#trilema | 2016-01-23

Logged on 23-01-2016 02:43:37; asciilifeform: 'electron moves because allah willed so'

00:20

http://log.bitcoin-assets.com/?date=23-01-2016#1382232 << even the simple step of "add flag to trb to dump mempool into file each time a block is found" would be helpful atm.

00:20

Logged on 23-01-2016 03:02:42; polarbeard: feed me :3

00:22

(speaking of which - de-stupidification of the logging process would be a fucking blessing already. debug.log, srsly ? how about error.log mempool.log connect.log blocks.log for starters ?)

00:22

^ agree, also hex logs

00:22

well, at least sanely tabulated.

00:23

in any case, then you could start it with -logs=emcb for all 4, or otherwise whatever combo thereof.

00:23

and for the love of christ implement either rolling or else a maxsize for files or something.

00:23

roger

00:24

no logrotate love?

00:24

or that. anything.

00:24

as it stands right now it's made by sheepherds.

00:25

it's a miracle it works

00:28

ah, and also - fucking timestamps on loglines omfg.

00:28

who did that!

00:29

"here's a logline. it's from fuck you o'clock. you're welcome."

00:29

fuck you o'clock lol

00:29

didn't phf do something with timestamps?

00:30

haha, 'just derive it from block height man'

00:30

it doesn't even dump that, which would be insane but at least it'd be insane.

00:30

polarbeard: good idea except that syncing reference implementations to the best of my knowledge don't know the current blockheight

00:31

/ironic off :)

00:32

I'd be happy with unix_time space hex_message sent_or_received_bit

00:32

the log history that you're missing is that blockheight is the only non usg-shared time.

00:33

ben_vulpes it could always curl deedbot.org for the latest.

00:33

yes let's compile libcurl into the reference implementation

00:33

after all... we have fucking json in there for reasons!

00:33

i'm sure polarbeard can get on that train

00:33

choo choo!

00:33

moar deps

00:34

#nodeps

00:34

lol aren't you a bucket fulla sunshine.

00:34

writing software fills a boy with hate.

00:34

mircea_popescu: what was the calculation you ran to determine the hashpower necessary to nail a block more or less every 24 hours?

00:35

#noderps #nodeps #nodes

00:35

#definitelynofuckinnode

00:35

that there's 6 blocks an hour and 24 hours in 24 hours ?

00:36

so one'd need ~ 1/144th of nethash?

00:36

well, yes, if one's tolerant with the variance

00:36

sure

00:36

as you divide, variance divides by less.

00:36

;;nethash

00:36

892506470.131

00:36

blocks per day = (my_share_rate / difficulty ) x 86400 / block_time

00:37

still, it comes to about 2MW's worth, more or less.

00:37

my_share_rate = hash rate / 2^32

00:38

polarbeard so basically myhash/nethash * 144.

00:38

someone missed dimensional analysis in school?

00:39

yep, but nethash is not exact I think?

00:39

neither is 144.

00:40

approximation is a nifty wunderwaffen.

00:40

mircea_popescu: schooled me rather hard on a good example around c3

00:40

i could hear you chuckling from here.

00:41

nethash is a joke is what it is

00:41

amusingly, the proper measure for nethash actually is PW or somesuch.

00:41

petawatts?

00:41

by the time it gets to 50%+1 all these approximations will be a lot less approximative

00:42

50 TW, something like that.

00:43

50 PW*

00:43

that has a lot of assumptions on mining chip efficiency boss

00:43

non?

00:44

nope. stable situation is where nethash is half the power generation.

00:44

chips irrespective, be they z80s or whatever startrek

00:44

but now.

00:44

not then.

00:45

universally.

00:46

"Hypothetically, according to Robert Engelman (Worldwatch institute), in order to prevent collapse, human civilization would have to stop increasing emissions within a decade regardless of the economy or population (2009)."

00:46

when you say that it comes to 2MW's worth.

00:46

ahh this shit's always such lulzyu lull.

00:46

ben_vulpes ah yes, yes.

00:46

currently we're only burning something to the tune of 30GW

00:46

which is scandalous, but hey.

00:47

mircea_popescu: 21co tried to make energy friendly chips, now they are trying to pawn them on users as "compooter"

00:47

;;calc 3* 10^10 * 24 * 365

00:47

Error: Something in there wasn't a valid number.

00:47

;;calc 3* 10**10 * 24 * 365

00:48

262800000000000

00:48

https://en.wikipedia.org/wiki/List_of_countries_by_electricity_consumption << the esteemed republic is #13, after italy and above most of the world.

00:48

List of countries by electricity consumption - Wikipedia, the free encyclopedia ... ( http://bit.ly/1ODTasQ )

00:49

it could light up almost 3000 chads.

00:50

always imagined some d00d with access to nuclear power building a reactor on some island and using all the power for mining

00:52

did anyone ever run the numbers on what it would naively cost to lease the hashpower required to mint a block a day?

00:53

;;help nethash

00:53

(nethash takes no arguments) -- Shows the current estimate for total network hash rate, in Ghps.

00:53

maybe 1% of the cost of a block, a day ?

00:53

of the subsidy of a block*

00:54

very simple mechanism to ACTUALLY improve your insurance : make a pool which pays 25.25 btc for the first block it finds each day.

00:54

this is a very neat way to solve an otherwise very thorny economical problem, and it will figure prominently in the future.

00:55

miners will not only optimize txn to include in blocks per se, but also "whose pool to mine nao at"

00:55

so what's the lowdown on the various pool payout maths?

00:55

what do you mean ?

00:55

every time i look into pool payout schemes my eyes glaze over and my fraud alert systems kick in.

00:55

[MPEX] [S.MPOE] 35017 @ 0.000556 = 19.4695 BTC [-] {2}

00:56

there's 3 or 4 principal ways to do it, 2 or 3 of which are very slight variation on the same theme.

00:56

it's a complex topic but no, they're not fraudulent.

00:56

arrived at through a lot of experimental as well as theoretical effort.

00:56

has anyone written this up anywhere?

00:56

well... miners.

00:56

[MPEX] [S.MPOE] 81283 @ 0.00055562 = 45.1625 BTC [-] {3}

00:56

worth your time to research it for your own sharpening, it's all pretty clever.

00:57

myeah

00:57

perhaps i'll do that while i install os x on this laptop for the umpteenth time

00:57

fucking operating systems

00:57

should take about a week.

00:57

depends how hard you push the "but why!" pedal.

00:57

lotta history.

00:59

man imma have to learn chinese at this rate

01:00

ideograms, man.

01:01

[MPEX] [S.MPOE] 35000 @ 0.00055963 = 19.5871 BTC [+] {2}

01:06

https://www.youtube.com/watch?v=v5R9SZM2jaE

mircea_popescu: and i'm going to guess most of this history is on bitcointalk.org

↖ ↖

01:17

mats

01:17

Ukrainian Armor (The Rolling Stones - Gimme Shelter) - YouTube ... ( http://bit.ly/1ODX1WS )

01:22

mats: badass

01:23

tis the age of the cmos

01:26

mats

the autoloader is cool

01:27

mats

for some reason the abrams doesn't have it

~ 2 hours 13 minutes ~

03:40

bon dia

03:40

https://s-media-cache-ak0.pinimg.com/736x/23/a1/38/23a138d0e5eeb772813bb196e4c5f31c.jpg

03:40

... ( http://bit.ly/1PoZ98p )

03:40

^ dirigible

03:52

http://log.bitcoin-assets.com/?date=23-01-2016#1382541 especially the earlier history, yes. possibly a good place to start: https://bitcointalk.org/index.php?action=profile;u=14118;sa=showPosts;start=6160

03:52

Logged on 23-01-2016 06:06:51; ben_vulpes: mircea_popescu: and i'm going to guess most of this history is on bitcointalk.org

03:52

Latest posts of: organofcorti ... ( http://bit.ly/1Pp0g8b )

03:54

;;diffchange

03:54

Estimated percent change in difficulty this period | None % based on data since last change | 10.264 % based on data for last three days

03:55

https://megabigpower.com/buybackinfo

03:55

MegaBigPower :: Info ... ( http://bit.ly/1PKOWNY )

03:55

"pleased to announce our ASIC BuyBack Program"

03:57

https://hedgy.co/mining.html "Buy Bitcoin Direct from MegaBigPower at a Discount on a Future Date. Auctions Occur Once Per Month."

03:59

"It has been quite a while since any new hardware became available, it is reasonable to assume that most small miners will have recouped the majority of what they initially outlaid for the hardware, and are now disinclined to continue ‘supporting the network’, whilst incurring unnecessary cost. So: MegaBigPowers ‘land-grab’, to acquire all this idle hashing power seems like a smart move. As,

03:59

once the next price run up begins, there will be a frantic scrabble to re-power all the mothballed rigs lying around."

04:00

from http://blockchainstuff.eu/thoughts-on-the-current-state-of-bitcoin-mining/

04:00

Thoughts on the current state of Bitcoin mining | Blockchain Stuff ... ( http://bit.ly/1PKP93J )

04:01

"Bitfury recently announced their 28nm, 0.2 J/Gh offering (with some impressive efficiencies claimed) but have declined to inform us when they will be available, either-ways they do not sell to the public anymore. They also alluded to a 16nm chip with efficiencies as low as 0.02 due out later this year."

04:01

"KNC are in a similar holding pattern, awaiting production of their 16nm ‘Solar’ chip, with astonishing claims of 0.07 J/Gh, They no longer sell to the public either."

04:03

they sold until it was garbage, now that further decreasement in process size are hardly possible, they hold

04:03

oooh just noticed "March 2015"

04:05

basically folks have been used as sort of catalyst for the reaction of them getting 16nm or so and are now no longer needed

04:06

hmm

04:09

[MPEX] [S.MPOE] 50750 @ 0.00055964 = 28.4017 BTC [+]

04:10

http://log.bitcoin-assets.com/?date=23-01-2016#1382516 << perhaps i am up too late, but i cannot make heads nor tails of this

04:10

Logged on 23-01-2016 05:53:27; mircea_popescu: maybe 1% of the cost of a block, a day ?

04:14

ben_vulpes: 1% extra

04:18

https://github.com/nothings/stb/blob/master/docs/other_libs.md "single-file public-domain/open source C libraries with minimal dependencies"

04:18

stb/other_libs.md at master · nothings/stb · GitHub ... ( http://bit.ly/1Pp2TXI )

04:20

[MPEX] [S.MPOE] 36700 @ 0.00055964 = 20.5388 BTC [+]

04:24

http://tweetnacl.cr.yp.to/software.html

ah?

04:39

BingoBoingo

~ 1 hours 15 minutes ~

05:54

[MPEX] [S.MPOE] 53000 @ 0.00055964 = 29.6609 BTC [+]

~ 27 minutes ~

06:22

[MPEX] [S.MPOE] 51500 @ 0.00056057 = 28.8694 BTC [+] {3}

~ 38 minutes ~

07:00

shinohai

Height 394621 \o/

07:01

*

shinohai pops cork on muscadine wine even though it is 7 a.m.

~ 42 minutes ~

07:43

[MPEX] [S.MPOE] 30800 @ 0.00055961 = 17.236 BTC [-]

~ 48 minutes ~

08:31

;;rate RagnarDanneskjol -10 3 months of unpaid salary, will remove rating if payment ever arrives.

08:31

Error: For identification purposes, you must be authenticated to use the rating system.

08:32

;;rate RagnarDanneskjol -10 3 months of unpaid salary, will remove rating if payment ever arrives.

08:32

Rating entry successful. Your rating of -10 for user RagnarDanneskjol has been recorded.

08:33

*

Azelphur sighs

08:38

Was really unsure whether to place that one (badmouthing former employer and all) but figured he'll probably look here for someone to replace me, and other people might not be so financially secure as I am.

08:43

told me he needed some contract work 3 months ago, but didn't follow through

08:44

;;seen RagnarDanneskjol

08:44

RagnarDanneskjol was last seen in #bitcoin-assets 44 weeks, 0 days, 2 hours, 50 minutes, and 19 seconds ago: <RagnarDanneskjol> trinque - i am occasionally using the nick for continued development on punks ver. if anyone here's partial to the name I can pass it over. Wow that xm42

08:45

punkman, yea for me I did the work, he just keeps saying he will pay and continues to delay / make excuses

08:45

he said he;

08:46

whoops, he said he'd pay me an installment of £1000ish on Monday, then said it was a bank holiday and that he would send it out Tuesday, ignored me till yesterday where he claimed he had sent it out, but honestly he has said it's sent so many times and it didn't arrive I don't trust it any more

08:48

yeah that's not very good

08:48

http://log.bitcoin-assets.com/?date=23-01-2016#1382541 << yea, actually, the one thing bitcointalk is a source for is miner stuff. they never got any other venue , well noit publicly anyway

08:48

Logged on 23-01-2016 06:06:51; ben_vulpes: mircea_popescu: and i'm going to guess most of this history is on bitcointalk.org

08:48

Indeed it's not, had no money over christmas. His misinformation has cost me money too

08:49

o.O

08:50

that didn't work out so well did it

08:50

apparently not, no

08:50

was great up until the whole not paying me thing

08:50

you were working with him for a while though, right?

08:50

punkman, yea, I started with him last October iirc

08:51

ie, October 2014

08:51

kakobrekla http://trilema.com/2016/bitbet-sbbet-december-2015-statement/#comment-116281 << anything to this ?

08:51

BitBet (S.BBET) December 2015 Statement on Trilema - A blog by Mircea Popescu. ... ( http://bit.ly/1Ppwpwi )

08:52

mircea_popescu: yeah I get a new cert on bitbet, issued 2016/01/21

08:53

ah kk.

08:53

thing ended up in spam, possibly because of the link

08:54

but yea, officially looking for new work now. Hopefully it's a good move and I'll get paid a little more (and at all :P)

09:00

http://log.bitcoin-assets.com/?date=23-01-2016#1382551 << ahaha no that's a balloon. dirigible is oval not buttshaped.

09:00

Logged on 23-01-2016 08:40:37; punkman: ^ dirigible

09:01

http://log.bitcoin-assets.com/?date=23-01-2016#1382554 << he also had a blog which was pretty good.

09:01

Logged on 23-01-2016 08:52:39; assbot: Latest posts of: organofcorti ... ( http://bit.ly/1Pp0g8b )

09:01

iirc the dude hated me for some reason.

09:02

http://log.bitcoin-assets.com/?date=23-01-2016#1382567 << well, actually, they sold back when they had to take the poisoned dollars of the masses, a few bucks at a time. once they n olonger need that, they don't. who eats garbage if they can help it.

09:02

Logged on 23-01-2016 09:03:04; kakobrekla: they sold until it was garbage, now that further decreasement in process size are hardly possible, they hold

09:03

Azelphur notsure why you're not using the assbot wot anyway.

09:04

mircea_popescu, haven't been around much recently so haven't familiarised myself with it

09:04

ah

09:05

anyway, there was a bank holiday on monday. in the us. some black speechmaker dude

09:06

http://log.bitcoin-assets.com/?date=23-01-2016#1382242 << tell a bit about the concentrated thing

09:06

Logged on 23-01-2016 03:32:24; mircea_popescu: http://log.bitcoin-assets.com/?date=23-01-2016#1382094 << you deeply misunderstand what "investor grade" means in this context. the point is that he is RARE. you understand this ? the fundamental property of capital control in any functioning society without exception possible is that it must be concentrated.

09:06

mircea_popescu, yea I know, dude is very good at excuses, as I say it has been 3 months, and even if he has sent a bank transfer it's still only for a small percentage of total owed

09:06

is it the '1024 chickens' idea ?

09:06

nope. it's the "brain resides in skullbox not spread out throughout cells like mitochondria".

09:07

same thing, no ?

09:07

i guess it can be viewed so, yeah. ok.

09:07

i'm trying to grasp how it isn't an argument for gosplan tho

09:07

the problem here is that both ox and chickens are separate from the thing they pull.

09:07

ah, this IS an argument for gosplan. the only one.

09:07

if it was just missed payment by a few days I'd be cool with it, heck I was fine with it up until about 2 weeks late, stopped working when it was a month late, and public -10 has only happened after 3 months late

09:07

nevertheless, the argument against gosplan is that "fuck you".

09:08

while you personally may not be inclined, i am, and without me it dies.

09:08

Azelphur i c.

09:08

started to get really annoyed when I had the choice of taking money out of my cold wallet or taking a loan from my parents to make rent over christmas - not cool :/

09:09

so mircea_popescu doesn't have an argument against gosplan ?! other than not liking the colour & shape on purely aesthetic grounds?

09:09

the argument against gosplan is exactly the argument against "hi guise i am ceo".

09:09

"who the fuck are you ?"

09:09

I feel bad -10ing him though, I like to work things out with people and it just sucks that he's not meeting his obligations or even being reasonable really

09:10

which is why all the red square nuclear bomb displays and etc. obligate pompousness.

09:10

lucky we are that 'hi guise im ceo' has no army and no navy.

09:10

it's a forced mistake.

09:10

quite.

09:10

(except when he does, e.g., obamitler)

09:10

problem being, they don't actually help, just serve to alienate me further.

09:11

only because the beast is immunocompromised

09:11

mno.

09:11

i am immune to it, it depends on me. this creates a certain situation of power in the field.

09:12

depends how ?

09:12

anyway, in the expression of the castrated subhumans that nevertheless are intelligent enough to notice the problem, this is usually called the "knowledge problem", ie poor state can't acquire and process enough information to make a reasonable gosplan.

09:12

nevertheless, this is a flattering if incorrect representation. like saying "unfortunately whale did not have enough evolution mojo to grow wings". yeasure.

09:13

asciilifeform i have no idea how to formulate this better than to refer you to the (oft mentioned, by me) worms dispute.

09:13

king can not invest by staff. he doesn't have a staff to give.

09:18

mircea_popescu: modern usg is going for 'eastern despot' caesaropapism.

09:18

which is why they keep inventing academic honors for the emperor. they work admirably, don't they.

09:19

hey it worked for the 5000yr egyptianreich!

09:19

not so.

09:19

your idea that "5000 yr egyptianreich" is not unlike a 1900s victorian idea that "5000 yr chinareich". wut ?

09:19

they had more dynasties than americans had sexual partners.

09:19

point^

09:21

ironically, the people who could be caesar ie emperor and pontifex max uniformly do not wish to be, so much so it's almost a heuristic.

09:22

and they do not wish to be not because they somehow give a shit about brutus and brutish jealousies.

09:22

but, at this juncture, augustus' memoirs are recomended re-read. might explain the vecchio rotinculo much closer than before.

09:24

http://log.bitcoin-assets.com/?date=23-01-2016#1382542 << ukrs are apparently really good at shooting into 1) air 2) earth 3) defenceless farm houses. who'dhaveknown.

09:24

Logged on 23-01-2016 06:17:35; mats: https://www.youtube.com/watch?v=v5R9SZM2jaE

09:25

spiffy firewurx.

09:25

and wake me up when they ~make~ a tank.

09:25

aww why the haet!

09:26

dja know that the AVERAGE ammo expenditure per enemy killed is in the 2mn range by now ?

09:26

people keep mocking hollywood marksmanship. mostly because they've never been in a warzone.

09:26

well also because did not go to kindergarten, never learned concept of suppressive fire

09:28

seeing how it's about 2kj per bullet, we find that each enemy combatant takes ~1MWh in bullets only!

09:29

and that's on the modest 39mm. pretty much a thing of the past, everyone doing 51 now amirite. so 2x that.

09:30

*

asciilifeform was very very confused until realized that mircea_popescu is counting the ~length~ of the shell !

09:31

well yes. 39 is the old ak/sks russian thing. nato has a bigger one.

09:31

today as in '70s the man-portable small arms fire the 'intermediate' things and the squad guns - the 1890s longer one.

09:33

[MPEX] [S.MPOE] 101766 @ 0.00055833 = 56.819 BTC [-] {3}

09:33

well, depends. the large nato round is really like the hunting .308

09:34

mircea_popescu: http://s242.photobucket.com/user/Akalvarez_2007/media/PANA0003.jpg.html << sov-new (ak74), nato-new (m16), sov-old (ak47), nato-old-fullsize, sov-old-fullsize (mosin, and all ru squad guns, 1890s round)

09:34

Ammo Side By Side Photo by Akalvarez_2007 | Photobucket ... ( http://bit.ly/1KwjWQb )

09:35

aha

09:36

notice the very archaic flanged shell and rimmed bottom on the last

09:36

yeah lol.

09:36

century old bullet!

09:36

oldest rifle cartridge still in mass use.

09:36

more than century.

09:37

*

asciilifeform played with a bag of empty shells of those, as a boy

09:43

http://log.bitcoin-assets.com/?date=23-01-2016#1382439 << i thought we had this ?

09:43

Logged on 23-01-2016 05:23:47; mircea_popescu: and for the love of christ implement either rolling or else a maxsize for files or something.

09:44

http://log.bitcoin-assets.com/?date=23-01-2016#1382433 << all i really want presently is to dump mempool. and to have some means of removing ANYTHING whatsoever from it. because if we had this, we COULD semispace-garbagecollect.

09:44

Logged on 23-01-2016 05:20:23; mircea_popescu: http://log.bitcoin-assets.com/?date=23-01-2016#1382232 << even the simple step of "add flag to trb to dump mempool into file each time a block is found" would be helpful atm.

09:44

(add what belongs in the pool to a new pool, then nuke what remains. also defrags...)

09:45

asciilifeform let the guy do something useful wouldja. step at a time.

09:45

http://log.bitcoin-assets.com/?date=23-01-2016#1382456 << ick

09:45

Logged on 23-01-2016 05:33:08; mircea_popescu: ben_vulpes it could always curl deedbot.org for the latest.

09:45

and drink humour vitamins!

09:45

l0lzamins

09:46

very fucking hard making jokes about bitcoin codebase, seeing the state it's in.

09:46

'who has been in the army, does not laugh in the circus' (tm) (r) (su)

09:46

like fucking pagliacci over here, just turns to sad.

09:46

ahahaha quite.

09:47

http://log.bitcoin-assets.com/?date=23-01-2016#1382549 << this looks disturbingly like BingoBoingo's material

09:47

Logged on 23-01-2016 08:40:14; punkman: https://s-media-cache-ak0.pinimg.com/736x/23/a1/38/23a138d0e5eeb772813bb196e4c5f31c.jpg

09:48

http://log.bitcoin-assets.com/?date=23-01-2016#1382561 << this seems like a deviously clever notion until you recall that most of the asic miners were - quite likely deliberately - not built to last

09:48

Logged on 23-01-2016 08:59:13; punkman: "It has been quite a while since any new hardware became available, it is reasonable to assume that most small miners will have recouped the majority of what they initially outlaid for the hardware, and are now disinclined to continue ‘supporting the network’, whilst incurring unnecessary cost. So: MegaBigPowers ‘land-grab’, to acquire all this idle hashing power seems like a smart move. As,

09:48

http://36.media.tumblr.com/06e1a4f30719f22ffe8bd2f8abe73b1b/tumblr_nd32f8rOuT1rqs8dwo1_1280.jpg << in other news, never trust a man with a garage.

09:48

... ( http://bit.ly/1KwkTrB )

09:49

(they never got the wear circuits to behave plausibly like a real-life bathtub curve, either. so the things would routinely ship with varying number of working hash cores, etc)

09:49

iirc the overall "delivered asic, all cores working" was in the <5% zone.

09:49

aha.

09:50

i've always wanted to do a bit of microscopy on those, see if there was a set of magicks for re-enabling the 'accidented' cores

09:50

deliberately? imo overlyconfident.

09:51

kakobrekla: the whole racket stood on a solid column of chump, and was designed around such

09:51

more or less. really, esp early on, it was just blindly flailing.

09:51

they didnt have the time to make anything past "this will prolly work"

09:51

easy to reinterpret what memory salvaged of the midden of history, but if you were there and saw the missing 95% from that story...

09:52

kakobrekla: if this were so, there would have been many asic co. sunk by a taped out dud

09:52

afaik there were none.

09:52

and so there were.

09:52

ahahaha don't kid, you make my heart jump so.

09:52

but i - obviously - wasn't there.

09:52

nor wanted to be.

09:53

of the people who scammed their "investors, at least half, to two thirds, were scammed in turn, by their "providers".

09:53

i did say probably, not possibly.

09:53

yes but any public intel suggesting a taped-out total dud ?

09:53

or just 'took the money & ran'

09:53

asicminer had a fully blown up run. so did bfl, twice iirc.

09:53

so did many others.

09:53

there were, up until early 2013 or so, more dead runs than live.

09:54

i am specificially thinking of ~ic~ rather than ~board~

09:54

i remember very well the smoking ~boards~

09:54

pre-packaged chips.

09:54

afaik there were 'duds' i dont recall specifics from top of the head

09:55

then yes.

09:55

http://log.bitcoin-assets.com/?date=23-01-2016#1382579 << this is a mega-l0l, worth reading! turns out, one day they decided to play 'obfuscated c contest' at ft meade

09:55

Logged on 23-01-2016 09:39:27; BingoBoingo: http://tweetnacl.cr.yp.to/software.html

09:55

actually a proper history of early mining would be a most respectable endeavour for a serious, diligent historian at this point.

09:55

(srsly who the FUCK uses 'salsa' etc)

09:55

of course they're busy fucking around doing nothing much. but hey, gosplan!

10:01

I'd like to contribute more debug-related patches, could maybe split the gui-snip stuff from debug_sanity patch, and start over.

10:02

does asciilifeform like anything from djb?

10:05

[MPEX] [S.MPOE] 68950 @ 0.00056063 = 38.6554 BTC [+] {2}

10:11

https://github.com/SergiusTheBest/plog interesting logging library, less than 1kloc, includes simple log rotation

10:11

GitHub - SergiusTheBest/plog: Portable and simple C++ logging library ... ( http://bit.ly/1KwmNbS )

~ 20 minutes ~

10:32

punkman: quite a few things

10:34

punkman start over, how you mean ?

10:35

punkman: i'm a fan of the man's cryptoanalytic work. but the sv-popularity of his cipher primitives is concerning

10:35

problem is that one doesn't get to 'make a name' in academe as 'cryptographer' by pushing rsa.

10:36

just like you don't become a famous small arms designer by refining the kalash

10:36

even if ~that would be the Right Thing~

10:36

so he's stuck with the ecc gargle.

10:37

and with devising block ciphers, which is very easy to 'do'

10:38

!s from:mircea_popescu originality

10:38

6 results for 'from:mircea_popescu originality' : http://s.b-a.link/?q=from%3Amircea_popescu+originality

10:38

^see also

10:39

djb = bernstein ?

10:39

right

10:39

guy's brilliant, hard to argue with that. but yes, seems he's being slowly ground to paste by "teh system"

10:40

one way to look at it is that he isn't much more of a 'free man' than, e.g., i am

10:40

asciilifeform actually may want to write him a letter, iirc he was also all about fits in head but arrived there on slightly different path

10:41

in that he has to work on certain flavours of shit icecream, in order to eat

10:41

didn't he end up basically rewriting sendmail because sendmail sucked ?

10:41

aha

10:42

mircea_popescu: start over, meaning make a new debug_sanity patch with more improvements (and without the gui snipping parts bundled together)

10:43

this with a view to better semantics, or with a view to reinstate gui ?

10:43

mircea_popescu: my original thumbs-down for his patch was on account of bundling multiple-unlikes

10:43

it does carry water.

10:44

the vestigial wx crapolade needs to go

10:44

but ~in own patch~

10:44

this makes it considerably easier to evaluate every path

10:44

*patch

10:44

yeah.

10:44

both gui-snip and debug-sanity are gonna touch a lot of files though, and I would prefer to do this once

10:44

'does this do X and ONLY X'

10:44

punkman: this is very much the wrong idea

10:44

we aren't trying to minimize patch-line-counts AS SUCH

10:44

punkman certainly a worthy attempt, even if mod6 prolly won't be able to put it in current release.

10:45

asciilifeform well yes, but not what he said.

10:45

what i took it to mean was that if he's gonna load his head with shit, would rather do two things than one and reload

10:45

ok to make the two patches in one sitting

10:45

but they gotta be separate.

10:46

that small logger thing actually looks pretty good

10:46

to my documentedly noobish eye.

10:46

can do the gui-snip thing first, work is already done

10:46

incidentally, why the fuck does bitcoin have json support ? do we actually want this ?

10:47

mircea_popescu: it is still there because that is how the rpc thing works

10:47

yes but fuck that, we have a protocol that then imports json ?

10:47

need to have a format for api output anyway right?

10:47

i very dearly wanted to rip it out, but then all the robotics folks are presently 'come to expect' will have to be scrapped.

10:47

this strikes me as the same fundamental insanity of "we'll replace bdb with leveldb in the sense that WE KEEP BOTH!!1".

10:47

if you recall, i wanted to replace it with unix signals.

10:47

or at the worst, a simple telnet-type protocol

10:47

i do but i don't recall why i wasn't so hot about that.

10:48

because it sux

10:48

but unix doesn't give us a rpc-thing that doesn't suck

10:48

yes because it sucks lol. if you're happy with that much.

10:48

standard format would be nicer than new invention

10:48

punkman only if "standard format" doesn't expand at pre-compile time into "here's a debalanced tree of 547657698 "standards""

10:48

punkman: when the standard does not fit-in-head and yields 200kB of heathen code - it can go die

10:49

it is ipso definitio not a standard.

10:49

there are some tiny json libraries

10:49

just a glorified mess.

10:49

the implementation is not the standard.

10:49

punkman: any written by wot men?

10:49

no

10:49

even before wot issue considered : there's two branches to corectness.

10:49

one is code, the other standard. it helps me nothing that you've made one small if this makes the other huge.

10:50

sorta what i meant in earlier thread, http://log.bitcoin-assets.com//?date=21-01-2016#1380326

10:50

Logged on 21-01-2016 22:16:42; asciilifeform: http://log.bitcoin-assets.com/?date=21-01-2016#1380266 << yes! and you can almost ~define~ 'sane system' this way. with the caveat that it can't be a non-orthogonal fudge like a program consisting of one line 'start_bitcoind()' nor anything in that spirit. ORTHOGONAL operators.

10:50

i guess so.

10:50

~genuine~ reduction in complexity specifically excludes shenanigans where it is 'swept under the rug' and hidden somewhere else

10:54

well, something for the foundation to decide, whether json is to be kept or replaced with something else

10:54

it is intrinsically retarded

10:55

and the logical replacement is sexpr.

10:55

which is 'standard' that fits in ONE MOTHERFUCKING PARAGRAPH

10:55

and correctly implementable by a stray dog

10:55

[MPEX] [S.MPOE] 88000 @ 0.00056176 = 49.4349 BTC [+] {5}

10:56

so if I talk to to bitcoind with python I get sexpr in ascii or otherwise?

10:56

in ascii.

10:57

json is really sexpr but was dropped as a baby

10:57

all sv garbage is of this general flavour, incidentally

10:58

could have both and break compatibility later

10:58

i.e. something built by the hands of ACTUAL people, that WORKED, but taken by jolt-cola-drinking jwz types and 'hacked!111 1333773d' into... what we have

11:00

anyway i did not touch the json thing for the same reason i did not touch the whitespace formatting

11:00

... because it does not solve any of the dire problems of the era

11:00

'deck chairs on titanic'

11:02

what is there to do on titanic other than abandon ship?

11:02

the correct model is that there are no other ships.

11:02

and the ocean is made of boiling oil.

11:03

http://log.bitcoin-assets.com/?date=23-01-2016#1382747 << ftr i do not like this one bit. WHY ADD 1000 LINES OF ????

11:03

Logged on 23-01-2016 15:11:54; punkman: https://github.com/SergiusTheBest/plog interesting logging library, less than 1kloc, includes simple log rotation

11:04

this is how we got prb.

11:04

'let's add this useful!11111 pheature'

11:06

and if prb had sane logging we wouldn't be having this discussion

11:07

mno.

11:08

it is bad enough that trb has 500 man-year of study baked into it. let's now add a 501st ?

11:08

fuck that.

11:09

ergo replacing, e.g., openssl, with a 1000 line thing THAT CAN BE UNDERSTOOD, is a worthy thing

11:09

adding an ~additional~ turd to the ball of shit, is not.

11:10

removing boost would be a worthy thing. BUT NOT if it multiplies the line count 2x, OR if it entails forcing gcc5 build

11:10

(as discussed in the c++11 thread)

11:11

~likewise~ any change that increases substantially the number of lines written by non-wot folks, is ipso facto a loss.

11:11

i thought all of this was elementarily obvious.

11:12

with that in mind, YES, the logs need timestamps, or better still, an adjustable format, verbosity levels, etc.

11:12

BUT NOT by adding some derp's 'log library'

11:13

'let's add the library' is, in mircea_popescu's well-chosen wordz, 'the hole through which the night slides in'

11:14

I didn't even dare suggest adding it of course!

11:21

[MPEX] [S.MPOE] 135000 @ 0.00056336 = 76.0536 BTC [+]

11:29

[MPEX] [S.MPOE] 47450 @ 0.00056336 = 26.7314 BTC [+]

~ 27 minutes ~

11:57

adlai

http://log.bitcoin-assets.com/?date=22-01-2016#1381103 << seems like everybody who doesn't lurk in #crypto-crackpots missed the joke

11:57

Logged on 22-01-2016 16:17:47; PeterL: http://log.bitcoin-assets.com/?date=19-01-2016#1376728 http://log.bitcoin-assets.com/?date=19-01-2016#1376731 << Later I realized why I missed the isomer joke: L is isomer of D, R is isomer of S, not quite interchangeable

~ 49 minutes ~

12:46

http://log.bitcoin-assets.com/?date=23-01-2016#1382818 << i think this is so.

12:46

Logged on 23-01-2016 15:55:03; asciilifeform: and the logical replacement is sexpr.

12:46

heh im bashing that baby stuff. alf huffed the right paint this morning :D

12:47

achtung, panzers! does 99.39.110.89 (99-39-110-89.lightspeed.gdrpmi.sbcglobal.net) belong to one of ye lot ?

12:49

http://log.bitcoin-assets.com/?date=23-01-2016#1382830 << not so. the first step of work is cleaning out the workspace. it is the work of truly inept management to impede on this "because it does not directly contribute to productivity"

12:49

Logged on 23-01-2016 16:00:29; asciilifeform: 'deck chairs on titanic'

12:52

re BingoBoingo 's https://voat.co/v/fatpeoplehate/comments/808993/4006420 << check out the ugly head of puritanism rearing its head yet again. "if i don't enjoy anything in the world i will surely be spared doom!!1". nothing's quite as resilient as idiocy.

12:52

Ruston's Guide to Spotting and Avoiding Fatfishing | fatpeoplehate ... ( http://bit.ly/1Qre8hh )

12:53

http://log.bitcoin-assets.com/?date=23-01-2016#1382843 << in italian accent "Did you ever think about what you're going to do on the toilet? (yelling) What are you going to do on the toilet!"

12:53

Logged on 23-01-2016 16:09:28; asciilifeform: adding an ~additional~ turd to the ball of shit, is not.

12:54

l0l! is this a ref to something?

12:54

https://www.youtube.com/watch?v=j2A9vRjwfQY << with interpretation.

12:54

Seinfeld - Edward Scissorhands BEST SCENE EVER - YouTube ... ( http://bit.ly/1PurcNu )

12:56

anyway. it's one thing to add "a library". it's another to "import this 1k lines that i read myself".

12:57

if gonna add 1k lines, make it... tinyscheme-89

12:57

implement tx debugger, sane logger instrumentation, and 101 other things, IN IT

12:58

now tinyscheme may be a good import

12:58

so perhaps i can now admit to having this quietly on my hdd.

12:58

but conceptually, in my head at least, more of a "bitcoind migrates into bitcoinos" thing

12:58

aha. slowly cut off chunks & replace with it.

12:59

until the whole thing, one fine day, starts just fine in a mildly massaged sbcl

12:59

and nobody even notices the cpp is gone.

13:00

anyway, as a general point : even if you did compute a fastest route, and even if in fact you are correct (which is fundamentally an undecidable problem), there is still no reason to panic if phenomena goes a slightly more circuitous path. there are at least two good reasons, both political. one's education (ie, benefit of the ingroup), the other ... well let's just say the other's for the benefit of the outgroup.

13:00

s/jsonrpc/ipc socket/g

13:00

asciilifeform: sbcl?

13:00

holy shit why does that substitution need recursion.

13:01

thestringpuller: http://sbcl.org

13:01

... ( http://bit.ly/1QreCnu )

13:01

thestringpuller: my personal favourite readily-available lisp runtime

13:01

and plug-in whatever you like

13:03

i barely "know" C, how am I ever gonna learn LISP?!?!?

13:03

LISP is like the masters saw.

13:03

thestringpuller: will have to spend a weekend, possibly risk opening a book

13:05

[MPEX] [S.MPOE] 123300 @ 0.00056234 = 69.3365 BTC [-]

13:06

"When you handle the master carpenter's tools, chances are that you'll cut your hand."

13:07

I want to be competent/sane before attempting to pick up master carpenter's tools.

13:09

mircea_popescu: unrelatedly, i now envision your toilet-of-the-future as looking rather like a cray-II.

13:10

lol!

13:10

for the record, no girl liked the idea.

13:10

cray 1 rather

13:10

oh? why not

13:10

"would this be a punishment toilet ? for like the dungeon ?" "do you think the jacuzzi is a punishment pool ?" "ewww!"

13:10

now doesn't want to go into jacuzzi.

13:11

l0l!

13:11

http://www.digibarn.com/collections/systems/crays/cray1/portrait/DSC08196.JPG << for reference

13:11

... ( http://bit.ly/1PusacC )

13:12

anyway, my idea was originalyl that it's a collaborative tool par excellence. girls better support each other's back, lest they fall in.

13:12

but i suspect this heavy... butted aesop is a little too hamfisted for the actual butts in question.

13:12

oh huh

13:12

with no backs, interesting.

13:13

i really pictured more of a cray.

13:13

(vs gigantic ivan iv - style cauldron)

13:13

http://static.dudeiwantthat.com//img/household/bathroom/jacuzzi-entertainment-system-5501.jpg << in other news, the new york apartment of the future.

13:13

... ( http://bit.ly/1PuslVi )

13:13

https://mattnt.files.wordpress.com/2011/12/cray.png << this one?

13:13

thestringpuller: yes. see also earlier link.

13:15

dude speaking of "provability" bs. FSB are "provably secure" because it can be proven they're at least as hard as regular syndrome decoding. which is np complete. while we don't actually know if THAT is in fact resolvable in polynomial time or not, nevertheless... PROVEDLY SECURE!!11

13:15

idiots.

13:16

"this bag is as heavy as that car. we don't know whether that car is just a cardboard cutout or what, but hey, provedly heavy bag! !1"

13:16

don't expect ~sane~ crypto out of the leprosorium

13:16

this is elementary.

13:17