Show Idle (>14 d.) Chans

← 2022-01-12 | 2022-01-14 →
00:02 whaack billymg: interesting, i got the msg 'ok, restarted' through asciilifeform, although all the other msgs from billymg i get directly
00:02 billymg whaack: how about this one?
00:03 billymg your message just now came through directly btw
00:10 whaack yeah it was just 1 packet seemed to have been drop
00:11 whaack dropped*
00:11 asciilifeform << nfi whether authentic, but lulzy
00:11 bitbot Logged on 2022-01-12 21:29:02 signpost[asciilifeform|billymg]: meanwhile lol @ finally leaking this
00:24 asciilifeform << we're gonna need some means of identifying keys for purposes like this w/out leaking bits. possibly e.g. crc32(sha512(key)). will amend spec after folx comment.
00:24 bitbot Logged on 2022-01-12 22:38:51 billymg: signpost: i have a key beginning in "OqlnKL" for you, and according to my logs i'm sending packets to you, but don't see any received from the ip:port that i have for you
00:25 asciilifeform ( prolly obv imho, but nobody but the 2 peers who are using a key have any biz knowing even leading N bits of it )
00:30 asciilifeform oughn't to commit the idjicy of e.g. gpg (to use this string for anyffin internally) but for 'do you have key X?' oughta suffice imho.
~ 58 minutes ~
01:29 signpost << yeah, could easily interpret as cover for something even more inept/idiotic having happened.
01:29 bitbot Logged on 2022-01-13 00:11:30 asciilifeform: << nfi whether authentic, but lulzy
01:30 signpost << seems reasonable to me. and defo considering any keys with bits in logs trash that'll be rolled to new keys after debugging
01:30 bitbot Logged on 2022-01-13 00:24:47 asciilifeform: << we're gonna need some means of identifying keys for purposes like this w/out leaking bits. possibly e.g. crc32(sha512(key)). will amend spec after folx comment.
01:31 asciilifeform at any rate when we've rekeying this will matter rather less.
~ 12 hours 16 minutes ~
13:47 PeterL << yesterday my blatta crashed, I'm not sure what this output means?
~ 27 minutes ~
14:15 asciilifeform PeterL: this is actually related to the icmp thing.
14:15 asciilifeform PeterL: seems that python's posix glue actually propagates a 'port unreachable' as a eggog.
14:15 asciilifeform awt: oughta wrap that thing in a catcher.
~ 1 hours 46 minutes ~
16:02 whaack a project i'd be interested in with the ppl here is to make a clone/mirror of the books on . I reached out to them to try to ask if i could mirror their site, but they are keeping their data proprietary. i'm sure one day they will be taken out and that site is a resource i would be quite sad to lose.
16:05 asciilifeform indeed, 1337 w4r3z
16:07 whaack asciilifeform: yeah, ty i think it was you who posted that link in b-a years ago, life changing
16:07 asciilifeform iirc did post at least 1
16:08 whaack crazy how much manual intervention must be put in place to hide that website from the search engines
16:08 asciilifeform lotsa reasonably fresh w4r3z there. possibl. the only quality english b00kw4r3z www atm
16:08 whaack although not unexpected
16:12 asciilifeform whaack: imho the current-day engl. warez culture is rather braindamaged -- recall e.g. the 'scihub' folx wanking instead of standing up mirrors
16:16 whaack will read thread later
16:16 whaack i'm guessing that some other academia paper website got nixxed and never took the time to make backups?
16:22 asciilifeform whaack: there's a 'dog in the manger' culture, they like to 'sit on' 'their' warez
~ 1 hours 7 minutes ~
17:29 signpost whaack: quite interested in working with you to make warez great again.
17:31 signpost impl wot-tronics, a dht, and metadata search on pest and it'll put historic solutions to shame.
17:31 * signpost recalls the glory of, tragedy that it died.
17:33 signpost on the subj, the "ratio" method used to encourage members of private trackers to "contribute" is brain-damaged, considers everyone equals, and *limits* distribution due to the poverty of the perspective.
~ 17 minutes ~
17:50 asciilifeform signpost: recall the days of bbs and warez ratios. folx would upload rngolade to 'make ratio', lol
17:51 asciilifeform yet anuther illustration of what happens when lacking wotronics
17:51 * asciilifeform not heard of ''
18:04 whaack asciilifeform:, from what i remember, was a private stash of music. it was not trivial to get access to this stash, iirc it was invite-only and bar getting a lucky invite, you had to contribute to be able to access the 'warez.' as i understand, the music stash was legendary, it had ~errything, including lots of old
18:04 whaack great interface for viewing and searching.
18:04 * whaack knew someone in college with access.
18:04 asciilifeform a
18:04 whaack i didn't know it had died though, and signpost probably knows more
18:08 jonsykkel was gr8, now replaced by with identical interface. tho only some ~25% of warez from
18:09 whaack signpost: maybe it'll be a good project after i'm done with trbexplorer
18:12 asciilifeform 100% wotronic warez on the proper foundation offered by pest is imho an obvious next step for the thing.
~ 25 minutes ~
18:37 billymg whaack: glad to hear you're back on the trbexplorer
18:40 billymg i'm interested in using it to generate something like the "chainalysis" folks generate for
18:40 asciilifeform billymg: i strongly suspect that these only look as 'readable' as they do on acct of manual fiddling
18:41 asciilifeform (rather than being purely mechanical output of whatever proggy)
18:42 * asciilifeform recalls the 'ohnoez pankkake sent coin to trump' or wat was it, tabloid thing
18:42 billymg asciilifeform: yeah, i suspect there is a lot of manual "tagging" involved, and a parallel db of all they know about extant addresses
18:43 * billymg in theory thinks it'd be nice to have a way to answer "what do they know about this address?" -- though perhaps not possible
18:43 asciilifeform billymg: not only manual tagging, but i suspect even the readable (i.e. not 'help mouse find the cheeze' mess of liquishit) in the graph is also manually produced
18:43 billymg ahh
18:44 asciilifeform (recall the time when asciilifeform asked for a callgraph of trb, and no one could produce anyffin other than a 'help mouse find cheeze' horror)
18:51 asciilifeform laying out a gnarly graph in 2d such that a human can make sense of it w/out a pen and magnifying glass and over9000 time is hard. (in some cases may be impossible)
18:52 asciilifeform simply getting the connections not to overlap may be np-complete iirc.
18:54 asciilifeform (it is a very similar problem to pcb routing, in fact)
18:54 whaack billymg: nice, it is near completition, i just need to add the tracking of a few other details (fees, fees paid to miners), adjust the view-block command so that it shows txns, and test the push command. aim is to finish by the end of February
18:55 whaack if there's some specific type of query that you would like added or that anyone thinks is missing from typical heathen explorers plz let me know
18:56 asciilifeform whaack: sumthing that exposes the 'who-gave' info of local noad would be nifty imho
18:56 asciilifeform (i.e. 'where didja get that block and when')
18:56 whaack ah, who gave that block?
18:56 asciilifeform aha
18:57 asciilifeform currently visible in debug log (if yer noad has asciilifeform's whogave patch)
18:57 whaack is that info available via the rpc?
18:57 asciilifeform nope
18:57 asciilifeform prolly oughta
18:57 asciilifeform it aint currently stored anywhere (other than log spew)
18:57 whaack hrm, not sure i'm going to include that in the genesis patch then
18:59 whaack not that it's not an important feature, but it doesn't quite fall in the scope of the current mvp, which is to replace all use of heathen blockexplorers.
19:01 billymg whaack: awesome, very much looking forward to this
19:01 billymg how big was the db again?
19:01 billymg iirc was like 1tb, on top of the trb blockchain
19:03 whaack billymg: it's currently sitting at 513GB
19:04 asciilifeform the big win will be when we have the 1 sane db for the thing, and not need sqlisms etc on top of it
19:04 whaack my HD is using 930GB, which includes the OS and a couple of other small files
19:05 asciilifeform ( ftr if stored via asciilifeform's scheme oughta weigh <half of what the traditional format weighs, while being readily convertible to/from it for protocol purposes )
19:05 whaack well how do we ever get that asciilifeform the whole problem as i understand it is we need to keep trbs db around because we need some authority as to what is the longest chain / correct fork and any attempts at restructing the database has a high probability to lead to a fork ala the berkly / level db one a while back
19:05 asciilifeform ( the trad format stores tx hashes all over the place rather than indices into the past like it oughta )
19:06 asciilifeform whaack: asciilifeform is satisfied that his scheme represents the original w/out info loss ( ran entire chain at the time through it and compared bitwise )
~ 1 hours 36 minutes ~
20:42 asciilifeform ( concretely -- had it eat ea. block which existed at the time, and shit back out, reconstituting the given one )
20:43 asciilifeform (it still needs the cross-indexing thing tho, not to mention the backing store per se, to be glued on)
~ 1 hours 24 minutes ~
22:07 billymg not sure if long message handling has been discussed yet (or if this qualifies), but i noticed that bitbot received a truncated copy of this message
22:07 bitbot Logged on 2022-01-13 18:40:22 billymg: i'm interested in using it to generate something like the "chainalysis" folks generate for
22:09 billymg in my main station's logs it shows the full message as outgoing, ending in "for the spooks". in the bot's station's logs it shows the truncated version as incoming, cut off right after "for"
22:13 billymg ah, it's in the spec (thought i remembered something about it)
22:13 billymg "Given that a Pest message is able to carry no more than 324 bytes of Text, a single IRC message entered into the console may require the station to originate two Pest messages. They must be appropriately chained, and their Timestamps must be equal. The receiving station will process them in the correct order using their S
22:14 billymg "...using their SelfChain values."
~ 29 minutes ~
22:43 asciilifeform billymg: indeed is in the spec
22:43 asciilifeform currently afaik blatta simply cuts irc msg into pestable chunks (or tries to at any rate)
← 2022-01-12 | 2022-01-14 →