log☇︎
281400+ entries in 0.175s
mod6: This year, the anniversary of the declaration is on Election Day in the US. lol. We should go and stand in places and read this outloud like they used to do after the signing in 1776.
Framedragger: *to include the new ssh_openpgp_diff_2016-07-13.tar is what i meant
Framedragger: asciilifeform: and signature (updated checksums file to include the new ssh_openpgp_all_2016-06-20.tar): http://95.85.10.71:8000/all/openpgp/SHA256SUMS.txt
mircea_popescu: http://deedbot.org/deed-2014-11-08-18-41-54.txt << other tyhings i found
Framedragger: (oh, and 1. will sign a checksum at some point (soon); and 2. the tarball contains three compressed files which expand into three dirs, with ~600k files per dir; one file = one openpgp key, same as last time)
Framedragger off to town
Framedragger: this concludes the ipv4 ssh key scan (the new keys are due to re-scan + the previously-excluded hetzner hosting ip ranges). i may rescan in a couple of weeks or a month to see how many new etc (and in general it would be a good and interesting exercise, etc.) some kind of writeup will follow...eventually
Framedragger: asciilifeform: i have moar ssh keys for ya: http://95.85.10.71:8000/all/openpgp/ssh_openpgp_diff_2016-07-13.tar (diff from that 10M bunch, i.e. only new ones) 920M file, 1.82M new ssh keys. ☟︎
jurov: trinque yes that sounds right
jurov: iirc it was in first deedbot incarnation and i had to pester the deeds.b-a second deedbot incarnation to reupload it
mircea_popescu: oh right was the contract
jurov: no it's not, tresurer contract was separate deed
trinque: Framedragger: myup, deeds.bitcoin-assets.com no longer points to deedbot.org
Framedragger: oh the sweet naivete and woes of internet's ephemeral aspects: has anyone (re)read the bitcoin foundations declaration of independence recently? http://thebitcoin.foundation/declaration.txt that link tied to taxation of bitcoin profits does not work anymore, for example. i guess the declaration's main matter was in spirit anyway
a111: Logged on 2016-07-13 16:04 mircea_popescu: http://btcbase.org/log/2016-07-13#1502490 << actually im sure linus would fucking love it, and ~the only reason he's not forcing it is that he doesn't grok it exists.
Framedragger: http://btcbase.org/log/2016-07-13#1502525 << hmm ok.. even his workflow wouldn't differ much now that i think of it.. he reviews patches sent via email etc. and he sure as fuck uses his wot fiercely. so, yeah.. ☝︎
a111: Logged on 2016-07-13 16:12 mircea_popescu: 90% of it goes into "getting a guy there to do it" and 1% or so into "cartridge price tag"
asciilifeform: but http://btcbase.org/log/2016-07-13#1502551 is a point, and it is why folks can't resist the desire to build gigantic cannons - expands the circle of 'there to do it' ☝︎
asciilifeform: afaik there are no ferrous metals in the prototype, nor do there need to be
mircea_popescu: rifle gets residue from the propellant. rail gun is magnetic, gets all residue from environment, also creates new residue out of solids.
asciilifeform: but in usg implementation, even ordinary rifle, recall, needs 50x the servicing anybody would expect.
mircea_popescu: more cleanning than with normal rifle.
mircea_popescu: then the capacitors need replacing, the diodes are burned out and the whole thing's gunked over.
asciilifeform: half the point was, no barrel, no cleaning, etc
mircea_popescu: asciilifeform yes. they are insanely expensive. you're thinking "oh, it's like a gun, hence railgun". no, it's not. your gun fires a magazine, then needs cleaning, then fires another and will do 10k cycles or more. the rail gun fires one thing, needs servicing, and will do maybe 3 ? 5 ? 12 ?
mircea_popescu: 90% of it goes into "getting a guy there to do it" and 1% or so into "cartridge price tag" ☟︎
mircea_popescu: but, in all war, the cost of ammo is mostly a cost of opportunity.
asciilifeform: most targets (e.g., sampans) weren't worth a torpedo.
asciilifeform: from pov of usg, rocket suffers from ww2 torpedo problem:
asciilifeform: bullet is cheap. hence the lure.
mircea_popescu: hence rocket rather than bullet.
mircea_popescu: so you're better off hiding the energy in mass than in speed.
mircea_popescu: just because they couldn't in 1920 doesn't mean they can't today.
asciilifeform: and rocket is every bit as subject to drag as bullet.
asciilifeform: actually it is because of the Hochdruckpumpe problem (ain't no barrel thick enough for icbm 'bullet') but also limit on g-force for complicated mechanism like nuke.
mircea_popescu: that's the only projectile of unlimited energy : a rocket. because it slides under the door of air friction.
mircea_popescu: which is why nukes are rockets rather than bullets, be those bullets accelerated chemically, magnetically or whichever other way.
a111: Logged on 2016-07-13 15:47 asciilifeform: ;;later tell mircea_popescu unlike any type of cannon, railgun imposes no limit on projectile energy. that is the whole reason for it.
mircea_popescu: http://btcbase.org/log/2016-07-13#1502520 << eh, in "theory", which is to say "in the peculiar fictitious narrative some people jack off to". the limiting factor for both is drag, and it ain't going away. "oh but what if we're shooting spherical chickens in a vacuum" "you're still better off with a rocket stfu". ☝︎
a111: Logged on 2016-07-13 15:38 asciilifeform: (why do 1,001 sound cards need to exist?)
mircea_popescu: http://btcbase.org/log/2016-07-13#1502503 << for the ~same reason 10`001 "approaches to interpreting the wot graph" must exist. ☝︎
mircea_popescu: and the fuckwit parade will consider it tantamount to the bad touch.
a111: Logged on 2016-07-13 15:34 Framedragger: watching linux kernel dev be managed with v may, i imagine, be a sight of entertainment, but then something tells me alf et al. are against such monolithic codebases in the first place
mircea_popescu: http://btcbase.org/log/2016-07-13#1502490 << actually im sure linus would fucking love it, and ~the only reason he's not forcing it is that he doesn't grok it exists. ☝︎☟︎
mats: mircea_popescu, mod6: thanks for your contribution to the courts project!
asciilifeform: ;;later tell mircea_popescu unlike any type of cannon, railgun imposes no limit on projectile energy. that is the whole reason for it. ☟︎
a111: Logged on 2016-07-13 04:47 mircea_popescu: http://btcbase.org/log/2016-07-13#1502308 << to put numbers on this : as it recently came out in debate with alf, diesel generators get ~40% efficiency if large enough (closer to my guess than yours! ha-HA!) ; depending on the coil design the resulting electricity will be transferred to projectile somewhere around 1 to 5% efficiency. to go higher air-cored assemblages are excluded because of the high reluctance of the damned
asciilifeform: ;;later tell mircea_popescu http://btcbase.org/log/2016-07-13#1502353 << efficiency has 0 to do with it! calculate, for lulz, the efficiency of making gunpowder and then launching a projectile with it. (it is in single-digit %, at best.) big fat whoop that the gunpowder-making happens on land. the real reason for railgun is illustrated by the failure of the german Hochdruckpumpe cannon. ☝︎
Framedragger: admirable architecture, that
Framedragger: hmh i guess they can...
asciilifeform: rings have 0 to do with it. in pc arch, every device on the bus can issue dma cycles and clobber ram.
asciilifeform: because they are all part of the kernel.
asciilifeform: which is to say, i can overwrite any byte in physical ram from within a driver.
Framedragger: i mean, why not maintain them separately etc
asciilifeform: on x86, or any other platform with dma, all drivers are 'part of the kernel' whether honestly stated or not.
Framedragger: well, it's hard to disagree, on second thought.
Framedragger: linus chose to do the whole 'driver is part of kernel' thing, yeah.
asciilifeform: (why do 1,001 sound cards need to exist?) ☟︎☟︎
asciilifeform: and most of it is only there because of hardware idiocy.
Framedragger: i guess the fact that you need to arrive at a single kernel executable file does not hinder this (cf. all those exokernel projects, not that exokernel architecture is the right approach)
asciilifeform: the problem is that there is more than ~10,000 lines of it. ☟︎
asciilifeform: linux is already largely broken into modules. that is not the problem.
Framedragger: so in the case of an os kernel, you'd just make sure to break down the source into codebase modules so to speak?
asciilifeform: there is no actual problem that warrants a system of that complexity.
asciilifeform: it is idiocy, just like the idea of a continent-sized office building is
asciilifeform: but i am against code bases of that scale, for any reason whatsoever
asciilifeform: not only this,
Framedragger: watching linux kernel dev be managed with v may, i imagine, be a sight of entertainment, but then something tells me alf et al. are against such monolithic codebases in the first place ☟︎
asciilifeform: ;;later tell Framedragger http://cascadianhacker.com/blog/2016/02/07_v-tronics-101-a-gentle-introduction-to-the-most-serene-republic-of-bitcoins-cryptographically-backed-version-control-system.html << ben_vulpes wrote a spiffy intro to v.
asciilifeform: consider: you wouldn't sign a contract that has convenient blank pages for the counterparty to add whatever the fuck he wants into , later
asciilifeform: the 'let's use git without central repo, and pass around signed diffs' thing was suggested ages ago, and i wrote v specifically to shoot it in the head; because signed standard diffs, that can be applied 'fuzzily', are evil. v diffs contain hash-on-entry and hash-on-exit, and if either doesn't match, it is a no go.
asciilifeform: nor does the thing force a particular type of crypto signature - any program that functions similarly to pgp, generating a detached signature, can be used.
asciilifeform: can write them himself, or send via carrier pigeon, if he likes
asciilifeform: it is entirely up to the operator how to go about getting v patches onto his box.
asciilifeform: Framedragger et al: consider, there is no concept of a 'repo' in v. nor of pulling, nor pushing, nor merging. in fact the thing (as properly implemented) doesn't make any use of network at all.
a111: Logged on 2016-07-13 11:48 Framedragger: this may sound like bait, but.... could you not just use signed commits in git, and enforce use of those commits, and use `git blame`? is the argument against this the usual "git is a shitload of code, best to start anew [and also without existing political associations re. github etc.]", or sth more than that?
asciilifeform: http://btcbase.org/log/2016-07-13#1502442 << the resemblance to git is wholly superficial. this probably deserves an article.. ☝︎
a111: Logged on 2016-07-13 06:01 deedbot: [Recent Phuctorings.] Phuctored: 230213975110617197067377 divides RSA Moduli belonging to 'Stephen Domorod III (Stephen at Domorod dot Org) <stephen@domorod.org>; ' - http://phuctor.nosuchlabs.com/gpgkey/CD894142C4313393598D2A7B35E2A61D9CD4C6ACC0214D508203C9CDA8A65195
a111: Logged on 2016-07-13 04:55 mircea_popescu: but otherwise - i have here this battery, contained in a thin PET cover, ready to do ~1,2 MJ at my command. go, construct battery, compressed coil, box of covenant or lavey's teeth assemblage that can do a mj and fit in a gallon.
mircea_popescu: yeah. just, i'm generally too lazy for non-verbal arts.
Framedragger: and prolly retink them laters as you relisten etc
Framedragger: 'cause you do have to really follow and think about the concepts used in the thing if you want to enjoy it beyond the superficial-level "wow such pretty noise"
Framedragger: heh. i listened to a weirdo almost-atonal musical piece yesterday by a friend who's doing sonology / electronic music (yeah such precise terms, i know). it was fkin great, lots of effort but worth it methinks. but then, the point is that this also counts as 'interpretation', possibly
deedbot: [Trilema] The humiliation intensifies - http://trilema.com/2016/the-humiliation-intensifies/
mircea_popescu: interpretation being just about the only art worth the time.
Framedragger: heh, yeah, i see what you mean. trilema allows for more elaborate.. structures
mircea_popescu: trilema's even worse, there i have the time to elaborately bait people. here i'm stuck with much shorter intervals.
mircea_popescu: makes perfect sense, which is why the process of searching etc is worthwhile.
a111: Logged on 2016-05-20 15:53 mircea_popescu: Framedragger actually, the process of creating github and then confusing git for github, aka embrace and extinguish / "rms-ing" is also amply discussed in teh logs :)
Framedragger: mircea_popescu: http://btcbase.org/log/2016-05-20#1469423 fwiw but it's now much. i'm now convinced i had originally misinterpreted. at that point i was assuming that you had basically thrown out git in your mind as useless as it had been tainted with too strong an association with github. if that even makes sense. ☝︎
mircea_popescu: Framedragger "" gives literal matches, otherwise words are or'd together.
mircea_popescu: it proved to be hands down the #1 most productive learning technique in the harem over the years. i expect it works universally.
Framedragger: how do i search for multiple words associated by OR, not AND? $s works for this but btcbase's search doesn't, right?
mircea_popescu is always keen to discuss later interpretations of what he actually said at some point.
shinohai: Don't worry, v and t make github obsolete
Framedragger: regarding the association: previously i got the impression that you thought git and github are now being confused and hence best to get rid of git and start anew anyway. maybe wrong impression
mircea_popescu: in such a situation we sometimes decide to hammer a source into usability (such as, bitcoin, eulora, etc). but just as often decide to write de novo. how exactly this decision is made is not exactly specified.
mircea_popescu: Framedragger as to your words, the political association of git is at best with linus & the linux folk, i dun see what it has to do with github anymore than sunsets have to do with bad photography. but as to the actual issue : on one hand there's a ton of code there ; on the other there's absent functionality.
Framedragger: this may sound like bait, but.... could you not just use signed commits in git, and enforce use of those commits, and use `git blame`? is the argument against this the usual "git is a shitload of code, best to start anew [and also without existing political associations re. github etc.]", or sth more than that? ☟︎
shinohai: V'ify all the things