200300+ entries in 0.069s
mircea_popescu: you know i have trouble picturing you telling some random woman to fuck off ?
mircea_popescu: lol. it is kinda the retard path. prolly come up with some fixes first tho.
mircea_popescu: nah, riccardo is a suspicious guy sending paypal donations in clear violation of maritime law.
mircea_popescu: "specializes" seems to have lost a meaning sometime in 2004.
mircea_popescu: "Sonya is also Founder and Director of two Hats a design and development company that specializes in corporate branding, print design, web design, web development, mobile/responsive web development, e-commerce services, business analysis, search engine optimization, social media, Google Analytics, Google Ads and E-Marketing."
mircea_popescu: so tell her "listen, i want to hear the speaker to i can mock him as well."
mircea_popescu: "Sonya is an entrepreneur and founder of Bitcoin Payments."
mircea_popescu: of course this would also trash the fixed size blocks-powered bitcoinfs
mircea_popescu: anyway, if anyone got a github account, leave a link to this convo on there, curious if sipa/anyone groks wtf's going on.
mircea_popescu: and back in the day you'd get flunked in undergrad sys design for shit like this.
mircea_popescu: but anyway - the fucking magic number should have been in block headers not in a c file.
mircea_popescu: a noob saved a word ("who would need 1000000 repeated each block!!!") and in the process made the design much dumber.
mircea_popescu: only sane way to have a data struct is if it begins with "hello, this is a struct, will take no more than x of your space kthx"
☟︎ mircea_popescu: and when you allocate memory, same ? you just count the bytes you used ? AFTER YOU DID ?
mircea_popescu: who the fuck looks at a list and goes "a it's ok, don't need header referencing the size, why would you."
mircea_popescu: (why the blocks were designed to not contain a summary bytecount of their own variable content i nthe first place is yet another example of the "satoshi was not much of a designer" thing)
mircea_popescu: but anyway, yes, if we're to keep state the only sane way is to put it in the block headers.
mircea_popescu: irrespective of what "the very talented" "core developers" think, this will not be a simple or easy change. it will require a lot of touching,
mircea_popescu: punkman you'll never get consensus behind "random number".
mircea_popescu: it'd have to look a lot more like the above than like anything i've seen,
☟︎ mircea_popescu: but, if a solution to this problem, admitting it is a problem, and admitting a solution can be had, is to be had,
mircea_popescu: but it horribru or not horribru - a) we are currently keeping state, just, as a constant and b) there is no way to make pie-in-the-sky arbitrary values work.
mircea_popescu: you can't allow it to reduce block size, because the noise will make the system unbound and it'll collapse.
mircea_popescu: that's the biggest problem with it, it introduces a state.
mircea_popescu: (in case it wasn't obvious - this wouldn't allow a decrease of blocksize. increase only)
mircea_popescu: (ie, 1mb exactly. because 990 mb blocks fails to allow for an increase by the above proposal)
mircea_popescu: there has not yet been as much as a signle day of straight full blocks.
mircea_popescu: if the fee market has not developed by the time the blocks are filled like that, it;s not happening.
mircea_popescu: it necessarily would. because unless two weeks worth of filled blocks, it does nothing.
mircea_popescu: 20 empty blocks in the two weeks interval nullify the growth.
mircea_popescu: maintaining full blocks for two weeks straight has not happened yet nor would it be trivial
mircea_popescu: that's the difference : i have no need for an arbitrary 18% no matter what.
mircea_popescu: actual bitcoin users can actually force the miners, by paying enough to make their transactions not-non-mineable, even if it fills blocks
mircea_popescu: gives everyone some leverage - miners that want to keep blocks smaller can mine empty blocks. miners that mine large blocks have the usual disadvantage due to speed. the thing itself allows very limited growth, in the sense of 1% growth if the previous two weeks were all full blocks.
mircea_popescu: a function that, after each retarget, allows blocks as large as say 1.01x the average actual size of the blocks in the 2k batch for instance.
☟︎ mircea_popescu: "Attempting to buy time with a fast increase is not wanting to face that reality, and treating the system as something whose scale trumps all other concerns. A long term scalability plan should aim on decreasing the need for trust required in off-chain systems, rather than increasing the need for trust in Bitcoin." etc
mircea_popescu: mostly for "Bitcoin's advantage over other systems does not lie in scalability. Well-designed centralized systems can trivially compete with Bitcoin's on-chain transactions in terms of cost, speed, reliability, convenience, and scale. Its power lies in transparency, lack of need for trust in network peers, miners, and those who influence or control the system. Wanting to increase the scale of the system is in conflict
mircea_popescu: certainly more than reddit modertatorship's paying hjim
mircea_popescu: weird that he didn't choose to take that stand three years ago on his own damned forum, but instead goes to break his neck on someone else's shit platform.