14:23 UTC

< May 2015 > Su Mo Tu We Th Fr Sa 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

- Console
- #amber
- #apicula
- #arm-graphics
- #arm-netbook
- #bitcoin-wizards
- #buildbot
- #bundler
- #cinch
- #coiniumserv
- #coiniumserv-dev
- #crystal-lang
- #cubieboard
- #datamapper
- #discferret
- #elliottcable
- #forth
- #glasgow
- #gridcoin
- #gridcoin-dev
- #homecmos
- #huawei-g300
- #imx6-dev
- #imx6-dongle
- #ipfs
- #jruby
- #libreelec
- #libreoffice-ru
- #lima
- #linux-amlogic
- #linux-exynos
- #linux-rockchip
- #linux-sunxi
- #lisp
- #litex
- #logarion
- #lowempire
- #maemo-leste
- #maglev-ruby
- #microrb
- #milkymist
- #mirage
- ##moved_to_libera
- #mutant
- #nanoc
- #neo900
- #nextbsd
- #nmigen
- #ocaml
- #opal
- ##openfpga
- #openwrt-devel
- #panfrost
- ##panfrost-offtopic
- #Paws
- #Paws.Nucleus
- #picolisp
- #ponylang
- #prjmistral
- #pypy
- #qaul.net
- #qi-hardware
- #racket
- #radxa
- #reasonml
- #river
- #rom-rb
- #rubinius
- #ruby
- #ruby-core
- #rubygems
- #rubygems-aws
- #rubygems-trust
- #ruby-lang
- #ruby-rdf
- #sandstorm
- #scopehal
- #skywater-pdk
- #slime
- #soletta
- #stellar
- #stellar-dev
- ##stm32-rs
- #symbiflow
- #systemtap
- #teamhacksung
- #teamhacksung-support
- #tinyqma
- #videocore
- #wallaroo
- #xiki
- #xtompp
- ##yamahasynths
- #yosys
- #zig

<frankenmint>
Their article said 'software' implementation that can work on existing chip infrastructure

<frankenmint>
yea I know, say sorry bout giving you a hard time about my gpu frame comming 5 months later :p

<frankenmint>
fluffypony: know anyone stateside who wants a vitalia standalone here stateside? I've got a brand new one for them

<fluffypony>
ok I'm out, speaking at a meetup in Brussels in 3 hours and presentation still needs tweaks

<Taek>
Would it be an acceptable endgame to have mining become completely inaccessable to the unwealthy, but instead dominated by 5-10 corporations, none of whom have more than 33% hashrate?

<kanzure>
no, because corporations are subject to the law and can be compelled by courts to compromise technical systems

<Taek>
certainly it's less ideal than 1-cpu-1-vote, but in a stable political climate it might be sufficient

<temujin>
bitcoin is perfectly parallel, meaning the hashrate landscape can change quickly as individual miners change pools (in the case of a disagreement with pool policy, overcentralization, etc)

<Taek>
A bigger mining investment also has more room to establish itself in places with inherent advantages (such as regions where electricty is cheap), has more ability to collect lots of talent (proprietary and significant ASIC improvements)

<Taek>
hopefully. Currently the most promising seems to be the idea of using the waste heat of ASICs as an advantage

<temujin>
haha, my two GPU rigs certainly heated my apartment quite nicely winter 2013... i can see this being true

<nsh>
all we need, and it can't be that much of an ask, is for the US DoD to reprogram their GPS satellites to act as entropy beacons so that it's possible to cryptographically attest geographic locations

<nsh>
well, you can't, because space doesn't exist. but you can attest to being in a certain locus within a certain timeframe

<nsh>
(what i mean is, you can fake being somewhere else, but you can't fake having the exact information for that position at some exact time)

<nsh>
presumably china will at some stage, so at least you can split the trust between rough geopolitical adversaries

<yoleaux>
"GLONASS (Russian: ГЛОНАСС, IPA: [ɡlɐˈnas]; Глобальная навигационная спутниковая система; transliteration Globalnaya navigatsionnaya sputnikovaya sistema), or "GLObal NAvigation Satellite System", is a space-based satellite navigation system operated by the Russian Aerospace Defence Forces." — http://en.wikipedia.org/wiki/GLONASS

<yoleaux>
"The BeiDou Navigation Satellite System (BDS, simplified Chinese: 北斗卫星导航系统; traditional Chinese: 北斗衛星導航系統; pinyin: Běidǒu wèixīng dǎoháng xìtǒng) is a Chinese satellite navigation system." — http://en.wikipedia.org/wiki/BeiDou_Navigation_Satellite_System

<nsh>
but aye, back in the real world, the military of nation-states does not have a great track record in positively responding to my suggestions

<Taek>
in all liklihood, the consumers would never touch their miner, the company they bought it from would manage all of the operations

<nsh>
certainly finding ways to convert sha256 entropy from a problem you have to deal with to an asset is likely to increase asic decentralization

<nsh>
specifically: "How to find smooth parts of integers." http://cr.yp.to/factorization/smoothparts-20040510.pdf

<yoleaux>
I saw gmaxwell 15 May 2015 19:13Z in #bitcoin-wizards: <gmaxwell> So yes, an unspecified different design may have totally different vulnerabilities.

<nsh>
who's going to explain smoothness and elliptic curves and how the latter helps find the former in integers despite having no intrinsic notion of prime divisor itself

<yoleaux>
I saw sipa 14 May 2015 16:17Z in #bitcoin-wizards: <sipa> i'll be on vacation the next two weeks

<nsh>
i guess i'll just mash my brain into wikipedia until i find new ways of crying blood through my eyes

<gmaxwell>
nsh: sorry, I've been super busy (people will be excited...) and the conversation in here was ... uh not at its best lately.

<nsh>
well, first i wanted to know how djb's fast batch-GCD algorithm works, but in order to understand that, i need to know generally how elliptic curves, which don't even have a notion of prime divisors, can facilitate the finding of small prime factors of integers

<yoleaux>
"The Lenstra elliptic curve factorization or the elliptic curve factorization method (ECM) is a fast, sub-exponential running time algorithm for integer factorization which employs elliptic curves. For general purpose factoring, ECM is the third-fastest known factoring method." — http://en.wikipedia.org/wiki/Lenstra_elliptic_curve_factorization

<nsh>
(Silverman I is available here, if anyone else is [masochistic and] interested: http://www.pdmi.ras.ru/~lowdimma/BSD/Silverman-Arithmetic_of_EC.pdf )

DrWatto has quit [Quit: Actually, she wasn't really my girlfriend, she just lived next door and never closed her curtains.]

<nsh>
i want to be hand-holded through difficult number theory because i am a lowly telephone sanitation engineer and they keep kicking me out of maths classes when i sneak into university :)

<andytoshi>
basically you do ECC operations as though you were in a prime field until they break, then you've detected non-primeness in a way that gives you a factor

<nsh>
the immediate integers to a large prime are likely to have asymptotically equivalent smoothness

<andytoshi>
there are actually two methods in the paper, the p-1 and p+1 ... they are quite similar, but there are no asymptotics since this is all discrete

<yoleaux>
"In computational number theory, Williams' p + 1 algorithm is an integer factorization algorithm, one of the family of algebraic-group factorisation algorithms. It was invented by Hugh C. Williams in 1982." — http://en.wikipedia.org/wiki/Williams'_p_%2B_1_algorithm

<yoleaux>
"Pollard's p − 1 algorithm is a number theoretic integer factorization algorithm, invented by John Pollard in 1974. It is a special-purpose algorithm, meaning that it is only suitable for integers with specific types of factors; it is the simplest example of an algebraic-group factorisation algorithm." — http://en.wikipedia.org/wiki/Pollard%27s_p_%E2%88%92_1_algorithm