Forum logs for 13 Jul 2016

Monday, 16 March, Year 12 d.Tr. | Author:
* mircea_popescu wonders at this novel tendency of us military types to call traditional warfare "kinetic". [00:05]
mircea_popescu: what the fuck are these people thinking ? they're seeing themselves as some sort of 17th century naval artillery dudes, firing kugels at one another ? there's nothing kinetic about weapons in use today, they're all thermic. [00:05]
mircea_popescu: even the fucking nukes are thermic weapons. what kinetic ? [00:06]
Joshua-I: rail guns? lol [00:10]
mircea_popescu: firing what, greatcoat buttons ? [00:10]
Joshua-I: Pretty kinetic until they're not [00:10]
mircea_popescu: kinetic usually denotes macroscopic moving items not fucking ions [00:10]
Joshua-I: def. macro? [00:12]
mircea_popescu: well you know, large enough to not have to apply qm. [00:12]
mircea_popescu: in fairness i guess they are researching rail guns to fire bullet-like items [00:12]
mats: it sounds kewl [00:23]
mircea_popescu: i guess that's as good a reason as any [00:23]
mircea_popescu: has the added benefit that it makes me think they have no fucking idea what they're talking about. [00:23]
mats: i wonder about the feasibility of railgun-assisted orbital launches [00:24]
mircea_popescu: mats you're not the first. that thing's been a solution in search of a problem since ~1916. [00:24]
mircea_popescu: it dun work worth a shit, electromagnetism is a self-limiting disease so to speak, but hey. [00:24]
ben_vulpes: not to mention the accelerations involved make it more or less entirely impractical for anything other than raw materials. [00:25]
mircea_popescu: 100g and over was it ? [00:25]
ben_vulpes: at that point, you may as well save on fab and engineering costs and submerge a hydrogen gun right under the water [00:26]
mats: being able to hurl a rock at enormous speeds over the horizon seems like a useful capability [00:26]
mircea_popescu: anyway, iirc us navy launched a breadbox sized item at about 10km/s or somesuch. cost a little more than a littoral com-sunkship. [00:26]
ben_vulpes: bonus, you can point the latter. not so much for either the cannon or railgun. [00:26]
ben_vulpes: mats: sure, but how do rockets not solve this better and more cheaply [00:26]
mats: its a solution to certain tactical problems, but none in the civilian arena [00:26]
mircea_popescu: none in the physical area, more like it. [00:26]
ben_vulpes: the *single best bet* on orbital advances in my lifetime is the mach-5 airbreather. [00:27]
mircea_popescu: not much air in useful orbit is there ? [00:27]
mats: ben_vulpes: well, they tend to be unguided [00:28]
ben_vulpes: mircea_popescu: once it leaves compressible atmo, it cuts over to internal o2 supply fed directly into the *same* combustion chambers. [00:28]
ben_vulpes: ssto [00:28]
mats: but idk anything about rocketry, so i'll stop here [00:28]
ben_vulpes: mats: why would one want an unguided projectile? [00:28]
mircea_popescu: sounds like some pretty intense engineering required. [00:29]
ben_vulpes: mircea_popescu: and how. [00:29]
ben_vulpes: the only new part of the whole system is the precooler to take air at mach5 and cool it without frosting everything over. [00:29]
mats: i suppose the real advantage would be the raw kinetic energy being delivered [00:29]
ben_vulpes: /maybe/ [00:30]
ben_vulpes: drag is proportional to v^2 [00:30]
mats: i wonder how useful it is as a bunker buster [00:30]
ben_vulpes: the further away your target is, the less useful the railgun and the more useful the classic icbm. [00:30]
ben_vulpes: ramjet if you want to stay low. [00:30]
ben_vulpes: (less useful because a) drag losses and b) accuracy) [00:30]
ben_vulpes: but like cold fusion, railguns keep usg contractors in the procurement pipeline. [00:31]
mircea_popescu: http://trilema.com/forum-logs-for-13-jul-2016#2128840 << this is just about not happening. mass conserves, water is stable. whatcha gonna do, add urea, make lsd in flight ? [00:31]
a111: Logged on 2016-07-13 04:29 ben_vulpes: the only new part of the whole system is the precooler to take air at mach5 and cool it without frosting everything over. [00:31]
ben_vulpes: add glycol, drop the freezing point. [00:32]
mats: you've got him now, he's going to talk about rockets for hours [00:32]
mircea_popescu: mats it's very hard to beat plain and simple fire-air bombs. you gotta put electricity through coils to make the projectile go magnetically, electricity which you got from a combustion generator in the first place.you're about five transforms away, might as well just put the same gasoline in a bottle and throw it at the enemy. [00:32]
asciilifeform: iirc attraction of railgun was : shooting at ~incoming~ rockets etc [00:33]
mircea_popescu: ben_vulpes it's a simple problem of mass conservation though. if your compression factor is 1:10, say, and you burn 1 ton of fuel, you then therefore get a cubic kilometer worth of air, which will contain a ton and a half of water, what do you do with it. it doesn't burn, it doesn't go away. nowwut. [00:33]
ben_vulpes: mats: rude [00:33]
asciilifeform: but go, aim [00:33]
mircea_popescu: asciilifeform that's pretty insane seeing how rocket can maneuver. [00:34]
ben_vulpes: mircea_popescu: turns to steam helps with isp [00:34]
ben_vulpes: srsly this is kindergarten rocket chem [00:34]
ben_vulpes: see also peroxide motors [00:34]
mircea_popescu: i never heard of an aspiring rocket. [00:34]
ben_vulpes: "horatio" [00:35]
asciilifeform: ramjet? [00:35]
mircea_popescu: nor for that matter of a water-oxidizer mix [00:35]
ben_vulpes: no silly it's not /water-oxidiser/ [00:35]
ben_vulpes: it's internal h2 plus ambient air [00:35]
mircea_popescu: $google horatio the rocket [00:35]
deedbot: http://www.iwhistory.org.uk/rockets/ << John Dennett: Isle of Wight Rocket Man | http://www.csifiles.com/reviews/csi/no_mans_land.shtml << CSI Files - CSI: Miami--'No Man's Land' | http://www.horatiotodds.com/dining/ << Dining — Horatio Todd's [00:35]
ben_vulpes: $google more things under heaven and earth [00:35]
deedbot: http://www.shakespeare-online.com/quickquotes/quickquotehamletdreamt.html << There are more things in heaven and earth, Horatio | http://www.enotes.com/shakespeare-quotes/there-more-things-heaven-earth-horatio << There are more things in heaven and earth, Horatio - Shakespeare ... | https://www.quora.com/Shakespeare-once-said-There-are-more-things-in-heaven-and-earth-Horatio-than-are-dreamt-of-in-your-philosophy-How-should-skeptics-best-address-this-ques [00:35]
mircea_popescu: ben_vulpes yes yes. the comparison of the engine to the rocket is flawed. because the rocket doesn't aspire [random gunk] from atmosphere. [00:35]
mircea_popescu: atmosphere contains water. this is a problem, because mass. [00:36]
ben_vulpes: mass in rocket exhaust is a /good thing/ ffs. [00:36]
asciilifeform: the n2 is also not good for much [00:36]
mircea_popescu: yes, if you get it to go there. yes. [00:36]
ben_vulpes: the only reason is because von braun was a savage. [00:37]
mircea_popescu: but your problem was that it dun wanna go, it wanna freeze over your assemblages [00:37]
ben_vulpes: it wants to freeze over the intercooler hooked to the h2. [00:37]
ben_vulpes: keep it from doing that, you have cold, high pressure ambient gas you can dump into a steady state reaction chamber. [00:37]
mircea_popescu: werll for that money, keep gas from growing fat after marriage, you got desktop cold fusion. [00:38]
ben_vulpes: and so long as there is o2, it *will* burn. [00:38]
ben_vulpes: so, keep water from precipitating out of the gas coming in, cool it rapidly by dumping the heat into the h2, react h2 and o2 (and yes, some h2o), you have an air-breathing rocket that (yes, nominally) can operate in atmosphere with out consuming onboard o2 all the way up to mach 5. [00:40]
ben_vulpes: leave atmo, cut over to internal fuel supply. [00:40]
mircea_popescu: it's a fine theory, now show me the item. [00:40]
ben_vulpes: if it works (and what, the brits can't do anything, clearly), it's the only improvement with legs over the classic rocket. [00:40]
mircea_popescu: i'll grant that it has more chances of being useful than the coil gun which isn't saying much. [00:41]
ben_vulpes: that's all i'm saying. [00:41]
mircea_popescu: :))) [00:41]
ben_vulpes: high stair! [00:41]
mircea_popescu: btw, you folks know teh epic bill cosby series on parenting ? [00:42]
ben_vulpes: "chocolate cake!" [00:42]
mircea_popescu: right. [00:42]
mircea_popescu: chocolate cake and braindamage. [00:42]
mircea_popescu: also in the news http://www.engrish.com/wp-content/uploads//2015/12/once-it-eats.jpg [00:43]
ben_vulpes: what, for the novice parents, is the braindamage in cosby's routine? [00:45]
* ben_vulpes has not heard the thing since his youth. on vinyl! [00:46]
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 [00:47]
a111: Logged on 2016-07-13 04:32 mircea_popescu: mats it's very hard to beat plain and simple fire-air bombs. you gotta put electricity through coils to make the projectile go magnetically, electricity which you got from a combustion generator in the first place.you're about five transforms away, might as well just put the same gasoline in a bottle and throw it at the enemy. [00:47]
mircea_popescu: thing - and proper cored items are way the fuck too expensive to be destroyed over three firings. [00:47]
mircea_popescu: by and large for every 100 J worth of gas you're putting in, you're getting out ~1J of ~muzzle velocity~. then as ben_vulpes pointed out correctly you lose most of this to air friction (longer range ? MUCH MORE FRICTION!), so you deliver maybe 1kJ or 2 per MJ expended. [00:47]
mircea_popescu: if you just dump the fuel on enemy, you get 500x bang for the buck. eventually you cook him, through whatever bunker. it's why there's no bunkers in the sun. [00:48]
mircea_popescu: ben_vulpes he says the words, "because kids have brain damage". [00:48]
ben_vulpes: aha yes! [00:49]
ben_vulpes: "it's BRAIN DAMAGE!!!1" [00:49]
mircea_popescu: aha. [00:49]
mats: there is the benefit of a nuclear reactor driving the gun [00:49]
mats: i see your point though [00:50]
mircea_popescu: oh and the caper : obviously the coil is a lc oscillator. the remainder 95 to 99% energy doesn't simply disappear. it COMES BACK in the circuit. going... the other way. [00:50]
mircea_popescu: mats yeah, in a post-oil world all sorts of things might make more sense. [00:50]
mats: what do you mean by post-oil [00:51]
mats: something better or scarcity? [00:51]
mircea_popescu: as in, if there's no oil. [00:51]
mircea_popescu: eh, there's not ever going to be something better. [00:52]
mircea_popescu: much like the current "something better" wall materials kids kick holes into aren't better than brick and concrete. even if some claim otherwise. [00:52]
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. [00:55]
mircea_popescu: we'll discuss stability afterwards, as a sign of me not thinking i'll have to. [00:56]
Joshua-I: this? [01:00]
mircea_popescu: it's just a bucket of gasoline. [01:00]
Joshua-I: Must be a big bucket [01:02]
mircea_popescu: one gallon. [01:03]
Joshua-I: Spoke too soon v. surprised [01:03]
ben_vulpes: gasoline's like 40 Mj/kg [01:03]
mircea_popescu: lol like 5 trillion!11 [01:03]
ben_vulpes: today we learn that mircea_popescu has tiny bukkits [01:04]
mircea_popescu: it's a puke bucket. [01:04]
ben_vulpes: ha! [01:04]
Joshua-I: Puke in MJ plz [01:04]
mircea_popescu: i have many types of buckets. [01:04]
ben_vulpes: i witnesesd a feller at burning man receive the moniker of 'princess buckets' one time. [01:04]
mircea_popescu: i even have a 6 ounce, two-man bukkake bukket. [01:05]
Joshua-I: Two man? Sounds too intimate [01:05]
ben_vulpes: i strolled out of my tent at ~9a to him drinking directly from the bottle. said, "i'm drunk! what do you think about that?" "psh, you're nowhere near drunk" [01:05]
mircea_popescu: well, a one man bukkake's just called "i don't want kids" neh ? [01:05]
ben_vulpes: two more hours of that, i leave to go toodle around on vehicles on the desert. come back, he's asleep on the couch, head deep in five gallons of orange. [01:06]
mircea_popescu: ew [01:06]
ben_vulpes: pulls his head out, whines plaintively, "may i have another bucket please?" [01:06]
Joshua-I: Orange? [01:06]
ben_vulpes: "another bucket?! what are you a princess?" [01:06]
Joshua-I: Oh [01:06]
ben_vulpes: "PRINCESS BUCKETS!" [01:06]
mircea_popescu: lmao [01:06]
Joshua-I: My brain is wacked out tn lol [01:06]
mircea_popescu: this sounds just like the shittier fraternity on campus. [01:06]
ben_vulpes: Joshua-I: i did not know that tungsten was psychoactive [01:07]
mircea_popescu: one man's tonite's another man's wolfram [01:07]
ben_vulpes: i didn't *make* him drink [01:07]
ben_vulpes: i just told him he wasn't drunk yet. [01:07]
mircea_popescu: tort law of portland begs to differ! [01:07]
mats: you devil you [01:07]
Joshua-I: Well it sounds pretty tame for burning man [01:08]
ben_vulpes: Joshua-I: well yes, i'm sharing it. [01:08]
mats: mircea_popescu: on the matter of post-oil, i do wonder what the world would look like in 10, 100 years after cold fusion [01:09]
mats: i'm not gonna volunteer, but i think humans wanna get off this rock and thats not a terrible idea [01:09]
mircea_popescu: prolly every cavortess or w/e her name was living in hypercube houses [01:09]
Joshua-I: I hope it looks even more ridiculous than it does now [01:09]
Joshua-I: Absurdist capitalism will dominate! [01:10]
mircea_popescu: doing post-quantum computing reddit. [01:10]
mats: your pessimism is refreshing [01:10]
mircea_popescu: it's not pessimism just because i didn't vote for change! [01:11]
mats: alright alright [01:11]
mircea_popescu: anyway, i'm off to bed. take it lezzy. [01:11]
Joshua-I: peace [01:11]
mats: good idea [01:11]
Joshua-I: I will return to ebbits and curry [01:12]
Joshua-I: http://www.ebbits.net/strip/2013.01.28 [01:13]
BingoBoingo: "Taylor appeared at an arraignment Monday in St. Louis County Circuit Court, telling Judge Joseph Dueker that he wanted to represent himself “pro per.” The legal term to represent oneself is “pro se.” Dueker told him he didn’t know what “pro per” meant and gave Taylor until July 27 to decide how he will be represented." [01:51]
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 [02:01]
mircea_popescu: this guy links the most retarded comics... [07:02]
shinohai: lel [07:02]
shinohai: I just don't get ebbits [07:02]
mircea_popescu: you mean other than "i want to make a comic and don't perceive i should invest into drawing or writing skills, whoever loves me loves for me!" ? [07:04]
mircea_popescu: it's the comic equivalent of a fat ugly woman that doesn't wash or ever shut up. [07:04]
shinohai: http://shop.heavymetal.com/shm/images/278.jpg [07:07]
mircea_popescu: i guess [07:22]
mircea_popescu: mats hanbot http://trilema.com/2013/so-whos-running-the-courts-circus/#comment-118065 cheers! [07:27]
mircea_popescu: $rated mats [07:28]
deedbot: mircea_popescu rated mats 2 at 2015/04/07 21:29:49 << http://trilema.com/2013/so-whos-running-the-courts-circus/#comment-113478 helping people get pogos, stuff. [07:28]
mircea_popescu: coo. [07:28]
Framedragger: http://btcbase.org/log/2016-07-12#1502117 << myeah, maybe it's not even that much of a problem just, lame.. [07:44]
a111: Logged on 2016-07-12 17:49 mircea_popescu: http://btcbase.org/log/2016-07-12#1502056 << this is what you get for sending ustards to college. they "adapt" what they read to the "practical needs" in their environment. problem ? [07:44]
mircea_popescu: quite lame. [07:44]
mircea_popescu: btw you looking into teh v ? [07:46]
Framedragger: mircea_popescu: not yet, and that's due to lack of time, and generally lazy vibes over the summer (a festival or two, reading books etc) i know i should take a careful look at v, and i dun want to botch it, hence terribly slow with it [07:47]
mircea_popescu: no prob. just didn't want for you to get teh idea i just said that to give you busywork or somesuch. the thing's actually important. [07:47]
shinohai: V'ify all the things [07:48]
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? [07:48]
Framedragger: mircea_popescu: yeah, acknowledged [07:48]
mircea_popescu: shinohai for srs, your wp eventually gets a v mode! [07:48]
Framedragger: ^ but re my q, i should first understand v... [07:49]
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. [07:49]
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. [07:50]
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 [07:50]
shinohai: Don't worry, v and t make github obsolete [07:50]
mircea_popescu: link ? [07:50]
* mircea_popescu is always keen to discuss later interpretations of what he actually said at some point. [07:51]
Framedragger: how do i search for multiple words associated by OR, not AND? $s works for this but btcbase's search doesn't, right? [07:51]
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. [07:51]
mircea_popescu: Framedragger "" gives literal matches, otherwise words are or'd together. [07:52]
Framedragger: aok [07:52]
mircea_popescu: http://btcbase.org/log-search?q=fuck+mother << mircea_popescu: "fuck your mother" [07:52]
Framedragger: searchin' [07:52]
mircea_popescu: and in other news, http://67.media.tumblr.com/eaa45c186f01e7a30c279a9234dff8a7/tumblr_nq3j82wF251sw3gf6o3_500.gif [07:56]
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. [07:56]
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 :) [07:56]
mircea_popescu: :) [07:56]
mircea_popescu: makes perfect sense, which is why the process of searching etc is worthwhile. [07:57]
Framedragger: for shizzle [07:58]
mircea_popescu: trilema's even worse, there i have the time to elaborately bait people. here i'm stuck with much shorter intervals. [07:58]
Framedragger: heh, yeah, i see what you mean. trilema allows for more elaborate.. structures [08:00]
mircea_popescu: interpretation being just about the only art worth the time. [08:01]
deedbot: [Trilema] The humiliation intensifies - http://trilema.com/2016/the-humiliation-intensifies/ [08:02]
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 [08:03]
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" [08:03]
Framedragger: and prolly retink them laters as you relisten etc [08:03]
mircea_popescu: yeah. just, i'm generally too lazy for non-verbal arts. [08:58]
danielpbarron: $up Bigbootycutefeet [10:11]
deedbot: Bigbootycutefeet voiced for 30 minutes. [10:11]
asciilifeform: http://btcbase.org/log/2016-07-13#1502371 << sorta why 'future with fusion' would still include synthetic liquid fuels (made, however inefficiently, from atmospheric co2 and h2 from water) [11:09]
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. [11:09]
asciilifeform: http://btcbase.org/log/2016-07-13#1502423 << mirrored crud [11:11]
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 [11:11]
asciilifeform: http://btcbase.org/log/2016-07-13#1502442 << the resemblance to git is wholly superficial. this probably deserves an article.. [11:12]
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? [11:12]
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. [11:16]
asciilifeform: it is entirely up to the operator how to go about getting v patches onto his box. [11:16]
asciilifeform: can write them himself, or send via carrier pigeon, if he likes [11:16]
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. [11:18]
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. [11:19]
asciilifeform: this is quite deliberate. [11:19]
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 [11:20]
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. [11:21]
gribble: The operation succeeded. [11:21]
Framedragger: thanks asciilifeform [11:33]
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 [11:34]
asciilifeform: not only this, [11:34]
asciilifeform: but i am against code bases of that scale, for any reason whatsoever [11:35]
asciilifeform: it is idiocy, just like the idea of a continent-sized office building is [11:35]
asciilifeform: there is no actual problem that warrants a system of that complexity. [11:35]
asciilifeform: it is purely a product of sloth, vanity, and deliberate вредительство. [11:36]
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? [11:37]
asciilifeform: linux is already largely broken into modules. that is not the problem. [11:37]
asciilifeform: the problem is that there is more than ~10,000 lines of it. [11:37]
asciilifeform: PERIOD. [11:37]
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) [11:37]
Framedragger: riiiight [11:38]
asciilifeform: and most of it is only there because of hardware idiocy. [11:38]
asciilifeform: (why do 1,001 sound cards need to exist?) [11:38]
Framedragger: linus chose to do the whole 'driver is part of kernel' thing, yeah. [11:38]
Framedragger: well, it's hard to disagree, on second thought. [11:38]
asciilifeform: on x86, or any other platform with dma, all drivers are 'part of the kernel' whether honestly stated or not. [11:38]
Framedragger: i mean, why not maintain them separately etc [11:39]
asciilifeform: which is to say, i can overwrite any byte in physical ram from within a driver. [11:39]
Framedragger: sure [11:39]
asciilifeform: because they are all part of the kernel. [11:39]
asciilifeform: because idiot iron. [11:39]
Framedragger: ring0 enuf for everyone [11:39]
asciilifeform: rings have 0 to do with it. in pc arch, every device on the bus can issue dma cycles and clobber ram. [11:40]
Framedragger: ah, right i see what you mean [11:41]
Framedragger: hmh i guess they can... [11:41]
Framedragger: admirable architecture, that [11:41]
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. [11:42]
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 [11:42]
gribble: The operation succeeded. [11:42]
asciilifeform: later tell mircea_popescu unlike any type of cannon, railgun imposes no limit on projectile energy. that is the whole reason for it. [11:47]
gribble: The operation succeeded. [11:47]
mats: mircea_popescu, mod6: thanks for your contribution to the courts project! [11:54]
mod6: o7 [11:58]
shinohai: o/ mod6 [12:01]
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. [12:04]
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 [12:04]
mircea_popescu: and the fuckwit parade will consider it tantamount to the bad touch. [12:04]
mircea_popescu: http://btcbase.org/log/2016-07-13#1502503 << for the ~same reason 10`001 "approaches to interpreting the wot graph" must exist. [12:05]
a111: Logged on 2016-07-13 15:38 asciilifeform: (why do 1,001 sound cards need to exist?) [12:05]
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". [12:07]
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. [12:07]
mircea_popescu: which is why nukes are rockets rather than bullets, be those bullets accelerated chemically, magnetically or whichever other way. [12:07]
mircea_popescu: that's the only projectile of unlimited energy : a rocket. because it slides under the door of air friction. [12:08]
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. [12:08]
asciilifeform: and rocket is every bit as subject to drag as bullet. [12:08]
mircea_popescu: phh. [12:08]
mircea_popescu: just because they couldn't in 1920 doesn't mean they can't today. [12:08]
asciilifeform: more so, because it large, carries fuel. [12:09]
mircea_popescu: asciilifeform not every bit, no. because drag goes up ^1/3 with mass and ^3 with speed. [12:09]
mircea_popescu: so you're better off hiding the energy in mass than in speed. [12:09]
mircea_popescu: hence rocket rather than bullet. [12:09]
asciilifeform: bullet is cheap. hence the lure. [12:10]
asciilifeform: fire 1,001 for cost of 1 rocket. [12:10]
mircea_popescu: sure. much like chinese shoes are cheap. [12:10]
mircea_popescu: same lure. [12:10]
asciilifeform: from pov of usg, rocket suffers from ww2 torpedo problem: [12:10]
asciilifeform: most targets (e.g., sampans) weren't worth a torpedo. [12:11]
mircea_popescu: if it's not worth hitting with a ~$1k rocket, it's sure as fuck not worth hitting with a ~200k railgun shot. [12:11]
asciilifeform: 200k?! [12:11]
mircea_popescu: but, in all war, the cost of ammo is mostly a cost of opportunity. [12:11]
mircea_popescu: 90% of it goes into "getting a guy there to do it" and 1% or so into "cartridge price tag" [12:12]
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 ? [12:13]
asciilifeform: half the point was, no barrel, no cleaning, etc [12:13]
mircea_popescu: then the capacitors need replacing, the diodes are burned out and the whole thing's gunked over. [12:13]
mircea_popescu: more cleanning than with normal rifle. [12:13]
asciilifeform: but in usg implementation, even ordinary rifle, recall, needs 50x the servicing anybody would expect. [12:14]
mircea_popescu: rifle gets residue from the propellant. rail gun is magnetic, gets all residue from environment, also creates new residue out of solids. [12:14]
asciilifeform: afaik there are no ferrous metals in the prototype, nor do there need to be [12:22]
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' [12:24]
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" [12:24]
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.. [12:24]
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. [12:24]
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 [13:07]
trinque: Framedragger: myup, deeds.bitcoin-assets.com no longer points to deedbot.org [13:15]
Framedragger: yeah i know (http://log.bitcoin-assets.com//?date=26-06-2016#1451011) [13:19]
mircea_popescu: Framedragger quite. [13:45]
mircea_popescu: http://deeds.bitcoin-assets.com/deed/9ULZPc7yeZ9fQEA1aZ73H6mcv1s2C4gYFAbNTb5urovj << bwahahah [13:46]
mircea_popescu: possibly it was http://deedbot.org/deed-2014-11-02-20-47-08.txt ? [13:47]
jurov: no it's not, tresurer contract was separate deed [14:17]
mircea_popescu: oh right was the contract [14:18]
jurov: iirc it was in first deedbot incarnation and i had to pester the deeds.b-a second deedbot incarnation to reupload it [14:18]
jurov: we are now at fourth deedbot or such? [14:19]
mircea_popescu: http://deedbot.org/deed-2014-11-03-22-31-27.txt ? [14:20]
trinque: jurov remembers nothing except maybe having complained about something or other, and it being very bothersome. [14:21]
trinque: lol [14:21]
jurov: trinque yes that sounds right [14:22]
jurov: and yes, deed-2014-11-03-22-31-27.txt is it [14:22]
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. [14:26]
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 [14:27]
* Framedragger off to town [14:28]
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) [14:33]
mircea_popescu: http://deedbot.org/deed-2014-11-08-18-41-54.txt << other tyhings i found [14:34]
mod6: yeah. [14:35]
ben_vulpes: 2014 [14:36]
ben_vulpes: dag [14:36]
mircea_popescu: oya. [14:36]
mircea_popescu: imagine, 2nd anniversary of independence coming up soonish [14:37]
mod6: not far off! [14:37]
ben_vulpes: awkwardly far from all astronomical cycle dates. [14:37]
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 [14:38]
Framedragger: *to include the new ssh_openpgp_diff_2016-07-13.tar is what i meant [14:39]
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. [14:42]
mircea_popescu: sure. know of any art spaces with hot chicks ? [14:43]
ben_vulpes: notbad mod6 [14:43]
mod6: There's maybe a few congregating spots dedicated to art, et. al. here where I'm at. Probably in most places. [14:44]
mod6: Hot chicks? If one is lucky. [14:44]
mircea_popescu: so pick one and go do a poetry reading [14:44]
mod6: Not a bad idea. [14:45]
shinohai: Hear Ye, Hear Ye [14:51]
mod6: ^ [14:53]
mircea_popescu: speaking of rockets etc, we ever do http://outerzone.co.uk/latest/plan_thumbs.asp ? [17:49]
asciilifeform: mircea_popescu: neato. old-school drawings [17:58]
asciilifeform: reminds me of ancient 'machine guns of the world' 133337 w4r3z rar, with scanz of old ink prints for same [17:59]
phf: that is awesome [18:02]
phf: we've been building kites with my girlfriend, i might pick up one of those plans for next project [18:03]
shinohai: later tell mod6 471a2ce942966a575f00d310ed4a0be4f5f58f57d6d56caaa4bf55d7ea7e72f3f79c36f88fd84bd6f72198b5a841fb8f568c7679940d85d3030afe1fb022d0ec bitcoind [18:03]
gribble: The operation succeeded. [18:03]
asciilifeform: shinohai: you are unlikely to get bitwise-repeatable binary: gcc embeds turds [18:04]
phf: asciilifeform: which reminds me, is that at linker level or assembler? because i wanted to try generating intermediate S files, and if that's where turds are, perhaps stripping them programmatically [18:06]
asciilifeform: iirc linker. [18:06]
shinohai: force of habit. anyway, it built. twice [18:07]
asciilifeform: possibly a properly-built rotor will give same turdwise output on all boxes. [18:07]
phf: right getting identical environment as far as gcc is concerned [18:11]
mircea_popescu: asciilifeform aha, serious collection too. [19:50]
mircea_popescu: phf many of them not really much work beyond "cylindrical kite". light wood structure + wax paper. [19:52]
asciilifeform: in other lulz, https://www.deepdotweb.com/2016/07/13/new-partnership-aims-to-put-an-end-to-child-porn-related-bitcoin-use [20:20]
mircea_popescu: still with this ? [20:20]
asciilifeform: 'Recently, the IWF provided Elliptic a database of bitcoin addresses that they identified with child porn. With this information, Elliptic can identify those illegal activities on the blockchain (bitcoin’s public ledger of transactions). James Smith, CEO of Elliptic said this regarding their involvement: “This is the first time anybody has started identifying these crimes in bitcoin and flagging them up in a system like ours. Thi [20:20]
asciilifeform: s is a great step … towards our goal of getting rid of any sort of illicit activity in bitcoin.”' [20:20]
asciilifeform: apparently it never runs out !1111 [20:20]
mircea_popescu: it's not even worth dignifying with a mixing campaign. [20:20]
asciilifeform: presumably there is a pile of payola available to favoured sons, in this particular flavour. [20:21]
mircea_popescu: me has had plan drawn up to include small donation to verboten list in all txn processed for what, 3years now ? as anti-this pill. [20:21]
mircea_popescu: has yet to actually justify deployment. [20:21]
asciilifeform: find list, fire away, why not [20:22]
mircea_popescu: and in other "venture capital backed entrepreneurial ideas re bitcoin aka blockchain technologees + their respective lamestream media coverage" news, http://67.media.tumblr.com/fdc9fa6c79e206b6caa9c8f1715cb2e8/tumblr_njzceuR9H71tonrdjo10_r1_500.gif [20:22]
asciilifeform: if the blacklist peddlers have half a brain, they distribute the thing as hashes [20:23]
asciilifeform: but hell knows. [20:23]
mircea_popescu: asciilifeform there's relatively few addresses actually in use, you can crack their hash in a minute or two. [20:23]
asciilifeform: tru. [20:23]
mircea_popescu: if they had half a brain they'd give up the pretense and go learn how to do plumbing or something useful more alligned with their intellectual capabilities. [20:24]
mircea_popescu: clearly, they do not have even that half a brain. [20:24]
asciilifeform: it's a 'tin woman' [20:24]
asciilifeform: 'i have a fine portrait of the pharaoh machined from pure shitanium' 'have $1m' [20:24]
asciilifeform: and, quite possibly, they envision it turning into $100m when 'made mandatory for legit exchanges' etc [20:25]
mircea_popescu: sure, sure, and then microsoft can buy it for a trillion and then etcetera. [20:26]
mircea_popescu: if not for hope, how would the jam-tomorrow socialism exist. [20:27]
asciilifeform: hoap. [20:27]
mircea_popescu: hoap & jew soap. [20:27]
mircea_popescu: june 15th : "nato runs massive baltic exercise, with little russian meddling" july 6th : "putin's military buildup in the baltic stokes invasion fears". [21:09]
mircea_popescu: teh lady doth protest too much, when the fuck did the "modern democracies" bloc turn into such a fucking lapdog ? [21:10]
BingoBoingo: bc,stats [21:10]
gribble: Current Blocks: 420625 | Current Difficulty: 2.133989253313239E11 | Next Difficulty At Block: 421343 | Next Difficulty In: 718 blocks | Next Difficulty In About: 5 days, 15 hours, 41 minutes, and 6 seconds | Next Difficulty Estimate: None | Estimated Percent Change: None [21:10]
BingoBoingo: ticker --market all [21:10]
gribble: Bitstamp BTCUSD last: 656.26, vol: 6512.33113135 | BTC-E BTCUSD last: 645.002, vol: 4973.66621 | Bitfinex BTCUSD last: 655.49, vol: 17213.09171584 | BTCChina BTCUSD last: 654.59771, vol: 23276.70420000 | Kraken BTCUSD last: 655.92, vol: 929.06917542 | Volume-weighted last average: 654.213758389 [21:10]
BingoBoingo: So, surveying the neighborhood after the latest round of wind thinned the tree canopy was in the right place and time to see powerline arc and catch fire [21:17]
BingoBoingo: http://btcbase.org/log/2016-07-14#1502631 << But drain/waste/vent side. Dun want those monkeys playing with natural gas and I'll do my own supply side tyvm. [21:19]
a111: Logged on 2016-07-14 00:24 mircea_popescu: if they had half a brain they'd give up the pretense and go learn how to do plumbing or something useful more alligned with their intellectual capabilities. [21:19]
mircea_popescu: sure. [21:19]
BingoBoingo: I find it disturbing how many "people" out in the wild have no idea that natural gas piping is a subset of plumbing [21:23]
BingoBoingo: https://archive.is/pRwfQ [21:35]
mod6: shinohai: hey thanks for giving that a try! [21:38]
mod6: so last week buildroot's site was down (for about a day) and currently the dang sourceforge site doesn't seem to be allowing curl to grab the boost lib. [21:39]
mod6: trinque was nice enough to host all four of the deps that we require on deedbot.org openssl, bdb, boost, and buildroot. [21:40]
mod6: So I've updated the forthcoming V99994 to pull these from deedbot.org as opposed to their respective places out in the tubes. [21:40]
mod6: If anyone wants to test this along with us, let me know. Will provide you a link. Want to ensure that it works well before we deedbot the new build script. [21:41]
mod6: As far as the Makefiles are concerned, we're having good success there (other than the remote sites for the deps), and will probably be making the same deedbot.org updates to that as well. [21:42]
shinohai: built thrice mod6 Deb, gentoo, and Arch [21:42]
mod6: great work! i build mine on gentoo, and seemed to work good this AM. [21:42]
mod6: or maybe that was lastnight. either way. [21:42]
mod6: 65 [21:42]
mod6: ^5 [21:42]
mod6: Will provide further updates on the makefiles project as I have them. Stay tuned. [21:43]
mod6: Salud! [21:43]
shinohai: if my router dies ill b back online whenever i whip comcast into shape. [21:46]
mod6: give 'em hell [21:52]
mircea_popescu: mod6 did you check/sign the copies ? [21:58]
mod6: they are not signed at this time. the build script in its current form, will pull the dep from deedbot.org and check the SHA512 against what is hardcoded into the script. if it matches, we continue, if not we die. [21:59]
mod6: I could create a clearsigned manifest that could reside on deedbot.org that could be also pulled down, verified and used. [22:00]
mircea_popescu: hm [22:00]
mircea_popescu: yeah what im thinking is, since this "we gotta import crap" thing is going to continue, might as well put some sort of deed process into it [22:01]
mod6: I'm a bit hesitant to "sign" a file outright that I don't have carnal knowledge of -- say openssl - at least without a disclaimer that says "I am only confirming the SHA512 of this artifact is ABCDEF1234... This does not mean that I have read that code and it ``fits in head''." [22:01]
mircea_popescu: hence why it'd be a deed rather than a v diff. [22:01]
mod6: So was thinking a clearsigned manifest could do the trick there. [22:01]
mod6: So, a clearsigned manifest that holds the URL and the SHA512 that I attest is correct then, deedbotted? [22:02]
mircea_popescu: yeah, it would in this instance, but it'll become unmanageable in short order. because it's not just one such item [22:02]
mircea_popescu: what i'm thinking is : the binary/payload in question, base64'd, deedbotted, and the build script modified to take an optional parameter to "allow deedbot import from known signatures" and then it can have a $ifdef for "buildoot"="deedbot.soandso" [22:03]
mircea_popescu: and it knows that if the flag is on, it goes to where deed so and so is and checks it, debases it, unzips it etc. [22:03]
mircea_popescu: make any sense ? [22:04]
mircea_popescu: could have a standard disclaimer up top, have it ignore #s or w/e. [22:04]
mod6: one caveat here, I want this to be the last release of the build script -- so I don't wanna do any heavy lifting here. Would rather put such effort into the makefiles instead. [22:04]
mircea_popescu: makefiles also works yes. i'm thinking more in the mid term than for the next version necessarily. [22:05]
mod6: Makefiles will also solve alf's complaint about "shouldn't pull these from the web at all." [22:05]
mircea_popescu: at least this'd allow some basis for proper management of this mess, rather than current adhocness [22:05]
mircea_popescu: (i'm not saying you're making a mess, i'm just saying - we're stuck with all this grandfathered in bullshit, such as boost, openssh, who the fuck knows what else even. qt ffs.) [22:06]
BingoBoingo: From the mines, a respite https://i.imgur.com/OrGKuRv.jpg [22:06]
mod6: mircea_popescu: yeah, for sure. i reading though the above here... [22:07]
mod6: i think you have the same idea as i've been pondering... [22:07]
mod6: let me see if I can make sense of it and say it back: [22:07]
mircea_popescu: aha! [22:07]
mircea_popescu: well it's just sense i say1 [22:07]
mod6: we make base64 encodings of the artifacts. we deedbot those, they are signed. [22:07]
mircea_popescu: right [22:08]
mod6: a person wants to build, presumably with the makefiles, but could be a build script... they pass the builder a flag that says "get these deps for me automatically", at which time, it'll go and fetch the deed(s) and verify them, decode them, and place them in the correct place and continue building. [22:09]
mod6: Or not, and they place the deps in there by hand, on their own. Then continue building. [22:09]
mircea_popescu: only if signed by people you trust the sigs of [22:09]
mod6: Ok, so .wot directory is required here. [22:09]
mircea_popescu: ya [22:09]
mircea_popescu: since we got it, use it neh ? [22:09]
mod6: Yeah, it certainly has to be there. [22:10]
mod6: for the buildscript and the makefiles. so that makes sense. [22:10]
mod6: trinque: is there a max length for deeds? (just curious about the above ^, some of these might be mightly large) [22:11]
mircea_popescu: an you put a number on that ? [22:12]
mod6: mircea_popescu: thanks for your input here, its high time that we get away from the adhoc nature of this build process. [22:12]
mod6: sure, lemme see here... [22:12]
mircea_popescu: slowly but wsurely you know ? [22:12]
mircea_popescu: mod6 specifically, iirc it's tens of mb not gb. amirite ? [22:12]
mircea_popescu: and in other ethedao news, today toelik butt-erin releases DA CONE! https://media.giphy.com/media/hYXKE2SGWmYIU/giphy.gif [22:13]
mod6: $ wc -l boost_1_52_0.tar.bz2.base64 [22:15]
mod6: 954767 boost_1_52_0.tar.bz2.base64 [22:15]
mod6: $ du -sh boost_1_52_0.tar.bz2.base64 [22:15]
mod6: 71M boost_1_52_0.tar.bz2.base64 [22:15]
mod6: $ du -sh boost_1_52_0.tar.bz2 [22:15]
mod6: 52M boost_1_52_0.tar.bz2 [22:15]
mod6: <+mircea_popescu> mod6 specifically, iirc it's tens of mb not gb. amirite ? << Yes, Sir. [22:16]
mod6: So hopefully that'll be just fine. [22:19]
mod6: <+mircea_popescu> at least this'd allow some basis for proper management of this mess, rather than current adhocness << the nice part about the makefiles and your preposed solution is that we would finally have a solution in place for all of this. [22:22]
mircea_popescu: ya not the end of the world. pretty big though. see what trinqwue says [22:22]
mircea_popescu: mod6 aha [22:22]
mod6: it really has been like herding cats to get all of the build infrastructure into place. [22:23]
mircea_popescu: i bet [22:23]
mod6: it'll be nice to be able to move forward from this part. [22:23]
mircea_popescu: but hoenstly seems to be actually coming together [22:23]
mod6: it is. slowly, but surely. [22:24]
mod6: which isn't a bad thing i'd rather do it slow & careful, than fast and reckless. [22:24]
mircea_popescu: ya [22:24]
mod6: building up tons of technical debt that I (or someone) has to fix eventually anyway. [22:24]
mod6: let's fast-forward for a minute and pretend that your solution is implemented. [22:25]
mod6: and the makefiles are complete. [22:25]
mircea_popescu: ok [22:26]
mod6: this really leaves one last question that has to be thought through before we move on Do we place the makefiles into trb's tree, or do we place them in their own tree? I've thought about this quite a bit. And there are tradeoffs to both. [22:26]
mircea_popescu: it seems to me they should be in trb tree. which tradeoffs do you see ? [22:27]
mod6: The one staring me in the face is this: If we place them in trb's tree, we'll be adding vpatch after vpatch to the source base to update the makefiles as they require to be updated. Having them in their own tree could elimiate cruft in trb. [22:27]
mod6: The thing is, they /are/ really a part of trb itself. [22:28]
mircea_popescu: this isn't properly called cruft tho. [22:28]
mod6: true, not actual cruft no, just changes. [22:28]
mircea_popescu: if item x, whatever it may be, can only be used to do Y then it belongs in Y's tree [22:28]
mircea_popescu: 1 thing, 1 tree. [22:29]
mod6: I started there for sure. Once I actually did this, it works, no problem with that. Was just trying to think, "what will the tree look like in 10 years, 20 years?" [22:29]
asciilifeform: http://btcbase.org/log/2016-07-14#1502685 << what has mircea_popescu been smoking?? there is no qt in trb [22:30]
a111: Logged on 2016-07-14 02:06 mircea_popescu: (i'm not saying you're making a mess, i'm just saying - we're stuck with all this grandfathered in bullshit, such as boost, openssh, who the fuck knows what else even. qt ffs.) [22:30]
mod6: Anyway, got some time to mull it over. :] [22:30]
mircea_popescu: eventually people will build management trees on top of this. much like phf's "make deedbot lines print out gray". [22:30]
asciilifeform: the deps are strictly: 1) gcc 2) some libc (musl works ok) 3) boost 4) openssl 5) bdb. [22:30]
mod6: I tend to agree, overall though. It should belong with trb. [22:30]
mircea_popescu: "make makefile changes print out blue" or w/e [22:30]
mircea_popescu: asciilifeform just sayin'! [22:30]
mircea_popescu: at some point, there actually was, iirc [22:31]
mircea_popescu: or was it the xwidgets w/e that thing is called ? [22:31]
asciilifeform: before the great cleansing [22:31]
mircea_popescu: i was making a point omaigerd. [22:31]
asciilifeform: my point was that we actually have a handle on ~all~ the deps [22:31]
asciilifeform: in the sense of enumeration. [22:31]
mircea_popescu: yes, we do. [22:32]
asciilifeform: http://btcbase.org/log/2016-07-14#1502694 << i'd make the downloader, if there must be one, a separate script. [22:33]
a111: Logged on 2016-07-14 02:09 mod6: a person wants to build, presumably with the makefiles, but could be a build script... they pass the builder a flag that says "get these deps for me automatically", at which time, it'll go and fetch the deed(s) and verify them, decode them, and place them in the correct place and continue building. [22:33]
asciilifeform: sign that. [22:33]
asciilifeform: this 'here are the turds' thing are not something that ought to become habit. [22:34]
mircea_popescu: meh. becomes opaque. [22:34]
mircea_popescu: "oh, it's the downloader". [22:34]
mod6: i can see this both ways too. i still wrestle with the whole "V Mirror" thing too. [22:34]
mod6: its the difference between "easy" and "manual". [22:35]
asciilifeform: it was the 'sign tarballs of liquishit' thing that led me to http://www.loper-os.org/?p=1545 [22:35]
mircea_popescu: yeah well. [22:35]
mod6: mechanically checked and verified, but something i wrestle with none the less. [22:35]
mircea_popescu: look, we can't not sign the things - these asswipes KEEP CHANGING THEM [22:36]
mircea_popescu: that's why i want them stored in deedbot and sealed up. [22:36]
mod6: I do tend to favor that, as opposed to signing my own death warrant. [22:36]
asciilifeform: pretty sure i signed a set of rotor dep hashes a while back. [22:37]
asciilifeform: possibly even deedbotted - though i have not been able to turn up the link [22:37]
mod6: As far as the "V Mirror" thing, that I know alf hates, I move forward on the basis that one doesn't have to use it, even though it is there. This may not be ideal, but it's bridge between both "easy" and "manual" [22:37]
mod6: A compromise. [22:37]
asciilifeform: there is a reason i dun like it [22:37]
mircea_popescu: mod6 which v mirror thing is this ? [22:38]
asciilifeform: my position is that the customary degree of automagic: 'auto-fetch latest widget from central repo!11' - is not something that ought to be kept. [22:38]
mod6: Yes, you believe that one should read and verify these vpatches on their own, place them in by what they say, and who wrote them. Instead of getting a big ball of wax automatically. [22:38]
asciilifeform: it is a psychologically-destructive habit [22:38]
mircea_popescu: asciilifeform gotta cut the branches in the proper order lest you break your teeth. [22:39]
mod6: I tend to agree with you -- but i believe that the entire thing reduces to a "downloader script" outside of V if I remove the mirrors part. [22:39]
asciilifeform: and in many cases, in particular the one now contemplated, is wholly unnecessary: fetching the trb deps is a once-per-machine thing [22:39]
asciilifeform: there is NO reason to download them on every build [22:39]
asciilifeform: you do it when you buy a new disk. that's it. [22:39]
mod6: So i feel like it's one or the other. People need a way to press one button and build -- not everyone is clued. [22:39]
asciilifeform: and ideally, from ~your other disk~ [22:40]
asciilifeform: not from the net. [22:40]
mircea_popescu: this is a point - script should prolly check for local cache first [22:40]
asciilifeform: aha. [22:40]
mircea_popescu: "we only need what wed don't have" [22:40]
mod6: Yeah, the makefiles will do this. It'll check the "shit" folder and what not, and look. If there continue, if not, halt. UNLESS, the -go-get-the-shit flag is passed in the invocation. [22:40]
mod6: Then it'll continue with Mr. P.'s perscription above. [22:41]
mircea_popescu: works. [22:41]
mod6: To me, this seems to satisfy the requirements we've been kicking around for a while. [22:42]
mod6: asciilifeform: agreed? [22:43]
mircea_popescu: yeah definite progress here. [22:43]
mircea_popescu: the fence is quasi complete [22:43]
asciilifeform: definitely improvement over 'calls curl on my box with no nic' [22:43]
mod6: fair enough, Sir. [22:43]
asciilifeform: also i have this notion that we already had this, in some draft version, and i lost it. [22:44]
mod6: Thanks for both of your input this evening. [22:44]
mircea_popescu: shoulda deedbotted it [22:44]
mircea_popescu: since you're too poor to have blog [22:44]
mircea_popescu: :D [22:44]
asciilifeform: l0l aha [22:44]
mod6: asciilifeform: i believe that perhaps your stator script did something of the like. [22:44]
mod6: or rotor? i'd have to look. [22:44]
asciilifeform: nah, it just expected them to be there [22:44]
asciilifeform: i never had auto-loader of anything iirc. [22:44]
mod6: oh thats right, it just expected you to do this manually. which i automated. [22:45]
mod6: got it. [22:45]
mod6: 'tis been a while. [22:45]
mod6: Anyway, this is great! I aught to write this up and put this into the ticket notes. [22:45]
mircea_popescu: *thumbsup* [22:46]
mod6: %h [22:59]
tb0t: <%a|%add> <project> <code> "<subject>" "<notes>" [ante] | <%e|%edit> <project> <id> <code> "<subject>" "<notes>" [ante] | <%r|%remove> <project> <id> | <%p|%print> <project> <id> | <%pp|%print_projects> | <%pc|%print_codes> [22:59]
mod6: %p trb 12 [23:00]
tb0t: Project: trb, ID: 12, Type: F, Subject: Makefiles for building full orchastra, Antecedents: , Notes: [23:00]
mod6: %e trb 12 F "Makefiles for building full orchastra" "http://btcbase.org/log/2016-07-14#1502667" [23:00]
a111: Logged on 2016-07-14 01:58 mircea_popescu: mod6 did you check/sign the copies ? [23:00]
mod6: %p trb 12 [23:01]
tb0t: Project: trb, ID: 12, Type: F, Subject: Makefiles for building full orchastra, Antecedents: , Notes: [23:01]
mod6: dafaq [23:02]
mod6: %p trb 12 [23:05]
tb0t: Project: trb, ID: 12, Type: F, Subject: Makefiles for building full orchastra, Antecedents: , Notes: http://btcbase.org/log/2016-07-14#1502667 [23:05]
a111: Logged on 2016-07-14 01:58 mircea_popescu: mod6 did you check/sign the copies ? [23:05]
mod6: weird. i'll have to look into that. [23:05]
deedbot: [Qntra] Windows Print Spooler Vulnerability Spent 20 Years Unaddressed - http://qntra.net/2016/07/windows-print-spooler-vulnerability-spent-20-years-unaddressed/ [23:41]
deedbot: [» Contravex: A blog by Pete Dushenski] Modernists: Fixing self-made problems since 1964. And before. - http://www.contravex.com/2016/07/13/modernists-fixing-self-made-problems-since-1964-and-before/ [23:45]
Category: Logs
Comments feed : RSS 2.0. Leave your own comment below, or send a trackback.
Add your cents! »
    If this is your first comment, it will wait to be approved. This usually takes a few hours. Subsequent comments are not delayed.