The trouble with that notion is this: imagining that a plastic-based ecosystem arises (horrifying thought!) it means that there are life-forms capable of deriving energy from plastics, breaking them down. That makes plastics useless to us humans, because any time we try to use plastics for all the things we currently do with them, those life-forms are going to come along and attack, break down the stuff we deem "useful plastics"; the critters will make no distinction between nurdles lost on the beach and the plastics holding your car/house/clothes/aeroplane together. i.e. It's Game Over for plastics use.
Beer certainly is an odd one! Some (craft) craft beers will age gracefully -- improving, even -- for years. Industrial lagers... not so much.
I scored once with an imported (Belgian) craft beer at the local supermarket, heavily marked down price because the sell-by date was approaching. Some days later, marked down by half again because the date had passed. Needless to say, I bought all I could carry.
Just another vote against Markdown. Markdown is OK for simple docs, but very poor for wiki where inter-page links are the Magic Sauce. Creole markup is a lot better for wikiing imho.
I wouldn't store the file format in the file extension; rather store metadata properly as metadata. Chances are that the application wants to hold a lot more metadata anyway, so you're going to need a metadata storage scheme anyway. (Yes, I am a lone crusader for eliminating metadata from filenames.)
Word of mouth. Chatting with people in my socials, their recommendations/reads, and... blogrolls appear to be making a minor comeback. We're all drowning in the LLM shit, so the only thing we can rely on is each other.
I think the point is that, rather than "requiring everybody on the planet to reevaluate how they interact with a computer", Maggie's idea is that computers should be reevaluated with regard to how they interact with people. Like another person earlier in these comments, I am much more skeptical than her that LLMs are "the answer" to enabling non-professional-programmers to create -- more importantly to modify and compose -- chunks of deeply custom software, but in principle I'm in sympathy with the sentiment. People do care. It's just that almost all software they interact with is deeply brittle, change hostile, black-box. You can change almost nothing about it beyond some superficial tweaks, and composing different pieces of software into a working "something else" is mostly impossible. I suspect that the very notion of an "app" is something we have to get past in order to make progress on this...
What (if any) are the differences, advantages, disadvantages of this implementation over the Guava library's EventBus? From a fairly cursory look it seems much the same...
"Let it rust"? If you live near the sea, cars rust and there's not a lot you can do about it. You can attend to visible rust timeously, but there are plenty of hollow spaces where it can fester for years, all unnoticable, before it pops out through the paintwork.
Trouble with my ('95) Corolla is that it's the body that's showing its age. (Rust, and not the good kind.) The ICE is just fine and probably good for another 1/2-million km, but the whole thing has got to go soon.
I think that at least some of this comes about because it's still relatively early days for the form-factor. As the industry matures, as it becomes more cutthroat everything will become more comodified and therefore standardised.
Look at some of the cars of (say) the late-19thC where not even the steering wheel was standard. So, while e-bikes are probably not quite that early stage right now, they've not advanced terribly far from the plain vanilla bicycle yet.
For a good portion of the world I think you'd have to add, "...at a price roughly equal to [insert large percentage, in places exceeding 100%] your monthly income," to get a more realisic response.