382200+ entries in 0.091s

mircea_popescu: kakobrekla i had asked 3 different people so far, which all promised they're going to do it.
mircea_popescu: i could share that for the low low price of 4506982753094509 btc
mircea_popescu: next time someone asks me how i know all this shit ima be all like "dude... my raids crashed".
mircea_popescu: dude with all these crashed raids, i now have the bitcoin/usd rate from unixtime 45788995934578218
mircea_popescu: There is a cult of ignorance in the United States, and there has always been. The strain of anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that my ignorance is just as good as your knowledge.
mircea_popescu: perhaps a surprise in degree, but certainly not in kind
mircea_popescu: and they had a pretty good solution for the byzantine generals, rtoo.
mircea_popescu: VanCleef you're not even kidding, i get a lot of (mostly, older, mostly white, mostly male, republican types) very indignant at how they knoiw btc to be a scam
mircea_popescu: the fact that it gives better answers to several classes of problems
mircea_popescu: but in more advanced applications, there's little way to distinguish between "tech" and "the rules of the world"
mircea_popescu: in a sense, the tech creates the ontological limits of the discourse. this is not visible as much in "classical" applications, because somehow it's obvious that your particular abilities to deal with weak nuclear force are involved.
mircea_popescu: this is why the qe does not improve the position of the us.
mircea_popescu: for that matter, demand does not create value. value creates demand.
mircea_popescu: yes, it does. being happy always boils down to 1. identify what to do to be happy and 2. doing that.
mircea_popescu: as the implementation becomes less centralised (for instance through the workings of democracy, and lawmakers pushing agendas etc),
mircea_popescu: in a centralised system, the assumption is that the implementation is friendly to the design, and the cop will do what the fuck the law says he should do.
mircea_popescu: what makes the actual cop at your door (implementation) follow this principle (design) ?
mircea_popescu: anyway, a very simple example to understand what i mean : suppose the law says nobody may search your house without a warrant.
mircea_popescu: this distinction doesn't work in bitcoin as you'd expect.
mircea_popescu: the particular way bitcoin forces this is through the blockchain.
mircea_popescu: if you take the latter, you can't so asume. the implementation has to be somehow forced to stick to the design
mircea_popescu: if you take the former, you can asume the implementation is friendly.
mircea_popescu: looky here. you decide to make a system. next step, is to decide whether it shall be centralised or decentralised.
mircea_popescu: those things are by and large irrelevant in this discussion.
mircea_popescu: the 51% attack is a weakness of the particular solution used to implement decentralisation.
mircea_popescu: the system is not based on "trust/demand/decentralization". the system is based on the blockchain because it has to use some solution for the byzantyne problem once it's decentralised.
mircea_popescu: i think you're jumping to conclusions on very little data, and the shape of those conclusions is visibly more influenced by your own ideas of the world than the actual reality of the matter
mircea_popescu: i mean "BTC can't survive without being decentralized enough". what's enough and how do you know ?
mircea_popescu: so now i will get a pretty expensive lab bill saying "power fluctuations caused bla bla bla"
mircea_popescu: well, see, before the 2nd drive went in and got fried, i had already sent the 1st to be cut up
mircea_popescu: but for all the shortcomings of distributed models, which exist, the centralised model has its problems.
mircea_popescu: blameless or blameful, you know ? what can i do, i'm doing it.
mircea_popescu: yes, 99% of the time 99% of your shit isn't at work / needed.
mircea_popescu: anwyay, mpex is very much an adaptive beast. it's the best i can do, honestly, because i don't know the correct way to build bitcoin systems
mircea_popescu: however, if you do noboduy cares you were < .1 seconds forever, cause well... days offline.
mircea_popescu: depends, right ? if you never have catastrophic failures of the sort, the incremental gain from the better server is everyone's benefit.
mircea_popescu: so would you have a faster & better main server, or a shittier main so the back-up is there ?
mircea_popescu: jurov yes, but then at some point you have machines in disagreement. how do you resolve this ?
mircea_popescu: at some point the dilemma is, "either people will complain becauyse it takes too long, or people will complain because x edge case wasn't covered in the security design"
mircea_popescu: bob_ yes, in principle. see, here's the thing : the more complex the security model becomes, by that very fact the more complex recovery processes become
mircea_popescu: in my case, i have to, everytime something unexpected happens, see why the fuck it happened before i can move on
mircea_popescu: however, if the result of errors is not nil, you can't just shrug and move on
mircea_popescu: you CAN seamlessly failover on errors, if the result of errors is nil and convenience of not having errors is what counts.
mircea_popescu: kakobrekla the problem is centralisation + security as a combo.
mircea_popescu: you know genial = friendly, happy people. genius is prolly what you mean
mircea_popescu: jurov because at some point your system is either cenrtralised or it isn't.
mircea_popescu: anyway, no attack, no data loss, no anything. just fucking obnoxious power.
mircea_popescu: that particular power strip apparently puts noise in burns drives. i have no spare powers because a lot of shit in there.
mircea_popescu: see, a drive popped yest. so this was suspicious, was looked into extensively. nothing found. new drive was put in.