Hide Idle (>14 d.) Chans
← 2016-05-11 | 2016-05-13 → ↓
mircea_popescu: $up indoor_jellyfish
deedbot: indoor_jellyfish voiced for 30 minutes.
mircea_popescu: asciilifeform nope.
indoor_jellyfish: Does anyone know of a document that explains the linux boot process in detail? I've lisp code that builds /a/ ResuceCD (http://paste.lisp.org/display/315833) but not /the/ ResuceCD needed for Masamune.
indoor_jellyfish: ie, it the current process doesn't package together the existing libraries / binaries /sources on disk into the bootable USB, and instead uses my re-write of "Roll your own ResueCD" (https://www.phenix.bnl.gov/~purschke/RescueCD/)
indoor_jellyfish: which uses and unpacked archive of binaries from the author of that blog post
indoor_jellyfish: I don't really understand how the C linker or the rest of the linux "toolkit" works (eg, chroot) and a detailed explanation of how I would arrange a set of files + binaries, bundle them and generate a minimal bootable USB would go a long way.
indoor_jellyfish is gabriel_laddel
indoor_jellyfish: without such a document, the masamune replication will get done, but it'll take a while..
gribble: The operation succeeded.
jurov: punkman these look like kernel innards, gabriel was asking more about userspace
jurov: funny that I can't remember any concrete source how I learned that stuff. it somehow accreted
jurov: ;;later tell gabriel_laddel you'll definitely need this: https://wiki.gentoo.org/wiki/Custom_Initramfs
gribble: The operation succeeded.
punkman: jurov, dunno looks like it covers the whole thing
mircea_popescu: jurov> funny that I can't remember any concrete source how I learned that stuff. it somehow accreted <<< thinking about it i dun think a proper source exists, outside of carder forums or w/e equiv.
mircea_popescu: maybe good move for gabriel_laddel to actually post that thing on his blog once he makes it.
mircea_popescu: o hey, paymium now joins bitstamp on the list of wanna-be bitcoin exchanges that email spam.
a111: Logged on 2016-05-12 13:09 mircea_popescu: jurov> funny that I can't remember any concrete source how I learned that stuff. it somehow accreted <<< thinking about it i dun think a proper source exists, outside of carder forums or w/e equiv.
asciilifeform: it is in the logz!
asciilifeform: my pogotron had it all...
mircea_popescu: oh right you are huh.
shinohai: I managed to make my own gentoo livecd a few weeks ago ... I used the roll-your-own initramfs thing like jurov posted earlier
a111: Logged on 2016-05-12 14:04 phf: asciilifeform: http://glyf.org/tmp/foo.jpg
phf: yes
asciilifeform: djvu or it neverhappened (tm) (r)
phf: there's really too much documents for my lowly eos 5d, but i'm going for teh ic schematics and such. it's mostly like "here's how you machine part 560345 front left hinge"
asciilifeform: how many schmidts you had to kill to get these!
mircea_popescu: check him out ?
mircea_popescu: phf care to qntra ? :)
phf: no, this needs further slow movement. i'm still hoping to get the whole set on another trip
phf: like there are cable assemblies, that are handy for repairs or new construction, that i just don't have time to capture
mircea_popescu: what's the constraint ?
phf: he still has open maintenance contracts
phf: while the bulk is software, a dwindling part of it is repairs and console upkeep
phf: the intended purpose is to help lispm owners maintain systems on their own since the number of available parts is basically two stables in a barn and those are going to the usual suspects
phf: and since asciilifeform is a lispm owner i don't see particularly reason why s.nsa shouldn't have access to it
asciilifeform: http://btcbase.org/log/2016-05-12#1466158 << there is one possible boojum that i know of:
a111: Logged on 2016-05-12 14:19 phf: no, this needs further slow movement. i'm still hoping to get the whole set on another trip
asciilifeform: most of the 'good stuff' was GAL listings
asciilifeform: rather than schematics
asciilifeform: as smbx was among the first outfits to use synthesized logic
asciilifeform: i'll be quite surprised if these are to be found in the papers david s releases to machine owners
asciilifeform: and they are what is needed to replicate the thing on fpga.
asciilifeform: (either that or the 'snap4' emulator source code)
phf: that's true, but those are not lost either
mircea_popescu: this is pretty cool.
mircea_popescu: $rated phf
deedbot: mircea_popescu rated phf 3 at 3668926147 << Teh desire has been expressed and heard!
phf: the main constraint is those same open maintenance contracts, but ultimately guy wants all that stuff preserved
mircea_popescu: $rate phf 4 Keeper of the logs ; and of Symbolics history.
mircea_popescu: $v 5B32D3412CCCD6B40CAE1CC564C953BE033C546D99BD1E35C62E449B93BCA1E2
deedbot: mircea_popescu updated rating of phf from 3 to 4 << Keeper of the logs ; and of Symbolics history.
asciilifeform: phf: last thing dks said to me was that the rights to the whole shebang were bought at auction by prof j mallery of mit/nsa.
asciilifeform: or rather, he admitted that they were sold, but did not say to whom, i discovered through other informants
phf: i know that mallery owns the entire software side, no idea where the vlsi falls, but afaiu that's the reason why nobody's doing anything with the assets
phf: but after shuffling "assembly" and "harness" papers for two days i know for a fact that some dedicated japanese man will be able to lovingly carve an smbx replica as his chosen otaku hobby
asciilifeform: phf: the vlsi falls squarely in the software side afaik
asciilifeform: it existed almost entirely as src text
asciilifeform: ditto the synthesis system
asciilifeform: (smbx had this, apparently, before ~anybody else, though today it is sop)
asciilifeform: a smbx box actually contained not 1 but ~3~ computers
asciilifeform: the lispm itself, the 'fep' (which did disk/tape/etc. i/o, and was its own m68k machine) plus another m68k in the console
asciilifeform: it might be possible to recreate the last 2 from the schematics.
phf: there are some funny comments about t.i. people on the margins
asciilifeform: the largest competitor, aha
asciilifeform: (ti sold what was mostly a straight miniature clone of the original mit lispm)
asciilifeform: i will note that there are no unusual cables (other than the long triax console cable) in my lispm
asciilifeform: the internals are mostly on the backplane.
asciilifeform: (disk is on a st-506 cable but the pinout is public and you can buy the cable on, e.g., ebay)
asciilifeform: dc power supply (a monster, iron box in the machine) was made by 3rd party and has labeled screw terminals
asciilifeform: the last true smbx box, xl1200, actually had standard scsi in it.
asciilifeform: (and the 'macivory' product that sat in a 68k mac, was simply a card with the smbx cpu in it... no cables at all)
asciilifeform: at one point i considered buying one, but was not able to swing the access to the needed microscopy
asciilifeform: $up luke-jr
deedbot: luke-jr voiced for 30 minutes.
asciilifeform: what brings you here luke-jr ?
luke-jr: so we're discussing whether we can get consensus for a hardfork with the community here
luke-jr: is there any possibility of that, or is it just impossible?
asciilifeform: luke-jr: what's in your hardfork ?
luke-jr: asciilifeform: not sure yet; ideally, only things that everyone thinks are acceptable (including people here)
asciilifeform: well nobody can answer this mega-question until the concretes are given, neh ?
luke-jr: (unreasonable people demand we support 2 MB old transactions)
asciilifeform: normally folks going hard-forking have some specific idea of why...
luke-jr: asciilifeform: to show the industry that a hardfork and consensus is a possible thing
mircea_popescu: luke-jr do your reading.
mircea_popescu: also, please don't refer to tmsr as "a community". it is not "a community", it is your liege.
shinohai makes popcorn
luke-jr: asciilifeform: things I'd like to see in it would be merged mining, additional inputs to the generation transaction, and maybe fix block withholding
mircea_popescu: $down luke-jr
mircea_popescu: as the romanian expression goes, prostu' parca nu-i destul daca nu e si fudul.
trinque: http://btcbase.org/log/2016-05-11#1465970 << serious wtf; I'll look into it
a111: Logged on 2016-05-11 21:38 mircea_popescu: world of tanks lol. trinque wtf the api can't be that broken ?!
trinque: http://btcbase.org/log/2016-05-12#1466068 << wrote my bot with the intention of making creating new commands very easy
a111: Logged on 2016-05-12 01:39 mod6: trinque: any thoughts here? should we just collaborate on this? phf?
trinque: could be that you expose some api - or SQL - my direction, or otherwise that I somehow sync (rsync?) tickets from you
trinque: http://btcbase.org/log/2016-05-12#1466098 << yeah, "commandbot" is WoTsauce
a111: Logged on 2016-05-12 01:49 mircea_popescu: quite possibly, see what trinque says. running two bots does not necessarily mean one must write two bots.
shinohai: Trying to get all that stuff mioved over to a site we control. I'm actually liking clwiki now, thanks for providing it!
trinque: sweet
shinohai: The bot section will need a whole new writeup though, since everything was merged into deedbot so gotta get a list of all those features compiled.
trinque: $help
trinque: shinohai: deedbot's can link to that, as I'll have that autogenerated from its sauce
shinohai: ah ok will do
PeterL: hi all
PeterL: so, I bought a house, I figured it is a good time to lock up a loan at 3.5% interest for 30 years
PeterL: rates can't go much lower than they are
shinohai: trinque: how do i insert a line break? <br> doesn't work it seems.
shinohai: and now I am getting 502 error as well :/
trinque: yeah I did that
trinque: it's back
shinohai: Good, I didn't break it
trinque: wait one sec
trinque: damned thing doesn't shit errors out anywhere
trinque: so I'll be digging for a minute
shinohai: kk no worries
asciilifeform: 'For instance, Diffie quizzed an MIT colleague named Dan Edwards, who would join the NSA after graduating. "He was extremely unhelpful," Diffie later reported, "failing to reveal things which were certainly not classified and which I later saw in the bibliography of his thesis." And when a colleague at Mitre went to work at IDA, Diffie asked him if he could share anything about his work. After a tan
asciilifeform: talizing pause: no.'
mircea_popescu: PeterL even if they do, if you can't use capital more productively than 5% a year it's time to go back to school.
mircea_popescu: (yes - that's the point of sending kids to school. to get them from where their average capital productivity is < 0% to where they get at least even with average. which should be within reach for an iq 100 fellow ; which is the only meaningful definition of iq available in the first place.)
mircea_popescu: asciilifeform goes right back to the whole "gestapo couldn't keep abreast of all the phonecalls from the "innocent public" they kept receiving". just like policeman rarely works there as a job, and generally works there as a hedge against his own inferiority complexes, derps working for nsa work there mostly for to be working there.
PeterL: so are you saying you wouldn't lend money at less than 5% because you have better things to do with it?
mircea_popescu: PeterL something like that. i wouldn't lend money in general, i'm not old and out of touch enough for that.
mircea_popescu: asciilifeform to look at it from a different perspective : attaching the attribute "classified" to a specified item has a number of effects on the world, some of which can have associated values that are computable. one such example is : that it triggers interest from the enemy, however defined. this is clearly a -. another one such example is : that it makes the workforce feel more important. since compensation always has to
mircea_popescu: = "opportunity cost" + "self-actualisation value", and since the sort of idiots that actually take jobs with the nsa have very simple "self actualisation" functions, which often evaluate to "be in a place where things are called 'classified'", then this necessarily is a +.
mircea_popescu: it stands then to reason to say that classification is more of a hr/personnel retention tool than a opsec instrument.
mircea_popescu: experience seems amusingly enough to confirm this theory.
asciilifeform: toilet cleaning schedule is classified
asciilifeform: but so is manhattan project.
mircea_popescu: and it ended up with the russians.
asciilifeform: so it did.
mircea_popescu: so that didn't help ; or even hindered.
asciilifeform: usa of the period was a culturally impoverished zimbabwe
asciilifeform: was stuck using foreign specialists.
mircea_popescu: meanwhile, it DID make the job of recruiters that much easier.
mircea_popescu: i am not so convinced the past 60 years made the us any less of a culturally impoverished zimbabwe. it's still not a nation, nor much of a country. it doesn't even make much of a hotel, used to be a sort of cote d'azur, now more like jamaica.
mircea_popescu: "the place where middleaged couples with strange ideas about what'd put the spark back in the bedroom go for massage-vacations"
asciilifeform: my point was, they use aryans now.
asciilifeform: wake me up when princeton leaks even 1 page.
a111: Logged on 2016-05-12 17:23 shinohai: trinque: how do i insert a line break? <br> doesn't work it seems.
shinohai: ^ kek
mircea_popescu: is this good for bitcoin ?
deedbot: [Trilema] The Megawatt standard - http://trilema.com/2016/the-megawatt-standard/
BingoBoingo: Meaningful rebellion in one's twilight years https://archive.is/hyTIq
mod6: <+asciilifeform> http://natashenka.ca/posters << lol these should go up everywhere
trinque: all the abstinence kids look like their genes are quite happy with calling their current derp the end of the road.
deedbot: [ยป Contravex: A blog by Pete Dushenski] Outsourcing fever. - http://www.contravex.com/2016/05/12/outsourcing-fever/
← 2016-05-11 | 2016-05-13 → ↑
Random(trilema) | Download hourly DB snapshot | Get Source Code