495900+ entries in 2.481s
ascii_field: but don't
try
to say 'you should like scrolling and pixellated graphics, forget about paper and author-specified pagination'
mircea_popescu: so because -unprocessed- pdf doesn't actually exist you figure
this is an argument ?
ascii_field: this is quite like proposing
that bitcoind be built -on-
the pogo.
ascii_field: but pointing out
that asking printer or
tablet makers
to build machines which can swallow .tex is a nonstarter
mircea_popescu: so does pdf, in same conditions. except you gotta use either a windows bundle or else more perl crud munged
together.
ascii_field: .tex plus illustrations, whatever perl crud etc. was used
to munge data into figures, possible latex preprocessing, etc. sometimes
takes GB of ram and hour of fast cpu
ascii_field: mircea_popescu: you are proposing equivalent of selling raw crude at
the petrol station
ascii_field: and,
to use gnu's
terminology, .tex is often 'not
the preferred format for modification' - as in, it is frequently machine-generated
mircea_popescu: the failure of widgets is a problem of
the widgetmaker. do not ask me
to solve it
till i market a widget.
ascii_field: mircea_popescu: aha lol 'tabletrons' etc. viewer gadgets expected
to render .tex now ?
mircea_popescu: and for
that matter, it's not only
the most valuable
thing you'll ever have - it's outright
the only one.
mircea_popescu: oldest scam in
the book,
this identity
transfer. keep your identity, it's yours.
mircea_popescu: now
that you wish
to be facebook inc, suddenly *they* gotta be you know "people",
to
take
the place of
the vague "we".
mircea_popescu: 'cause
that's all
they know, facebook model. pretend like "we" are interacting with
the stakeholders but stay anonymous, and
then if/when
this gets big enough, assert a new identity and whoopdeedoo, claim you're worth a billion.
mircea_popescu: "o look,
they're hippy dippy cool and got cat pictures". what "they" ? and it's not
THEIR fucking cat, either.
mircea_popescu: no dude,
that's how *they hope
to make money*. later on. once "we" and "contact us" have got enough of
this fuzzy baseless
trust made by appealing
to a certain naivity of
the intelligent.
funkenstein_: mircea_popescu,
their (arxiv / vixra) competion is journals
that charge
the author per page
ascii_field: mircea_popescu: in
the sense of accurate and resolution-independent mapping
to flattened dead
tree
mircea_popescu: then people wonder about kaminsky. EXACT SAME
THING. you lot have been so mentally stunted by
the stupid welfare state, you do
this sort of shit. "oh, i wonder if arxiv backwards . org paid funkenstein_ anything". no dude,
they didn't, he's just silly like
that. like everyone in
that country.
ascii_field: it would help if
there were actually a proper replacement for it.
mircea_popescu: fuck
them. you want me
to
tihnk "i wonder if you have more"
mircea_popescu: you don't want me
to
think "hey
that was a great piece, i wonder IF ARXIV ORG HAS MORE"
mircea_popescu: own your content, AND EVERYTHING ABOUT IT. you don't need someone else deciding anything whatsoever. not what url
to use. not when
to "put up a warning page". not. ANYTHING.
mircea_popescu: the difference between
this stupid diqus shit and just writing for gawker is 0.
mircea_popescu: and im not reading a pdf. what
the fuck funkenstein_ .
mircea_popescu: <mats> conned or what? lent his name for a buck
to some
trash? << intelligent people have a lot of
trouble avoiding locklin's mistake.
mircea_popescu: <mats> such incompetence. how hard is it
to use sed? << real muppets use excel.
mircea_popescu: bitstein
the observer makes
the valid code replace point, but STILL names its article "the race
to replace bitcoin". as fucking if. "the race
to waste imbecile capitalist money,
the new hot valley
term for what was once known as a venture capitalist" was radioactive or something.
assbot: viXra.org e-Print archive, viXra:1504.0072, “The Majority is Enough” a Rebuttal of
Two Proposed Vulnerabilities of Bitcoin Mining ... (
http://bit.ly/1aqXbRI )
mircea_popescu: you're like
the guy going "so mp you're saying
the perpetuum mobile could not exist because X ?" sure, x if you want it.
mircea_popescu: i'm never going
to be able
to show you ALL
the
things
that are wrong wqith it, being
that it's nonsense.
mircea_popescu: model
this attack and i'll show you one
thin
that's wrong with it enough
to sink it.
[]bot: Bet placed: 2.166 BTC for No on "Bitcoin
to drop under $150 before July"
http://bitbet.us/bet/1133/ Odds: 14(Y):86(N) by coin, 15(Y):85(N) by weight.
Total bet: 15.79172615 BTC. Current weight: 86,260.
[]bot: Bet placed: 3.5 BTC for No on "Gold
to drop under $1000 before August 2015"
http://bitbet.us/bet/1131/ Odds: 11(Y):89(N) by coin, 13(Y):87(N) by weight.
Total bet: 9.21237 BTC. Current weight: 82,570.
funkenstein_: did kaminsky do anything other
than allow his name listed as
technical advisor?
mats: conned or what? lent his name for a buck
to some
trash?
mats: i'm confused as
to Kaminsky's role in all
this
bitstein: gave a long answer describing
the fork and giving lip service
to
this new consensus mechanism in
the works, but failed
to address
the second part. So before
they moved on, I followed up, “But right now, Stellar is a centralized system?” He begrudgingly responded, “It runs on one node, yes.” After my question
they all went right on back
to
talking about how awful centralization is and how great
the decentralized future is.
bitstein: At SXSW, I snuck into a Bitcoin 2.0 panel Jed McCaleb was on. During
the Q&A I got up and asked, “Last December
the Stellar blog had a post called ‘Safety, liveness and fault
tolerance—the consensus choices’
that described how
there was a fork in
the ledger and subsequently, Stellar had essentially collapsed into a centralized system. Can you elaborate on what happened and why people should
trust Stellar in its aftermath?” He
Adlai: concern
trolling in a fancy
typeface
Adlai: painting
this picture of a mining commons on
the brink of
tragedy
Adlai: according
to
the papers' model:
there is no incentive
to be
the first attacker, but
there is incentive
to be
the Nth, N+1st, etc
lobbes: 'but evidently its search-and-replace missed many instances where
the code still says 'ripple.' < l0l
mats: such incompetence. how hard is it
to use sed?
bitstein: Still cracks me up: "There are at least 177 instances in which ripple and stellar’s code matched up. In its haste
to get its currency established, Stellar simply copied ripple’s open-source code but evidently its search-and-replace missed many instances where
the code still says 'ripple.'"
http://observer.com/2015/02/the-race-to-replace-bitcoin/ bitstein: "Is
the new system live? Not yet."
lobbes: so
the 'attacker' is disincentivized due
to
this rising cost, and
the fact
that 'statistically' he is being outed by not finding his shares at
the expected rate (thus, reputation suffers)
Adlai: this is expensive for
the pool operator because he needs
to send redundant work units
to various miners,
to get
them checking eachother
lobbes: hmm, okay
that makes sense
Adlai: with "the defense"
taking some form such as a pool operator negrating a miner who has an unreasonable dropoff of share-finding luck just below
the difficulty
threshhold
Adlai: if miners and pool operators choose
to conduct such attacks against eachother (note
that
the attacks are against eachother, not against "bitcoin"),
they can be fought...
the attack becomes more expensive, and mining becomes more expensive, due
to
the defense
lobbes: but
to Adlai's point, could
there not be one derp out
there who 'withholds' enough until
that 'one day'
to make it worth it (in his eyes)?
lobbes: re: withholding blocks. So basically, mircea_popescu, you are saying
that anything gained from withholding said blocks would be outweighed by
the potential destruction of one's reputation (which is expensive
to build)?
mircea_popescu: mats well, it's
training, i gather.
trainee pieces aren't either useful or valuable, per se. so you're not really "buying" anything.
mircea_popescu: "In order
to encourage participation we would like
to be able
to award more substantial prizes. If you would like
to donate
to
this cause, please contact us!"
mats: heh.
tacit admission
that you can't buy
this kinda
time either way, imo.
mircea_popescu: dude with
the article must have had some pretty odd searches in
the past. my "google images" for princess leia consists of 20 hits of
the woman wearing some sort of white overalls, with
the 17th being her in actual atire.
mircea_popescu: fwiw, i had no idea leia is supposed
to be anything BUT
the slave chick.
mircea_popescu: no it's not fucking anonymous. it's pseudonymous. different fucking
things. identity still exists!
mircea_popescu: the rub, of course, being
the entire "oh but bitcoin is anonymous"
mircea_popescu: which is what
this is : inexistent problems
that don't exist being misunderstood as existing by people who don't understand how anything works, and
then failing
to see
that even should
they exist,
the solution also exists.
assbot: Logged on 09-04-2015 12:34:19; Adlai:
the only way
to _verify_
that your miners haven't withheld blocks, rather
than
trusting
their word for it, is
to check
those nonces yourself
Adlai: as you like
to say - mkay.
Adlai: a when
the same merkle root shows up again?
mircea_popescu: so i give you nonces 1-100
to hash , and fluffy nonces 101-200. and you report nothing found, and he reports nothing found. and
then one day a block on nonce 76 is found and you are forever fucked.
mircea_popescu: because identity being expensive, it's no longer productive
to attack.
Adlai: the only way
to _verify_
that your miners haven't withheld blocks, rather
than
trusting
their word for it, is
to check
those nonces yourself
☟︎ Adlai: the really funny bit from
the (second) paper is "This would push miners
to join private pools which can verify
that
their registered miners do not withhold blocks"
mircea_popescu: and
that's my point here :
these are discussions of no substance used by reddit dorks
to be discussing
things.
mircea_popescu: fluffypony yes fluffs, in a game
theoretic universe where cows are spherical and it rains glass, one could accidentally make a cow snowglobe.
fluffypony: also, as I understand it, we're
talking about a future
time when running a pool *is* more profitable
than it is now
mircea_popescu: Adlai
the "withholding"
thing is idiocy of prime order. it existe,d historically (ie, cca 2011) and was a problem for early pools. but it was SHARE withholding, not block withholding. it was fixed meanwhile.
Adlai: why is ddos an issue here?
this attack is not a ddos
fluffypony: so all
they have
to do is inconvenience
the pool and miners will failover
fluffypony: miners configure
their mining setup with failover pools, but it failsover based on latency as well as a pool being unreachable
mircea_popescu: do you have any idea how long it'd
take a pool (just
that, nothing else)
to make back 100 dollars after server costs ?
fluffypony: DDoS isn't a solved problem for Stratum, and especially not for mining where latency is of
the utmost concern
mircea_popescu: the cost of 1) is what, 100k ? 500k ?
the cost of 2) is what, 100k ? 500k ?
the benefit from
the gained market share is... what ?
mircea_popescu: as a result of 1 and 2, 50% of people mining on B move, leaving it with 11% hash rate.
the resulting 24% difference moves
to pools, 90% of which are A's.