a-dub

44 ms·

661.
by a-dub 4y ago
this is cool. throwing firecracker at CI is something i've been thinking about since i first read about firecracker. i was thinking more along the lines of, can you checkpoint a bunch of common initialization and startup and then massi
662.
by a-dub 4y ago
expected value makes some sense when you're sampling a lot, but basically none for one offs. personally i think if you're making big bets/one-offs, you probably should be computing something like minimax trees for all the po
663.
by a-dub 4y ago
ironically, freebsd was late to the multiprocessor party, with a global kernel mode lock that existed well into the mid-2000s.
664.
by a-dub 4y ago
it can play MULTIPLE VIDEOS AT THE SAME TIME (!)
665.
by a-dub 4y ago
this was nicely written. it's amusing to see the name of the blog, a popular column in pc magazine in the '90s bore the same name. i forget who wrote it, but it was one of the really well known computer journalists of the time, if
666.
by a-dub 4y ago
was just recently wondering what happened to blogrolls...
667.
by a-dub 4y ago
> He goes on to describe the murder as the sort of thing that would happen "up on the Embarcadero in Frisco by the wise guys," a phrase no one from San Francisco has ever said. definitely had me lol'n
668.
by a-dub 4y ago
it amuses me greatly that 30 years in to the great commercialized internet experiment, we still don't have durable mechanisms for publishing, spam filtering, identification, reputation, moderation and discovery.
669.
by a-dub 4y ago
interesting read. i was a bit confused at first, thinking that the malicious package names were being inserted into cut and pasted pip/apt/etc install strings. i see now that's not happening in this scenario, but curious if t
670.
by a-dub 4y ago
knowing your opponent's hypothetical moves exactly seems like cheating to me! what would be interesting would be an agent that bluffs (and punches above its weight because so), since so many game playing ais assume that their opponent
671.
by a-dub 4y ago
of course they are going after unofficial distribution channels, it's signal, their whole schtick is owning everything end to end so that people can trust their brand.
672.
by a-dub 4y ago
this will increase the "productivity" economic indicator.
673.
by a-dub 4y ago
i don't think the use cases for c and c++ have changed all that much in the last decade. the sheer quantity of application stuff that has been written in very high level languages has.
674.
by a-dub 4y ago
p50 standard of living seems like the obvious one to me.
675.
by a-dub 4y ago
i think that's right. perl is ridiculously easy to learn, it just hasn't been as popular in the stack overflow era. also, i think there's a lot of very old and rotted code out there now that is probably very confusing to try
676.
by a-dub 4y ago
spdy for local rpc. interesting idea. seems the easiest way to get it out there would be support for it in grpc?
677.
by a-dub 4y ago
i'm not sure if i buy his argument that the "delta is not big enough." i have some experience with realtime ai systems and i've noticed something interesting about them. they have a non-smooth capability curve, where the
678.
by a-dub 4y ago
i didn't say it was a bastion of communication for human connection, i said it (the internet) represents "a change in the fabric of how humans communicate and organize." the number of edges per node in the walk in the woods w
679.
by a-dub 4y ago
spiking neural networks are interesting though, and new computational substrates that allow for experiments at larger scales could produce some interesting results. today's sum n' squash (sometimes not even squash) graph networks
680.
by a-dub 4y ago
content moderation in any form is policy, is political, is governance. they're talking about setting up a council with diverse viewpoints, that's an attempt at representative government! they already support blocklists and most of
681.
by a-dub 4y ago
but what about "full self moderating?!" jokes aside. there actually could be some opportunity to do some really cool stuff like some kind of experimental political science. like, mechanisms for population clustering and distributi
682.
by a-dub 4y ago
oh right. "let's cut billions in research and technology development funding, and murder own european financial capitol status because brussels tells us we have to throw out our crap produce before it's expired" please.
683.
by a-dub 4y ago
i suspect they will go back, but probably will lose (or willingly give up) their special "you get to keep your own currency" privileges. doing so would probably be a boon for all of europe and the west. and hopefully a history les
684.
by a-dub 4y ago
have a look at julia. as i understand it uses llvm to jit compile a high level language in an interactive interpreter where changes to modules can be made and components hot recompiled (and relinked into) a dynamic runtime. it aims to repl
685.
by a-dub 4y ago
my first job ever did this. qa and customer support were combined which made it the best place to learn all the ins and outs of a very complicated product. after 3-6 months i had excellent knowledge of how the product actually functioned on
686.
by a-dub 4y ago
it's interesting to me that anyone would try to consider absolute results over the last few years. I feel like there should be some kind of consumer behavior/overall market beta factor that normalizes results with respect to the e
687.
by a-dub 4y ago
when did they add this?!?
688.
by a-dub 4y ago
so where does the machine code for the TUI live? does it get compiled into your binary? how big is it?
689.
by a-dub 4y ago
boomer: kids these days. gen-x: whatever. millennial: do not ever call me a millennial. zoomer: what a mess. the olds are kinda crazy.
690.
by a-dub 4y ago
> I think you've badly misunderstood what's going on here. that sounds right. been almost a decade since i've dug into sha-2. i do remember the lengths in sha-256 and sha-512 though and i remember not being a huge fan of t

More ›