09:56 |
shinohai |
$ticker btc usd |
09:56 |
busybot |
Current BTC price in USD: $20081.69 |
09:57 |
shinohai |
http://logs.bitdash.io/pest/2022-10-03#1013785 << thx again signpost, knobs set and bot station updated. |
09:57 |
bitbot |
Logged on 2022-10-03 18:35:00 signpost: shinohai: http://logs.bitdash.io/pest/2022-07-11#1009305 |
| |
~ 1 hours 21 minutes ~ |
11:19 |
shinohai |
$uptime |
11:19 |
busybot |
The bot has been up for: 1 hours 24 minutes and 6 seconds |
| |
~ 4 hours 13 minutes ~ |
15:33 |
phf |
to the previous thread http://logs.bitdash.io/pest/2022-10-02#1013744 volume 4 pre-fascicle 7a https://cs.stanford.edu/~knuth/fasc7a.ps.gz says, unsurprisingly, that it's from fasc7a.dvi, TeX output 2022.08.01:1918 |
| |
↖ |
15:33 |
bitbot |
Logged on 2022-10-02 23:52:21 asciilifeform[6]: then again whoknows just with what baked |
15:33 |
phf |
as opposed to e.g. pdftex |
| |
~ 18 minutes ~ |
15:52 |
jonsykkel |
http://zzz.st/up/I78PRY3h/ ur mesage as i se it contains 0x0a |
15:53 |
jonsykkel |
after it not shows up in logs |
15:55 |
phf |
ah, cute |
15:56 |
phf |
well i picked up #x0a from slopy copy/paste, but we kind of know that blatta is funny about newlines, because of the whole irc compat |
15:57 |
jonsykkel |
inded wont work wiht any irc interface |
15:59 |
phf |
that's not true, one could figure out reasonable semantics for the boundary between on the wire pest, and the payload presentation through irc interface |
15:59 |
jonsykkel |
u mean spliting the mesage into seprate irc commands |
16:00 |
jonsykkel |
sure |
16:00 |
phf |
yah |
16:00 |
phf |
it will kind of break the logs, because logs are talking irc, rather then directly to database, so you'll have separate messages for non-separate messages, but honestly i don't see any reason to be particularly pedantic about this |
16:02 |
phf |
i'm too lazy to look at the spec but pest special-cases messages that don't fit into a single pest packet, by splitting them into multiple packets. presumalby on render one could recognize that the message is the same (identical timestamp, direct selfchain) and merge them together |
| |
↖ |
16:04 |
jonsykkel |
could, but stamps are second resolution so u will then have messages incorectly glued all the time |
16:05 |
jonsykkel |
need "more fragments" bit to solve it properly |
16:06 |
phf |
jonsykkel, i think you'll very rarely have messages "incorrectly" glued, since you're matching by self-chain and timestamp being identical |
16:07 |
jonsykkel |
well selfchain always gona match up if u send from same station, and then u just need 2 quick ones wihtin 1sec |
16:07 |
phf |
right |
16:08 |
phf |
i'm arguing the "all the time" vs "rarely", rather than whether it can happen at all |
16:09 |
jonsykkel |
sure, but i think "rarely" will be like 5times in a active day |
16:10 |
phf |
cool |
16:10 |
jonsykkel |
if u have fast finger |
16:10 |
jonsykkel |
like myslef |
16:21 |
phf |
well, according to my tests you're not in danger, the main culprit is deedbot |
16:24 |
phf |
i don't have much packet backlog, september 30th is the first packet, but over that time if one were to merge packets like that, would get http://paste.deedbot.org/?id=MhQR |
16:30 |
asciilifeform |
http://logs.bitdash.io/pest/2022-10-04#1013802 << unsurprising, but raises the q re outta what he bakes the very clean graphs etc (raw tex shat outta a pascal proggy?! or wat) |
| |
↖ |
16:30 |
bitbot |
Logged on 2022-10-04 15:33:04 phf[awt]: to the previous thread http://logs.bitdash.io/pest/2022-10-02#1013744 volume 4 pre-fascicle 7a https://cs.stanford.edu/~knuth/fasc7a.ps.gz says, unsurprisingly, that it's from fasc7a.dvi, TeX output 2022.08.01:1918 |
16:30 |
jonsykkel |
posible that irc client flood protection will save ppl |
16:32 |
asciilifeform |
http://logs.bitdash.io/pest/2022-10-04#1013815 << per current (well, not for much longer..) draft : |
16:32 |
bitbot |
Logged on 2022-10-04 16:02:37 phf[awt]: i'm too lazy to look at the spec but pest special-cases messages that don't fit into a single pest packet, by splitting them into multiple packets. presumalby on render one could recognize that the message is the same (identical timestamp, direct selfchain) and merge them together |
16:32 |
asciilifeform |
'Given that a Pest Message is able to carry no more than 324 bytes of Payload, a single IRC command received from the console may require a station to originate two text Messages. These must be appropriately chained, and their Timestamps must be equal. Receivers will place these Messages into the correct order using their |
16:32 |
asciilifeform |
... their SelfChain and NetChain values.' grr |
16:32 |
phf |
:} |
16:33 |
* |
asciilifeform did not bother to point out that only msgs from given $handle with == times & consec. chains oughta be glued, but hopes is obv |
16:34 |
* |
asciilifeform erring on the side of pedanticism in rewritten spec, can always 'diet' it down later |
16:35 |
phf |
asciilifeform, well, jonsykkel points out that anything that spits out messages under 1s< rate will get magically glued. i suspect that it will mostly happen in bot situations and ocassional "lol spam" |
16:35 |
asciilifeform |
yea was eating log & finally saw, lol |
16:36 |
phf |
per my test above the only "accidental" one is deedbot, and i suspect the way it happend is that two messages that were at distance arrived and sat in the same buffer,then hit the automatic machinery at the same time, so deedbot produced answers in a tight loop, that is in the same second |
16:37 |
phf |
the main issue with "gluing" is whether or not there should be a newline in between glued lines. otherwise the result is visually etc. identical to individual messages |
16:39 |
asciilifeform |
indeed prolly oughta've specced 'piss <= 1/s unless you want'em glued' |
16:40 |
asciilifeform |
hypothetically could've 'glue this' flag etc., but protocol is gnarly enuff w/out such things imho |
16:41 |
jonsykkel |
maybe not such a problem oustide of paste 10lines into chat = have to wait 10sec |
16:42 |
phf |
well, no, "paste 10 lines" should exactly be glued! |
16:42 |
jonsykkel |
or i supose those would be glued yes |
16:42 |
asciilifeform |
aaha |
16:42 |
jonsykkel |
nvm! |
16:42 |
asciilifeform |
was abouttasay |
16:43 |
jonsykkel |
that stupid ircness gose away indeed |
16:43 |
asciilifeform |
eventually 100% of ircism , theoretically, goesaway. (liekly 1st for folx willing to live in emacs / shell) |
16:45 |
* |
asciilifeform in new spec, moved all ircism to an annex |
16:45 |
jonsykkel |
how to even live outside shell |
16:45 |
asciilifeform |
old one is imho horrid mess from this pov, with chaotic mix of front and rear end |
16:46 |
phf |
http://glyf.org/screenshots/pest-as-she-sail.png |
16:46 |
asciilifeform |
jonsykkel: asciilifeform likes frontends where can cleanly copypaste text a la xchat |
16:46 |
dulapbot |
Logged on 2022-07-22 21:01:56 asciilifeform: tk is rather barbaric, but beats the shit outta raw console and its spurious linebreaks in pastes etc |
16:46 |
jonsykkel |
thought about even spliting pestron into pure item with no irc/tcp, with "rpc" inteface (but bidirectional) then have temporary rpc<->irc device as separte item |
16:47 |
jonsykkel |
true copybaste is masive asspain |
16:47 |
asciilifeform |
phf: nifty, loox like has editbox (& presumably cleanly pasteable-from log) |
16:48 |
asciilifeform |
jonsykkel: from shell / curses proggy, afaik lethal pain |
16:48 |
* |
asciilifeform not seen any 100% working examples at all |
16:48 |
asciilifeform |
(and if can't do it w/ middleclick, it aint 'working example' per asciilifeform's lights) |
16:48 |
phf |
yeah, i haven't really tested obnoxious pastes, but should also work |
16:48 |
phf |
Stately, plump Buck Mulligan came from the stairhead, bearing a bowl of lather on which a mirror and a razor lay crossed. A yellow dressinggown, ungirdled, was sustained gently behind him on the mild morning air. He held the bowl aloft and intoned: |
16:49 |
phf |
nd called out coarsely: |
16:49 |
phf |
gurgling in his throat and shaking his head. Stephen Dedalus, displeased and sleepy, leaned his arms on the top of the staircase and looked coldly at the shaking gurgling face that blessed him, equine in its length, and at the light untonsured hair, grained and hued like pale oak. |
16:49 |
asciilifeform |
phf: what didja sew the gui outta ? |
16:50 |
phf |
hmm, i was certain that i've fixed this issue, apparently not. i guess that's how this thread started to begin with |
16:50 |
phf |
asciilifeform, tk |
16:50 |
asciilifeform |
a. |
| |
~ 18 minutes ~ |
17:09 |
phf |
Stately, plump Buck Mulligan came from the stairhead, bearing a bowl of lather on which a mirror and a razor lay crossed. A yellow dressinggown, ungirdled, was sustained gently behind him on the mild morning air. He held the bowl aloft and intoned: |
17:09 |
phf |
—Introibo ad altare Dei. |
17:09 |
phf |
Halted, he peered down the dark winding stairs and called out coarsely: |
17:09 |
phf |
—Come up, Kinch! Come up, you fearful jesuit! |
17:09 |
phf |
Solemnly he came forward and mounted the round gunrest. He faced about and blessed gravely thrice the tower, the surrounding land and the awaking mountains. Then, catching sight of Stephen Dedalus, he bent towards him and made rapid crosses in the air, gurgling in his throat and shaking his head. Stephen Dedalus, displease |
17:09 |
phf |
d and sleepy, leaned his arms on the top of the staircase and looked coldly at the shaking gurgling face that blessed him, equine in its length, and at the light untonsured hair, grained and hued like pale oak. |
17:11 |
phf |
while that fixes an immediate problem, it introduces a new one: if one were to stitch it back together, will lose newlines, because in addition to fitting into a message i also intorduce a split on newline |
17:12 |
phf |
so really blatta should handle newlines in incomming pest messages better, by splitting them into multiple privmsgs, rather than having other clients accomodate this particular ircism |
17:13 |
phf |
awt |
| |
~ 30 minutes ~ |
17:44 |
asciilifeform |
lol! asciilifeform had nfi, wouldn't have dreamed in nightmare that one can stuff >1 newline into irc msg |
17:44 |
asciilifeform |
evidently can. |
17:50 |
phf |
tbh i aways found pastes to be distasteful. i suppose problem could be solved with attached mime annotated files and a whole lot of dwim interface patterns, but in practice haven't seen it done right anywhere |
17:51 |
phf |
telegram used to support ``` foo ```, but then for whatever reason they disabled it |
| |
↖ |
17:52 |
phf |
in the ideal world i want my chat window to be a multiplayer mathematica: inline images, inline graphs, multiline inline text pastes from terminal outputs and such |
| |
~ 4 hours 19 minutes ~ |
22:12 |
asciilifeform |
^ |
22:12 |
asciilifeform |
pastes, naturally, oughta be nelsonian 'transclusions' neh. |
22:13 |
asciilifeform |
http://logs.bitdash.io/pest/2022-10-04#1013884 << what did that notation do ? |
| |
↖ |
22:13 |
bitbot |
Logged on 2022-10-04 17:51:30 phf[awt]: telegram used to support ``` foo ```, but then for whatever reason they disabled it |
22:13 |
* |
asciilifeform not used subj |
22:15 |
* |
asciilifeform has it filed in head as 'aol/skype clone popular in orcistans' but not knows details |
22:17 |
asciilifeform |
imho an 'adult' pestron oughta support inline img, optionally displayed. we aint there yet tho |
22:17 |
asciilifeform |
( and, naturally, they oughta ride on signpost's luby scheme ) |
22:19 |
asciilifeform |
and re: 'multiplayer mathematica', per asciilifeform's lights, it'd follow quite organically from a sane 'single-player mathematica', i.e. 'the computer that aint a mistake' |