Hide Idle (>14 d.) Chans


← 2018-02-20 | 2018-02-22 →
a111: Logged on 2018-02-20 19:25 mircea_popescu: speaking of, what's teh eta on history look like trinque ?
BingoBoingo: ty shinohai, got some edits but good stuff
deedbot: http://qntra.net/2018/02/maduro-government-in-venezuela-attempts-ico/ << Qntra - Maduro Government In Venezuela Attempts ICO
shinohai: ty BingoBoingo !
mod6: Mornin'
shinohai: Heya mod6 :D
shinohai: How goes today?
mod6: Ah, not bad, you?
shinohai: Pretty good here, enjoying the 72 degree weather @ 10 am lol
shinohai: Southern weather is ridiculous
trinque: Houston went from 70s to 30s in one day this winter
shinohai: Wow. I fully expect that to happen here though, it being only February. At least 1-2 more weeks of cold at some point.
mod6: damn 70's eh?
shinohai: I almost want to go fishing today.
mod6: Sounds like fun.
mod6: Right now, it's 7 here.
deedbot: http://qntra.net/2018/02/bittorrent-inc-milks-window-to-let-code-execution-bug-live/ << Qntra - "Bittorrent Inc" Milks Window To Let Code Execution Bug Live
trinque: glad to see qntra in my morning reading again, btw.
mod6: ^
BingoBoingo: trinque: ty, I am trying to get that habit into habituation again
lobbes: Phew, been thinking through this archive download process the past few weeks. For one, I realized that I needed to do -some- type of "deduping" on the contents of each zip I've downloaded. Ended up building a process that hashes the hashes of each file in each zip. Identified some 30k dupes this way. Currently building the front-end queries to show only distinct results in searches.
lobbes: This exercise lead me to think about many other small but important details regarding how download requests will work. I'ma thinking I ought to bake a blog post for general input
trinque: neat lobbes, will read
lobbes: ty trinque
mircea_popescu: http://btcbase.org/log/2018-02-21#1785449 << do you understand merkle trees ?
a111: Logged on 2018-02-21 17:59 lobbes: Phew, been thinking through this archive download process the past few weeks. For one, I realized that I needed to do -some- type of "deduping" on the contents of each zip I've downloaded. Ended up building a process that hashes the hashes of each file in each zip. Identified some 30k dupes this way. Currently building the front-end queries to show only distinct results in searches.
mircea_popescu: the only sane way to go about his is to do poor man's (rather, "artificially intelligent") document ASTs in the shape of merkle trees of all their contents.
mircea_popescu: normally i'd have alf explain it to you ; but he's busy. so what i want you to do is write a blog article about "how to use merkle trees to avoid duplicate problem in web page archiving", and then i'll comment on it.
shinohai: https://www.emsec.rub.de/media/crypto/attachments/files/2011/04/becker_1.pdf <<< Paper on subject of merkle trees I found handy lobbes, for your perusal
lobbes: mircea_popescu: aha. Other than the subj. popping up in logs occasionally and my looking it up I really don't have a firm grasp on merkle trees. I will happily digest and produce blog post once completed with homework.
lobbes: And ty shinohai for reading
mircea_popescu: it'll be quite empowering, so well worth your time.
mod6: lobbes: cool
mircea_popescu: phf is the search brokened ?
mircea_popescu: o wow i hit it exactly ?
mircea_popescu is terrible at time.
deedbot: http://trilema.com/2018/the-republic-without-mp/ << Trilema - The Republic without MP
deedbot: http://qntra.net/2018/02/buttfunder-founder-montroll-arrested/ << Qntra - Buttfunder Founder Montroll Arrested
deedbot: http://barksinthewind.com/2018/vtools-genesis/ << BARKS IN THE WIND - vtools genesis
phf: ^ comments welcome, hanbot let me know if you have problems getting it working, feedback from mp-wp would be appreciated
mod6: phf: was able to press just fine, I did get an error on build, but it still spit me out a vdiff: http://p.bvulpes.com/pastes/5Szrt/?raw=true
mod6: fwiw I used that binary produced, even with the error, to test and see if it would yield the same vdiff as the original would, and so far, looks like it works just the same as printed on the box. I don't have as good of a test case as hanbot will have tho. Mine was very basic.
mod6: here's what my test looks like: http://p.bvulpes.com/pastes/7Y0RU/?raw=true
mod6: again, uber simple, i encourage everyone out there to test with wider testcases ofc.
phf: http://btcbase.org/log/2018-02-22#1785473 << thanks for fast testing! the warning is harmless, i frankly got desensitized to it, and it slipped through the cracks. the entire error.c needs to be cut anyway, but it's a low hanging fruit
a111: Logged on 2018-02-22 00:18 mod6: phf: was able to press just fine, I did get an error on build, but it still spit me out a vdiff: http://p.bvulpes.com/pastes/5Szrt/?raw=true
mod6: alright! My bad before, that problem wasn't an "error" proper, it was a Warning, a Warning in 'error.c'. If I'd read more carefully I'd probably have noticed.
mod6: Anyway, phf did a quick change. It's all good now:
mod6: If anyone is curious what the sha512s are for those vpatches & sigs I used in my latest test, they are here: http://p.bvulpes.com/pastes/vV3ti/?raw=true
mircea_popescu: exciting times.
mod6: yeah, indeed.
mod6: how goes today?
mircea_popescu: pretty great. on the tail end of a "hey, wanna go to the beach ?" "sure!"
mod6: that sounds nice.
ben_vulpes would dearly like to arbitrarily hie to the beach
mod6: ^
mircea_popescu: it's only like 50 miles or so
mod6: that's not too bad. bout an hour drive eh?
mircea_popescu: oh, and also, the police here utterly don't gaf. car doing 150 in an 80 zone, highway patrol right there, couldn't be arsed.
mircea_popescu: it's like a better romania of the 90s.
mod6: nice :]
mircea_popescu: so sunset on teh beach.
deedbot: accepted: 1
← 2018-02-20 | 2018-02-22 →