log☇︎
88200+ entries in 0.052s
mod6: BingoBoingo says he's on his way to the interwebs, was out shopping for hot water heaters...
mircea_popescu: i don't mind helping with the unobvious, but exactly what occured will occur again, "omfg, that was so obvious, where is it" sorta problem.
mircea_popescu: asciilifeform there's an impedance mismatch, sadly. i've not managed anything without extensive and reliable ops orgs in... well basically forever. reason i was an eff expert/consultant rather than manager was specifically that i never liked minding the middle.
mircea_popescu: better variant thereof, i guess. so travel, best thing for the suicidal.
ben_vulpes: mircea_popescu: i'm spending a bit of time keeping her from walking across a highway, if that counts as "walking on you"
ben_vulpes: mircea_popescu: if BingoBoingo can pick up the statements, that shrinks the delta; i've been flailing at transforming btc into amazon-compatible usd for the ssd shipment over the past week, he may be able to assist with that
asciilifeform: dunno, mircea_popescu is actually quite effective manager, when he's helped with e.g. minutiae of iron. and asciilifeform appreciates his unofficial help to date.
mircea_popescu: ben_vulpes so is she walking on you, is the idea ?
mircea_popescu: he seems rather reliable to date.
mircea_popescu: what's the delta between what you need and what BingoBoingo can do / has been doing ? ☟︎
mod6: ben_vulpes: ty
ben_vulpes: i will do as right by pizarro as i can; i will continue to do what i can. i must deliver notice that pizarro needs to find additional help.
mod6: Yeah, we're looking for L1, perhaps L2 if there is a good fit.
mod6: But more over, no, most roles you need to give a months notice or some such thing. It's not my preference, but what can I do?
ben_vulpes: mircea_popescu: i will do my best to continue supporting pizarro, i won't diappear immediately. but i need to find someone to take over my responsibilities, and i'm going to be stretched very thin.
mod6: I'm not, but I understand that the man needs to help his family.
mircea_popescu: so then why are you encouraging it ?
mircea_popescu: mod6 you ever been involved in any sort of multi-person anything where you could walk like that ?
mod6: ben_vulpes let us know that he needs to step back due to a family crisis. Not sure what more we can do, it's not ideal, but he's going to be wrapping up the report for June and transfering all of the rest of the information and other things to asciilifeform and myself.
a111: Logged on 2018-07-02 18:05 ben_vulpes: lords and ladies, june pizarro statement will be late; my wife is experiencing an all-consuming major depression that will take all of my attention going forward. i am working with mod6 and asciilifeform to ensure continuity at pizarro.
mircea_popescu: http://btcbase.org/log/2018-07-02#1831355 << dude, wake the hell up. you can't quit on a day's notice, what the fuck is this. ☝︎
mircea_popescu: wait a second, this is what, at will employment ? don't even have to put in a full year or w/e, can just walk whenever the fuck it no longer looks pleasant ?
mod6: You were disconnected when the news was announced, if you didn't see mircea_popescu .
mod6: One thing that I'd also like to mention regarding the role for pizarro manager: We are looking to de-couple the manager role from the accounting portion. We're really looking for two roles at this moment. One person to handle the accounting role, and another person to handle the day-to-day management.
mod6: I'll stand in as acting manager for the time being, but now begins the official search for a new pizarro manager. Please let us know if interested in the role.
asciilifeform: sad to hear. duty of commander then falls to mod6 . asciilifeform will take over the sysop duties that ben_vulpes has been volunteering for. the search for treasurer is on.
mod6: Thank you for the update ben_vulpes. All the best to you and yours.
ben_vulpes: lords and ladies, june pizarro statement will be late; my wife is experiencing an all-consuming major depression that will take all of my attention going forward. i am working with mod6 and asciilifeform to ensure continuity at pizarro. ☟︎☟︎
asciilifeform: mircea_popescu: in principle. in practice, very difficult ( in asciilifeform's understanding, entirely impractical ) to avoid your model being poisoned at the heart by the retardation of the iron ( dma, interrupts, whole orchestra of horrors that prevent anything like effectual isolation of processes , esp in kernel, and -- more insidiously -- prevent sane model of computation even in head )
asciilifeform: they wrote strictly bicameral thing, '0' (kernel, do whatever to whomever) and '3' (on unix planet, aka userland, process lives in own map)
asciilifeform: kernel already runs in (x86ological terminology) ring0
asciilifeform: ( where there are, for obvious reason, no memory protections etc )
mircea_popescu: if this were strictly true, nobody'd know what a bsod is.
asciilifeform: ( to carry on the analogy, in x86's retarded memprotections, you can ~spy on cunt~ of neighbour. but not insert into, unless she is kept in your potato cellar. )
asciilifeform: only when she is kept with the same potatoes.
mircea_popescu: i dunno, seeing how all the "bug bounties" ever return is "o wait, off by 1, overwrote other guy's girl"
asciilifeform: far better, to dispense with 'accidental overwritings', period.
asciilifeform: i have somewhat difficult time picturing the win from auto-overwriting. 'ooops you brought home 1 too many potatoes, you just overwrite gurl's spleen'
mircea_popescu: asciilifeform i dunno, i'm just asking him things. i wasn't trying to wrap some tellings into a pretense to questions.
asciilifeform: what's the win from auto-wrapping , anyway
asciilifeform: as reliable as the iron. and if you want to mod your pointers, you can, writers of e.g. nintendo, etc emulators, do
mircea_popescu: just about, though seems somewhat unreliable. and no mod.
mircea_popescu: asciilifeform how about the implict model we kept implicitly discussing, "program can ask for memory chunks, os will treat all writes to allocated memory as mod allocated size".
asciilifeform: all known approaches to gc , fall into 1 of these 3 cuts.
asciilifeform: efs forbidden) , copies to new space. aka 'semispace'. 3) errything else. high-tech , complicated, 'multi-generational' gc, with 'nurseries', etc. imho best implemented in iron, and quite difficult to get it right.
asciilifeform: there are 3 basic cuts of gc. 1) simplest -- no gc. you cons N times, and you're through. this is actually practical ( see the last coupla gc threads ) for scripting lang, where proggy runs for a coupla seconds, on inputs of known size, etc 2) 'stop & copy'. simplest gc, implemented in 1950s-1970s lisps, and by erry student in school. proggy runs out of space -- stops, finds what is alive and what is dead via refcounting (circular r
mircea_popescu: the original "let's unwind it from the FS end" didn't actually turn out capable to carry water, upon discussion.
mircea_popescu: i'm looking for roads into the os. could be from phf's emacs, could be from spyked 's lisp, could be from your arms, could be from whatever the fuck ends up working. ☟︎
asciilifeform: mircea_popescu: plox to expand ?
mircea_popescu: and what the hell is wrong with "let byte = 1010001101001011010111010111100" producing byte = 01111001, because 10100011/01001011/01011101/0111100
a111: Logged on 2017-01-03 20:34 mircea_popescu: all three queues to be implemented as ring buffers of user specified size.
mircea_popescu: (there's a lot of discussion of this topic in the logs, but to fish one example out of the many : http://btcbase.org/log/2017-01-03#1595771 ) ☝︎
mircea_popescu: spyked re http://thetarpit.org/posts/y04/074-adalisp-prototype.html what if anything do you plan to do re memory management ? gc in particular, but not just. ☟︎
trinque: minor niggle, http://ave1.org/tarpit/readme-2018-06-01.txt has the build script's name backwards
trinque: ave1: herp, it even said that. must've been tired. chugging along now, ty
mircea_popescu: meanwhile all quiet on the weird front. https://78.media.tumblr.com/c2b692b8132b4cbf2ff59a6ab0c1e4ba/tumblr_p3ivxcMy6N1rwygwno1_1280.jpg
mircea_popescu: fromloper we know that, by your name, you see.
asciilifeform: fromloper: i recommend to read the logs, http://btcbase.org/log/
fromloper: Currently (re) learning LISP and was simply scouting out places to lurk ;)
fromloper: Curious traveller. Got here by the 'Contact' link on the loperos blog.
asciilifeform: fromloper: you're on the air! who might you be ?
spyked: ^ phf, can you pl0x add this to the btcbase patch list? and http://lucian.mogosanu.ro/src/trilemabot/ ? ☟︎
deedbot: http://thetarpit.org/posts/y04/075-adalisp-prototype-ii.html << The Tar Pit - The return of the lost son of the father of adalisp, now in a genesis
hanbot: cheers lobbes, thanks for the service.
a111: Logged on 2018-07-02 02:29 esthlos: http://btcbase.org/log/2018-07-02#1831237 << interesting. seems like implementing the tags wouldn't be hard, but choosing the right ones would be
diana_coman: http://btcbase.org/log/2018-07-02#1831260 <- heh, like always; the summaries page looks useful to me ☝︎
lobbes: ty trinque
ave1: trinque, The directory for the script needs to absolute, i.e. ./build-ada.sh output --> ./build-ada.sh $PWD/output
a111: Logged on 2018-07-01 20:49 lobbes: incidentally, trinque, next time you're in the boiler room plox to try adding http://blog.lobbesblog.com/feed/ once more
deedbot: http://blog.lobbesblog.com/2018/07/arming-your-arm64-rockchip-gentoo-against-the-hordes-of-mindless-bots/ << lobbesblog - Arming your Arm64 RockChip Gentoo against the hordes of Mindless Bots
deedbot: http://blog.lobbesblog.com/2018/06/getting-php56-and-mysql-to-play-nice-on-arm64-rockchip-gentoo/ << lobbesblog - Getting PHP5.6 and MySql to Play Nice On Arm64 RockChip Gentoo
trinque: ave1: http://p.bvulpes.com/pastes/ggQ4D/?raw=true << getting this error when I run your latest. any idea where I'm going wrong?
mircea_popescu: ie, just like the seals directory, except the seals aren't binary but whatever, a number.
a111: Logged on 2018-07-02 02:41 lobbes: with re: to curation, you could record the nick for everyone that submits a tag, then just allow others who view the tagged data to filter based on nick. (Other option of course would be a gpg challenge to submit a tag, but I doubt people will want to unsheathe their key every time they want to tag a thread.)
mircea_popescu: http://btcbase.org/log/2018-07-02#1831265 << nah, the correct approach is to leverage the wot and the v. basically, allow user to give weights to tag authors, obtain a weighted average tag cloud per that spec. ☝︎
mircea_popescu: somewhere in logs /me explains the screw.
mircea_popescu: esthlos no rush anyway, as one grows it drives the other
lobbes: anyways, I'm just spit-balling here. You might find a much better way to do all this
lobbes: with re: to curation, you could record the nick for everyone that submits a tag, then just allow others who view the tagged data to filter based on nick. (Other option of course would be a gpg challenge to submit a tag, but I doubt people will want to unsheathe their key every time they want to tag a thread.) ☟︎
lobbes: indeed, the benefit of the tag method is that -you- don't have to personally tag each thread (not saying you can or cannot, just... it is an immense amount of effort).
esthlos: I'd be happy to be the librarian, though could imagine a system where the bot allows trusted users to "push" summaries or tags. curation is a necessity, but I don't doubt others can produce much better summaries than I
a111: Logged on 2018-07-02 01:52 lobbes: pretty nifty esthlos. Didja see the !tag thread btw? I always thought that would be a good way to approach the 'chronicle the logs' problem >> http://btcbase.org/log/2016-03-21#1437059
esthlos: http://btcbase.org/log/2018-07-02#1831237 << interesting. seems like implementing the tags wouldn't be hard, but choosing the right ones would be ☝︎☟︎
a111: Logged on 2018-07-01 21:22 asciilifeform: lobbes: iptables will work if you build the kernel ( there is a tarball with patched kernel on all pizarro rockchip boxen ) with the requisite flags
lobbes: http://btcbase.org/log/2018-07-01#1831138 << ty for this info. Some of my research on subj. indeed hinted at 'rebuild kernel' but it would have taken me a bit more time to actually confirm this for myself ☝︎
deedbot: lobbes rated esthlos 1 << reads log, keeps blog, does useful things. See blog.esthlos.com
a111: Logged on 2018-07-02 01:57 mircea_popescu: esthlos this then leads to the next stop, esthlos bot for reading out the content of a bullet :D
esthlos: http://btcbase.org/log/2018-07-02#1831241 << well I fell into that trap, huh! sounds like a good next project ☝︎
mircea_popescu: esthlos this then leads to the next stop, esthlos bot for reading out the content of a bullet :D ☟︎
mircea_popescu: certainly as far as i recall one of the "nutty" things alf said early on that signalled him away from the crowd to my eye.
mircea_popescu: http://summaries.logs.esthlos.com/#2018-06-27_4 << item is actually ancient, 2014s. in fact, predates "FIH", possibly lead to it.
a111: Logged on 2016-03-21 07:43 mircea_popescu: hanbot mod6 maybe an extension to deedbot where people could go !tag <string> and then a collection of clickable tags is published somewhere, each leading to a list of loglines ?
lobbes: pretty nifty esthlos. Didja see the !tag thread btw? I always thought that would be a good way to approach the 'chronicle the logs' problem >> http://btcbase.org/log/2016-03-21#1437059 ☝︎☟︎
esthlos: also, my piece shoulda mentioned: about 900 days since the bitbet blowout, which is my starting place
hanbot: <mircea_popescu> hanbot you got copies somewhere ? << that'd've been wise, but i didn't think to make a copy.
esthlos: hmm yep, I can generate those. let's see...
mircea_popescu: broken though patterns << " broken thoughT patterns. tho i have no good way to reference it.
mircea_popescu: esthlos another item is that some bullet-addressing mechanism particularly useful on your summary page. put anchors ?
mircea_popescu marks 2019 june on his calendar for the followup
mircea_popescu: "My attempted pace will be to read 4 days of logs every day (3 older logs, and the present log). The average log takes me about 30 minutes to read (including following links), and there's a bit over 900 logs to catch up on, so I'll be caught up in 300 days, working 2 hours a day. Starting late is a bitch, ya know." << i am SO FUCKING CURIOUS how this works out!