tmsr - Search: pruning

mircea_popescu: if you actually, inconceivably but actually, start pruning, can just as well do it as "will keep last n searches that weren

mp_en_viaje: guish a pruning kit from a mohel's tools.

mp_en_viaje: http://btcbase.org/log/2019-04-17#1908807 << guaranteedly no thought went into any of this. the http://btcbase.org/log/2016-05-16#1467290 proprietor of a previously transylvanian saxon-owned property including an acre of orchard that wasn't properly pruned failing to properly prune the damned trees isn't proof positive "nobody could turn tree pruning into +ev activty". the fucktarded gypo doesn't even know what the fuck an orchard is, nor could distin ☝︎☝︎

asciilifeform: 'pruning of middle' happens, in re konsoomer tech, but consider also that moar goes into garage, or even iraqi roadside mind,

PeterL: so I have been looking at v. (specifically asciilifeform's python version) it seems to me that there should be a "pruning" step after the flow is laid out to remove any patches that are not ancestors of the desired leaf, so that you would not have to manually remove patches.

ben_vulpes: well now i have to look into this and figure out what drives the memory of getting stuck. very foggy memories suggest it was the pruning of spent outputs i was unsure of.

asciilifeform: mircea_popescu: we had this thread. if 'pruning' is a thing -- monetary mass is no longer demonstrably capped.

asciilifeform: a pressed-blockchain aluminum disk set would help to cool the enthusiasm of idiots pushing for 'pruning', 'tx replacement', whichever lunacies crafted for 'let's make'em forget history'

asciilifeform: tx replacement , and the mere contemplation of 'pruning', are a carefully crafted enemy rot weapon.

davout: i think pruning would compare better to cutting everyone's throat "just a little" bit than "cutting one's own throat"

davout: so yeah, i'm satisfied with the answers in the sense that i did not miss any particular argument against pruning

davout: mircea_popescu: not really because i'm not particularly interested in discussing pruning from a position where i'm somehow supposed to defend it

asciilifeform: proponents of 'pruning' want to replace 'the weight of 400k blocks' with a promisetronic 'checkpoint list'. which, i imagine the game plan is, to eventually include deviations from the usual proof of work, per gavin's unabashed declaration where 'WE say what the blocks were'

mircea_popescu: blockchain replacement in a "pruned" scheme costs... whatever your pruning interval is.

davout: asciilifeform: yeah, sure you're not serving peers the historical data, that's the single thing that seems very wrong with pruning, at least to me

asciilifeform: 'where did this money come from, how much exists in total' 'dunno, in the Great Pruning of 2050 the scrolls of the ancients were lost'

asciilifeform: 'The list of all used transactions isn't readily available, and once pruning shows up, it might not even exist at all. So, it only makes sense to compare the new coinbase to the list of transaction hashes that are unspent at the time ... ' << holy fuck the 1) idiocy 2) nobody challenged it, afaik

asciilifeform: 'Fully-spent transactions are allowed to be duplicated in order not to hinder pruning at some point in the future. Not allowing any transaction to be duplicated would require evidence to be kept for each transaction ever made.'

danielpbarron: is it possible to do a lossless pruning on this lubby thing? say some chunk of data kept in an early blk.dat can be removed and the whole thing still verifies because the missing piece can be re-produced from data in more recent blk.dat ?

mircea_popescu: there's a reason i said 2 in 2 out. anything else is either nonmoving (1-2) or else tree pruning (2:1 ends up with a single coinbase ; 1:2 ends up with an infinity of satoshi sized coinbases you might as well stop lying about and issue outright)

mircea_popescu: because, for any given genesis block, the possible patches are universally available ; and the only way to obtain variant builds is through pruning the sigs.

brg444: Next, future Bitcoin "optimizations": Removal of transaction data from blocks - Blockchain pruning UTXO set commitments - Invertible Bloom Lookup Tables (IBLTs) - Moore’s Law - Kryder’s Law - Nielsen’s Law - Yet-to-be-discovered improvements"

pete_dushenski: "Block pruning works during initial sync in the same way as during steady state, by deleting block files “as you go” whenever disk space is allocated. Thus, if the user specifies 550MB, once that level is reached the program will begin deleting the oldest block and undo files, while continuing to download the blockchain."

pete_dushenski: "The minimum was chosen so that Bitcoin Core will be able to maintain at least 288 blocks on disk (two days worth of blocks at 10 minutes per block). In rare instances it is possible that the amount of space used will exceed the pruning target in order to keep the required last 288 blocks on disk." << RARE !

pete_dushenski: highlight : "Block pruning allows Bitcoin Core to delete the raw block and undo data once it’s been validated and used to build the databases. "

ascii_field: pruning the resulting include-fixed directory. Thus, the only people who have to deal with fixincludes are people who build GCC from the source packages, or who are setting up build scripts for their own deployment/distribution.'

mircea_popescu: 'pruning' offers a guarantee that the node won't be able to offer you a block which contains all spent outputs.

williamdunne: BingoBoingo: Pruning nodes can still validate transactions with recent inputs can they not? Or do they store all inputs. Need to read how it works

BingoBoingo: I really fail to see much difference between the pruning nodes and the pseudo-nodes

assbot: /orionwl just merged pruning support in /hashtag/Bitcoin?src=hash Core! Run a (no wallet) full node with 1.3 GB storage. Thanks to all who contributed.

assbot: Pepco angers Montgomery County residents by pruning beloved cherry trees - The Washington Post ... ( http://bit.ly/1xR9Qss )

decimation: http://www.washingtonpost.com/local/pepco-angers-montgomery-county-neighborhood-residents-by-pruning-beloved-cherry-trees/2012/07/18/gJQAqPqMvW_story.html < "As the men in hard hats hacked away, Julie Marcis and her husband confronted the crew, pleading with them to stop ruining the trees.“You feel like your insides are crumbling when you look at what they did,” Marcis said. “You have no control, you can’t do anything, short

ben_vulpes: anyways asciilifeform how does the backported orphan-pruning situation handle the full tree?

mod6: jurov looks at pruning code in utter disbelief... was thinking that everyone konws the cp --parents trick to copy files named in manifest << i /knew/ there had to be an easier way. i seem to remember asking, but anyway.

jurov looks at pruning code in utter disbelief... was thinking that everyone konws the cp --parents trick to copy files named in manifest

Adlai has started to think of pruning in a different manner - rather than reclaiming hard disk space, 'pruning' just means you can provably flush certain transactions from various caches. it's a time/space optimization, not a change of the glob's outward behavior

BingoBoingo: Yes, shouldn't indeed means shouldn't. https://github.com/spesmilo/electrum-server is the address history beast I was thinking of. Note that can become rapidly unfeasible on Gavincoin as does not take kindly to any pruning.

gavinandresen: asciilifeform: again, go read the technical road map post, section on pruning the chain

asciilifeform: pruning << the shitgnomes don't dare a hardfork. so they'll happily try a 'lossy softfork' - see how much functionality can be degraded (in this case, how much of the blockchain can be alzheimered) without hardforking.

mats_cd03: mircea_popescu: i haven't been using the west key system to search. instead, i've been filtering for proceedings involving USG and then pruning for terms like wiretap, warrant, email, hard drive, paperwork, and suppression of evidence.

BingoBoingo: Implement “UTXO commitments” blockchain pruning and “fully validating but not full-history” nodes << Killing this is basically the goal of Fall 2014-2015

ThickAsThieves: Implement “UTXO commitments” blockchain pruning and “fully validating but not full-history” nodes

mircea_popescu: well, it's definitely the blockchain-friendliest approach, seeing as only a hash is stored in the first place. moreover, i don't see how nonspendable tx helps anything other than pruning

mircea_popescu: it's a middle pruning control strategy, the exact equiv of raising taxes "for social good"