like controls and instrumentation, a usable trier, batch size, build quality and availability of replacement parts, budget, and training. No mention of specific brands/models because that would be huge and quickly outdated.
re: software gripe
I think this supposed epidemic of software not getting updated at the user end isn't really so much a result of lazy users as it is a result of commercial software updates not being trustworthy.
Look at Windows - updates are opaque, are pushed out without warning, occasionally include adware, and put the system out of action for an inordinately long period of time. It's not poor planning that causes industrial systems to be running Windows XP SP1, it's shoddy software quality that does.
When responsibilities get chopped up to the point that nobody is responsible for anything, problems that have an easy and obvious fix become impossible to solve because nobody has the authority to just do the needful and everything gets hung up because the only person with the authority to do some tiny part of the task got laid off 3 months prior and the person who would have noticed stuff not getting done was eliminated during the merger. But at least it's all documented and CYA'd.
Author of Typica software for coffee roasters.