06:48 |
crtdaydreams |
http://logs.bitdash.io/pest/2022-11-02#1015216 << putting the name to the face is initially quite jarring |
06:48 |
bitbot |
Logged on 2022-11-02 10:43:52 asciilifeform[4]: shinohai: photo is defo of same fella ( met him in at mp's thing in '14 ) |
06:54 |
crtdaydreams |
serves as a pertinent reminder there is a human behind the bitstream |
| |
~ 4 hours 27 minutes ~ |
11:21 |
phf |
http://logs.bitdash.io/pest/2022-11-02#1015413 and http://logs.bitdash.io/pest/2022-11-02#1015349 what it looks like and i've never seen something like that before, is that it's a wp targeted attack, with a sprinkle of text recognition. could be for example a custom template, or a template patch that's based on recognizing |
11:21 |
bitbot |
Logged on 2022-11-02 23:07:03 crtdaydreams[jonsykkel|awt|signpost]: http://logs.bitdash.io/pest/2022-11-02#1015379 << dun make sense, site itself still links to fpga stuff and shithub links to site |
11:21 |
bitbot |
Logged on 2022-11-02 17:35:16 phf[awt]: shoulda used mp-wp |
11:21 |
phf |
existing template conventions. so you plop it in and get a spam/legitimate hybrid, that it's hard to figure out what's going on. really quite clever |
| |
~ 24 minutes ~ |
11:46 |
asciilifeform |
phf: erred on side of pure spamola tho ( asciilifeform looked twice, and not noticed the older material, until 3rd time when cdd pointed out ) |
| |
~ 2 hours 16 minutes ~ |
14:02 |
signpost |
http://logs.bitdash.io/pest/2022-11-02#1015391 << dang, xorshift is simple! |
14:02 |
bitbot |
Logged on 2022-11-02 19:27:14 asciilifeform[4]: signpost: try marsaglia |
14:03 |
signpost |
working it into the thing atm. |
14:10 |
asciilifeform |
signpost: had notion it came up in ye olde #t; and sure enuff, diana bolted a marsagliator to asciilifeform's udp tester back in the day |
14:11 |
asciilifeform |
or hm wasn't a marsaglia was it |
14:11 |
asciilifeform |
was 'twister' |
14:11 |
signpost |
yeah, xorshift's like 5 lines of logxor and ash eh? |
14:11 |
signpost |
well in lisp |
14:12 |
asciilifeform |
aaha, iirc fastest prng with decently long period |
14:12 |
* |
signpost just needs a decently normal distribution so should be great |
14:15 |
signpost |
https://www.jstatsoft.org/article/view/v008i14 << paper for the interested; tried to huck the pdf link into archive.is but it wouldn't eat it, perhaps because it's a download. |
14:16 |
asciilifeform |
aaha, for this use case only needs decently long period ( and uniformity, which can then hammer into normal bell ) |
14:17 |
* |
signpost back in a bit |
| |
~ 4 hours 59 minutes ~ |
19:16 |
phf |
i'm surprised every time i visit cmu ai archive AND IT'S STILL UP |
19:25 |
asciilifeform |
lol went to look, and it dunload |
19:30 |
phf |
well for some reason it's very very slow, but… |
19:37 |
phf |
а ведь мог бы и полоснуть! |