Seems we just can't resist hating on anything new, no matter what it is these days. Drew has some good ideas but they're often lost in his reactionary rejection of anything not his narrowly defined vision of UNIXy enough. His inexplicable dislike of containers being just one example.
There can be nuance to our view of new things. It's important we stress our ability as humans to think this way. LLMs are not all or nothing good bad polarities. They can be just another tool in your toolset. And that's fine. No hatred required.
I think he doesn't like the way Unix containers are done, preferring other ways to get them. From that URL:
> Recall that everything really is just a file on Plan 9, unlike Unix. Access to the hardware is provided through normal files, and per-process namespaces do not require special permissions to modify mountpoints. Making a container is thus trivial: just unmount all of the hardware you don’t want the sandboxed program to have access to. Done. You don’t even have to be root.
Where does he express a dislike for the concept of containers? After all, he's working on a new microkernel OS, with a permissions model that seem designed to make containers easy to implement.
Yeah, but they are not being stressed enough or considered at a greater level. The same can be said with stuff such as Global Warming for instance, whereas every other company knows that stuff they're doing is terrible, but don't give a shit and we continue to follow as if nothing is going on. AI >right now< is not doing anything specifically dangerous, BUT people adding that "tool to their toolset" is not us, mere mortals but people who intend to use AI in crappier ways. Thinking about those things is not "shitting on anything new" but to think critically about stuff.
There can be nuance to our view of new things. It's important we stress our ability as humans to think this way. LLMs are not all or nothing good bad polarities. They can be just another tool in your toolset. And that's fine. No hatred required.