If you read the colorforth source (there’s only around 1000 instructions of assembly, it’s not a long read), there’s a real sense that much of the design (pretokenization, color tags, etc) are built around the punchline of a single (with rep prefix) instruction being used to linear-search the dictionary — going to trivial hardware-supported data structures here constrains the implementation a bunch, but there’s magic in self-imposed constraints.
Slowing refers to a change in the derivative, in this context. Slowing growth would be a decrease in change in GDP per year — a decrease in growth. But the claim is that the growth (first derivative) is what’s slowing — that is, the second derivative of GDP w.r.t. time went negative, which does seem to be the case in mid 2018 from the linked chart.
There are institutes scattered around the world — “colleges” and “universities” — where people spend a lot of time thinking about what set of knowledge should be considered part of the baseline understanding of a subject. These are published as “syllabi”, “core courses”, “majors”, etc. I recommend finding a college or university you respect, a major that interests you (maybe “computer science” or “software engineering”), looking at the core courses for that major, reading each syllabus and coming up with a strategy to learn the material in it. The colleges and universities even offer help with that last bit!
Always thought that “disc” was the original word for an object of a certain shape. As they evolved for computer storage, we got smaller diskettes… which were abbreviated to disks.
Or, they have freed up time for more useful endeavours, that may otherwise have spent on drudgery.
I don't discount the value of blood, sweat and tears spent on debugging those hard issues, and the lessons learned from doing so, but there is a certain point where it's OK to take a pass and just let the robots figure it out.
What do you see as the alternative here? Conductive epoxy is way less repairable than solder. Sockets are… components; and tend to be more expensive and higher failure rate than what’s socketed in them, except for extreme cases of very large ICs. Press fit requires special tooling, so repairability is much worse… what’s left?
It’s early February. Have you really read so many articles you couldn’t understand in one month that you have a “usual” way of dealing with it? You should consider whether you would benefit from curating your sources better, or if use of AI as a crutch has already decayed your ability to understand stuff on your own unrecoverably…
try curating the hacker news commentary when there are 800+ texts. no, really, what the hell are you talking about? having someone figure the insights that are relevant to oneself, among 800 texts, DOES solve a problem, which otherwise is unsolved unless you do it manually. which, the manual thing, as we all know, does not necessarily result in significantly better insights.
and yes, my job is to read technical slop dusk till dawn, and I care very little who wrote it, but whether it is relevant to my research. its a lot of reading, it causes me pain, so OF COURSE I would love to short cut it somehow, given most of it is slop anyway - no matter if its human or synthetic slop.
> a compiler introducing bugs into code it compiles is a nightmare thankfully few have faced
Is this true? It’s not an everyday thing, but when using less common flags, or code structures, or targets… every few years I run into a codegen issue. It’s hard to imagine going through a career without a handful…
How about we restrict airport and aircraft access based on individual's ability to do harm, rather than on the information in some trusted database? It sure seems like the major incidents in my lifetime would have been better prevented by keeping people with guns and bombs out than people with poor paperwork skills…
reply