log☇︎
152900+ entries in 1.144s
asciilifeform: yes, this means that martians can land, mine the remaining blocks in five minutes, and leave earth a smoking and bitcoinless crater
asciilifeform: it was never a measure of time
thestringpuller: BingoBoingo: Really as fragile as the thing is atm fuckers ruined my dreams << twas a good dream
asciilifeform: advertised in a node userid, '/btko.in:0.11.99/ (70002)'
asciilifeform: on account of being the only node on the planet that passes the idiots' 70001 filter while ~not having a useragent string~
asciilifeform: apparently we don't even ~have~ a 'useragent' string in 0.5.x
mats: am i supposed to parse the html or something for a response?
asciilifeform: it takes a standard POST
asciilifeform: tried to suppress abortion, even (a laughable nonstarter in ru) and 'fight alcoholism', etc, etc.
asciilifeform: i want a chart that i can regenerate.
decimation: but all you want is a chart
assbot: graphviz - Create a call graph for a file with clang - Stack Overflow ... ( http://bit.ly/1H5oriF )
asciilifeform: http://stackoverflow.com/questions/15458738/create-a-call-graph-for-a-file-with-clang
asciilifeform: anybody got a clang build?
decimation: asciilifeform: one would think the compiler writers would need such a thing
decimation: the virtual functions thing probably eliminates all halfass attempts to use a C based-tool
asciilifeform: i can't use a graph which contains munged names or misses virtual functions.
asciilifeform: the other thing is that we DON'T NEED A DYNAMIC CALLGRAPH FOR ANYTHING !!!!
asciilifeform: unrelated: folks competing for mircea_popescu's 1 btc prize for therealbitcoin callgraph can strike 'egypt' utility off the list. it is a worthless pile of shit, which produces worthless piles of shit (at least for cpp proggy) ☟︎
asciilifeform: decimation: my pogo didn't come with a gimbal telescope; did yours ?
decimation: unfortunately blockheight is only a solid concept in retrospect ☟︎
assbot: Logged on 06-07-2015 23:23:15; danielpbarron: http://log.bitcoin-assets.com/?date=05-07-2015#1188147 << now that you mention it, it occurs to me that my 0.7.2 node had been fully synched (until the recent wedge) running on a single 5400 rpm drive and 8 gigs of ram
assbot: Logged on 06-07-2015 22:36:48; danielpbarron: ;;later tell asciilifeform I get a segmentation fault http://danielpbarron.com/blkcut.txt
danielpbarron: http://log.bitcoin-assets.com/?date=05-07-2015#1188174 << it's a +1 ! those usually come with "details" like "fresh blood" ☝︎
danielpbarron: http://log.bitcoin-assets.com/?date=05-07-2015#1188147 << now that you mention it, it occurs to me that my 0.7.2 node had been fully synched (until the recent wedge) running on a single 5400 rpm drive and 8 gigs of ram ☝︎☟︎
TomServo: pete_dushenski: Ah cool, I'll give it a go. Thanks.
pete_dushenski: first you unplug the usb drive from inside the edgerouter, plug it into your computer, install openbsd on it, unpack it, reinstall the drive in the edgerouter, establish a serial connection, and start talking.
TomServo: But last I read booting OpenBSD required booting over the network, rather than install on the USB drive due to lack of a USB driver
pete_dushenski: TomServo i found another computer with a serial port that actually talked to the serial cable i have. the usb adapter i bought was b0rked.
TomServo: pete_dushenski: Didja find a way to boot OpenBSD on the edgerouter from USB or are you using the nfs boot option?
danielpbarron: http://log.bitcoin-assets.com/?date=06-07-2015#1189808 << can attest to this having opened 12 in a row today, although it goes sorta smoothly once you get the hang of it ☝︎
danielpbarron: ;;later tell asciilifeform I get a segmentation fault http://danielpbarron.com/blkcut.txt ☟︎
shinohai: ^ 10 BTC is steep for a domain name, but I like it
ascii_field: punkman: keyserver is presumably for use as a keyserver, rather than place to hang the severed heads of pwned folks ?
ascii_field: speaking of which, i've often thought of adding an 'i know a private key!' box to phuctor
assbot: Logged on 06-07-2015 21:11:18; thestringpuller: ascii_field: did you get past the wedge yet in 0.5.3.x? or do you still not have access to a node?
gribble: Error: "ba,bull" is not a valid command.
thestringpuller: ascii_field: did you get past the wedge yet in 0.5.3.x? or do you still not have access to a node? ☟︎
mircea_popescu: ascii_field gotta be a lot of magic numbers
pete_dushenski: "Most of those new lines of code come from “the new amd gpu register description header”, code that Torvalds says comprises “41 per cent of the entire patch” and has created a “... somewhat odd situation where a single driver is about half of the whole rc1 in number of lines.”"
ascii_field: a place to farm to death, to the bedrock
assbot: Logged on 04-07-2015 22:28:31; mircea_popescu: anyway, the main sufferance in my head atm is that reading the chinese stuff (in translation) clarifies in my head an objection that may well be a second major flaw to the protocol, after the "relay nodes gotta do it for the glory" : its altogether unclear a purely financial incentive is the correct solution for miners.
trinque: is this a worthwhile experiment?
ascii_field: jurov: let me guess, nobody's written a multicore verifier ?
ascii_field: and need a couplea catastrophic megadeaths
jurov: estimated time to fully verify a block is 20 seconds
mircea_popescu: this is not a fully virtual system
trinque: the message processing story in bitcoin is a shitshow
pete_dushenski: "Reports that the situation has passed are not correct. Please continue to wait 30 more confirmations than you usually would wait before accepting a transaction."
ascii_field: there's gotta be a way out.
pete_dushenski: "6 July 04:00: A new fork occurred starting 5 July at 21:30 with three blocks before the valid chain again became the strongest chain."
mircea_popescu: just as long as we put THE REAL TIME as a txn in every block, pogos can be fine
decimation: it's come up a couple of times here, but not in the context of the 2h epoch
decimation: who knew that we needed a #b-a astronomer
decimation: right, but that requires a trusted way to ask a bitcoind "what's the timestamp of the last verified block'
ascii_field: is there a notion of p2p time.
decimation: essentially there isn't such a thing as p2p time
ascii_field: they will get apologies and a coupon for free blowjob at community whorehouse
trinque: vs if you are set to yesterday and I give you a block from tomorrow
trinque: say I though a block from last year at your node, I don't see why it would matter that your clock is set to tomorrow
ascii_field: because right now, hitler has, essentially, a button that makes just about the whole btc network shit its pants
trinque: you can of course accept a block from the past
punkman: fried ice cream is a thing
trinque: anybody ever looked at tinc as a starting place for a gossipd?
ascii_field: the latter turns the node into a kind of boat anchor while active
assbot: Logged on 06-07-2015 19:10:06; punkman: is this a first?
ascii_field: trinque: adjusting anything whatsoever on pogo is nontrivial for a 'civilian'
trinque: having bitcoin disregard time, which I'm sure has all kinds of implications around when you may consider a given block valid
trinque: ascii_field: yeah, I think that is the only possible way to avoid having a magic central something
ascii_field: back to a magic public key a la gavin ?
decimation: I find the magic box unneeded when I can get rsa signed time from a variety of places via ntp
trinque: and consider certain addresses a valid source of time?
ascii_field: decimation: gonna ship a gigabit switch also?
decimation: that serves time via ntp server and has a push-button input on the front panel
decimation: if we are this paranoid about it, the only solution I can think of is to ship a separate box with an ethernet port
decimation: it was already such a turd, we are just trying to find out what to do about it
ascii_field: and now we've turned bitcoin into a centralized turd
decimation: alright, apparently we need a #b-a radio clock transmitter too
decimation: maybe it can come with a plug-in sextant?
danielpbarron: a few of the inner plastic clips got cracked in the de-shelling process but they should still click back together no problem
ascii_field: jurov: not sure if it is so much of a corner. perhaps it is possible to get a 'good enough' picture of political time via some clever means
jurov: then you painted yourself to a corner and need tiny webserver for user to set time manually
ascii_field: whether a block can be 'latest' depends on local time!
ascii_field: there is no such thing as a 'trusted node'
decimation: why not write a lightweight client that gets the time from trusted bitcoind node network?
decimation: actually happens on linux as I understand. there's a thing
shinohai wants to start a pogo bitnode factory, ensure domination of therealbitcoin
ascii_field: kakobrekla: even so, it is a bog-standard consumer product and could be obtained with massive bulk discount if we were to try in earnest
kakobrekla: ascii_field> kakobrekla: ssd is often 'free' - rip out of a retired laptop < maybe that is true for you and me but not the majority i dont think
ascii_field: kakobrekla: ssd is often 'free' - rip out of a retired laptop
kakobrekla: <ascii_field> trinque: 50 is almost TRIPLE the cost of a pogo < if you dont include ssd you dont have a working node
ascii_field: it'd have to warm up with a few 'addnode's pointing to working boxes, to run at all, yes
ascii_field: given that it turns your node into a boat anchor
ascii_field: even if the thing had a battery-backed otc, there would remain the question of how to set it
trinque: maybe without the rtc it's a leaky boat
trinque: but actual cost of a vessel is not yet determined
ascii_field: a solution that HALVES our fleet is ispo facto retarded!
ascii_field: trinque: 50 is almost TRIPLE the cost of a pogo
punkman: the parasitic timestamp fetching is not a bad idea though
trinque: there's a github repo out there for everything