340900+ entries in 0.21s
ascii_butugychag: if your box pipes it
to bash and executes,
that is a problem with you, not 'blockchain'
hdbuck: just was quite amused by
the latest Kaspersky Interpol Forbes combo virus
thestringpuller: So no, I don't buy storage often. I usually
try
to buy enough for 5 years at a
time.
thestringpuller: My most powerful box runs i5 a 1.5
TB array. Best
thing in
the box is a 960 GTX
ascii_butugychag: about
to upgrade cpus for
the 1st
time, with.... circa-'09 ones
thestringpuller: When I get promoted
to plutonimum mine from coal mine, maybe I'll buy SSDs.
thestringpuller: ascii_butugychag: Not all of us are rich like you and can afford
toys :P
trinque: the SSD has *massively* sped up
the process, which doesn't surprise me
ascii_butugychag: ;;later
tell mircea_popescu your scheme (could call it... nodecoin?) finally clicked in my head. at last, an end
to
the cn pond scum !!
trinque: the same node was at 350k before I had
the SSD installed, so we're gaining ground quickly.
trinque: trb node is at 384k; I will be switching
to
that with anti-heathen command patch very soon
trinque: still seeing something
that knocks btcd into a stuck state
trinque: jurov: I got it
to squirt your
txn earlier, after finding where my db lost
track of
the blockchain. site will update when btcd catches up, in roughly 90blks
ascii_butugychag: ;;later
tell mircea_popescu in what sense is adoptinc keccak a rejection of usg standards? it was actually adopted as sha3...
☟︎ polarbeard: you can press
the corresponding hunks,
the graph is showing
that
phf: there needs
to be some patches in between
phf: polarbeard: you can't press polarbeard_better_log_messages directly on
top of asciilifeform_maxint_locks_corrected period. so
the ~transition~ makes no sense
phf: ascii_butugychag:
thanks!
ascii_butugychag: it is my considered judgement
that maintaining
the illusion of 'we got it right
the first
time' is a pointless
thing.
polarbeard: phf: I don't follow, db.cpp is what generates
that relationship
ascii_butugychag: and now, e.g., shiva is hosed, imho needlessly, and will have
to be reground
ascii_butugychag: mod6: i found myself blowing
the very
tiny
time budget i had, on regrinding
mod6: and a similar situation in V would be like 69 patches
to patch
the original patch.
mod6: ascii_butugychag: one fix every now and
then isn't horrible or anything, id just hate
to see
the situation (like at usg-like job) where people create a
turd, create a PR, and
then have
to keep chainging
the PR until "ready"
ascii_butugychag: if it were given a place
to pump in patches, we could finally give
turdatron a proper burial
ascii_butugychag: phf: your viewer is mighty spiffy, it is exactly what i wanted
to make in september (and never had
time)
☟︎ phf: in other words
there might be an antecedent, but
there's also other hunks in
that file
that should've prevented
the linkage
ascii_butugychag: i still
think regrinds are a
thing
to be avoided whenever possible, but whatever
phf: polarbeard: it's not correct, since
the graph is supposed
to only show
transitions, rather
then descendants/antecedents. in
this case polarbeard_better_log_messages
touches util.h and util.cpp
that are modified in polarbeard_remove_shrink_debug_file
polarbeard: I
think it's correct
though, 565faf3ef371f5e2178ae30c45b08b93415eeb92263486e68f2ac2e8f4c7900056e628804bf5c0707a90be946e0aeaebfcd0a391aab40de2e5d56e6bcbdccb1e bitcoin/src/db.cpp
phf: there's a bug of some sort, pretty sure your polarbeard_better_log_messages shouldn't press from asciilifeform_maxint_locks_corrected, i'll look into it
tonight
phf: which kind of broke
the previous ip only links, i might fix
that at some point
mod6: so
that looks like its doing
the right
thing, and
the log was updated as well.
mod6: compiled just fine, just kicked it off for a quick
test...
mod6: anyway, i created a new patch for PVS and did a regrind on malleus (since
this one depends on main.cpp)
mod6: probably was just going
to fast.
mod6: hey ascii_butugychag, so i figured out what my dumbass did yesterday. somehow, when looking at your fix patch - I missed
the portion where you changed main.cpp altogether.
kakobrekla: which are
the golde age ones according
to alf?
mircea_popescu: "all 1 comments - sorted by: best
there doesn't seem
to be anything here" << lol reddit.
ascii_butugychag: in other nyooz,
turns out
that golden-age opterons (just
the chip) can be had for
their weight in mere gold on 'ebay'
jurov: eulora je
trochu ako randomizovany eve online :D
mircea_popescu: +/** Add entropy
to
the pool directly. Use
this for seeding or on-demand entropy. */
mircea_popescu: yes usg loves
to present "fixes" of
this nature.
the odds of
them making it in are about
the same as
the odds of obama being smart.
mircea_popescu: funkenstein_ <-- afaik
the only ways are 1) require coinbase receiver's signature in block header <<
this is rank nonsense. i'm not signing
to receive bitcoin.
ascii_butugychag: the measures against
toy choppers is 100% 'majesty of
the might of
the state' crapolade
theatre
assbot: Logged on 01-02-2016 16:37:42; punkman: ascii_butugychag: did you figure out how
to "non-outsourceable"?
ascii_butugychag: it remains
to be shown
that, like carnot cycle, mining cannot be escaped from.
ascii_butugychag: introduces exactly
the wrong incentives, and empowers
the folks you least want in
the driver's seat
ascii_butugychag: the more i
think about it,
the more it seems
to me
that mining is a catastrophic bug per se
mircea_popescu: ascii_butugychag yes but
the solution is not
to kill all
the sheep.
BingoBoingo: <punkman> BingoBoingo: dunno, wasn't
that interesting in retrospect << As a general rule cases become far less interesting when pleas happen
punkman: ascii_butugychag: did you figure out how
to "non-outsourceable"?
☟︎ ascii_butugychag: pools full of sheeple who can be steered,
turned on a dime, are a Bad
Thing
mircea_popescu: why. because
that's how
the whole world was built. and how it works when it works.
mircea_popescu: currently
the problem appears
to be "it being outsourceable", but really,
the problem is
that only blind outsourcing can be done.
mircea_popescu: see, i suspect ascii_butugychag
this is
the major contradiction here. your idea of safety is
to make it non-outsourceable, but i suspect it'd be both better and safer
to make it FAIRLY outsourceable.
adlai: i just want
to understand what ascii_butugychag suggested
there, instead of each block header including
the previous header's hash. i probably got
thrown off by
the word "nonce" leaking in
there
mircea_popescu: the wunderbar advantages of being explicit! others have something
to
talk
to!
mircea_popescu: "do you know
that currently, all a miner needs is
to know
the header of
the previous block
to mine" "yes"
adlai: i didn't ask what
the implications are, or why, i'm asking what exactly your proposal is, implementation-wise
mircea_popescu: wouldja simmer down and speak explicitly ?
these are
the logs!
ascii_butugychag: at
the root of both, is
the fact
that anything produced by 'coderz' is braindamaged.
adlai: ascii_butugychag: what did you mean by "requiring whole-block hash in
the nonce"?
ascii_butugychag: cpp is braindamaged and is
the reason why winblows is what it is.
adlai: punkman: right-shifting
the nonce bits by N
to make
the hashed data update every 2^N blocks