hnrj95

“Afterwards search will be linear along child[0]”

Yes, I thought this was clear from my statement.

“Rotation would not make a difference”

It would. The offsets generated by the hash would repeat every 32 shifts, but the absolute addresses given in the collision cases are a random construction based on the history of the tree at that point, so despite the offsets’ repeating, the tree’s invariants along the lookup are likely to be preserved.

Great article, but slight issue. The iteration expression shifts left by 2, effectively limiting the height of the tree to 32. You probably want a rotation instead; otherwise you’ll be locked into child 0 from depth 32 down. I suppose with a good hash the manifestation would be rare.

i’m not sure this is correct, but to each his own

surprised there’s no mention of kdb+ in here. iirc it’s older and more performant than most of these

fwiw, and surely anecdotal, but i don’t know anybody who knows the dark arts of the git cli and uses some client (magit, lazygit, etc) who prefers to use the former

i’d argue that you can’t properly define probability without notions in measure theory, which is obviously far too advanced for a high school student. i’m not an educator, but some middle ground needs to be struck. i think it’s clear to many that the quality of education in american colleges far exceeds the quality of education in the average middle or high school. that’s the issue, imo

try org roam

are there benchmarks against kdb+ and/or shakti?

axler’s text is phenomenal, but it’s probably not what you’re looking for if you want an “applied” view into computational techniques gleaned from linear algebra. the text centers on finite dimensional vector spaces, in the standard, mathy, axiomatic way—which is far more general than the prototypical numerical usage in standard programming problems in most swe jobs

i’m not vegan, but the egg industry is probably one of the most atrocious concerning animal welfare iiuc. they do slaughter chicks in droves—just not for meat

well deserved. prot is a fantastic person and his emacs packages are all stellar. can’t use any theme besides those in modus, for example

+1 for eglot. it just works. vibes well with core emacs (and therefore minimizes bloat, among other advantages)

shameless plug for the church of emacs, but my productivity as a programmer has increased vastly from two things: org mode, which to those unaware, is kind of like a super powered literate programming system, and magit, an unrivaled git client. there are other things, eg consult, marginalia, vertico, orderless, etc. the ecosystem is rich, and dev is very active. also, if you are a common lisp programmer, there is no ide-like setup that comes close to slime/sly, in my opinion, for any language/ide pair

can’t say i’ve seen a better title recently

anyone use the pentel orenz? i’ve used the kuru for a while and love it

+1 for karnataka style filter coffee

i actually think icomplete-vertical is great—except one thing. the rotation of candidates i find more visually disturbing than what vertico or selectrum does. +100x for modus’ being amazing. cheers to prot on his great work

lsp mode is great, but i prefer eglot. both are definitely worth looking into

i’d recommend you try a model f. despite the model m’s quality, it was actually made as a cheap alternative to the model f. the f is basically an m on steroids. i don’t prefer it to the hhkb for 2 reasons: (1) it’s ridiculously heavy, and i take my keyboard around with me (2) the layout doesn’t suit my workflow very well

100%. i’ve tried everything. model m, model f, cherry, etc. nothing feels the same way the hhkb does for me. and the layout, as an emacs user, is sublime

not sure how familiar you are with south indian cuisine, but the machine we use to make dosa batter also serves reasonably well for this purpose

as much as i dig protonmail, the rate at which they bring features from web -> mobile is staggeringly slow. it’s a bit frustrating, honestly

notmuch is great, too

i agree. the mental overhead is too much for me to justify, personally, for my own machine

it’s worth mentioning that there’s been great work wrt emacs and its packages very recently. it seems somewhat revitalized, i think

as mentioned in the article, blink is a remarkable tool. it really is indispensable for any dev work on an ipad. i absolutely love it. worth every cent

i can’t tell if this is satire, but just for explicitness: this is nowhere near any level of acceptable mathematical rigor

is it supposed to safeguard against dns leaks? in my cursory tests, it looks like it doesn’t

i agree. working on your own when learning a language seems like a great way to pick up bad habits, especially if you’re moving paradigms. eg going from c to go probably isn’t that dangerous, but moving from java to ocaml is likely quite dangerous

mullvad is absolutely excellent. they’ve definitely stuck to their guns, too