sipa changed the topic of #bitcoin-wizards to: This channel is for discussing theoretical ideas with regard to cryptocurrencies, not about short-term Bitcoin development | http://bitcoin.ninja/ | This channel is logged. | For logs and more information, visit http://bitcoin.ninja
Ylbam has quit [Quit: Connection closed for inactivity]
thrmo has quit [Read error: Connection reset by peer]
goksinen has joined #bitcoin-wizards
belcher has quit [Quit: Leaving]
CubicEarth has joined #bitcoin-wizards
<andytoshi>
gmaxwell: i haven't heard anything about digishield and MW except for fluffypony's comment just now
<andytoshi>
(nor had i heard anything about digishield period :))
oleganza has quit [Quit: oleganza]
jouke has quit [Ping timeout: 258 seconds]
<gmaxwell>
minor idea, with pay-to-contract transfers you can have outputs that commit to some properties (like ownership) in an auxiliary system. There is no reason these transfers couldn't be joint. E.g. two people write one transaction that creates one output belonging to the aux system, with their payment commiting to a tree of outputs that they desire in the aux system. This might make it more econom
<gmaxwell>
ical to move dust into a system specialized for microtransactions.
jouke has joined #bitcoin-wizards
jouke has joined #bitcoin-wizards
jouke has quit [Changing host]
NewLiberty has quit [Ping timeout: 258 seconds]
PaulCape_ has quit [Ping timeout: 258 seconds]
Nightwolf has quit [Read error: Connection reset by peer]
Nightwolf has joined #bitcoin-wizards
cluckj has joined #bitcoin-wizards
mountaingoat has quit [Ping timeout: 260 seconds]
wasi has quit [Remote host closed the connection]
wasi has joined #bitcoin-wizards
mountaingoat has joined #bitcoin-wizards
pro has quit [Quit: Leaving]
mountaingoat has quit [Ping timeout: 240 seconds]
wangchun has quit [Quit: leaving]
Dyaheon has quit [Ping timeout: 260 seconds]
jtimon has quit [Ping timeout: 240 seconds]
Giszmo has quit [Quit: Leaving.]
wangchun has joined #bitcoin-wizards
Dyaheon has joined #bitcoin-wizards
Noldorin has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
legogris has quit [Remote host closed the connection]
legogris has joined #bitcoin-wizards
mn3monic has quit [Ping timeout: 255 seconds]
oleganza has joined #bitcoin-wizards
TheSeven has quit [Ping timeout: 255 seconds]
TheSeven has joined #bitcoin-wizards
mountaingoat has joined #bitcoin-wizards
PaulCapestany has joined #bitcoin-wizards
dodomojo has joined #bitcoin-wizards
talmai has joined #bitcoin-wizards
goksinen has quit [Ping timeout: 255 seconds]
dodomojo has quit [Read error: No route to host]
goksinen has joined #bitcoin-wizards
anon616 has quit [Remote host closed the connection]
goksinen has quit [Ping timeout: 255 seconds]
anon616 has joined #bitcoin-wizards
UnrealLife has joined #bitcoin-wizards
UnrealLife1 has joined #bitcoin-wizards
UnrealLife has quit [Ping timeout: 252 seconds]
oleganza has quit [Quit: oleganza]
talmai has quit [Quit: mining]
RubenSomsen has joined #bitcoin-wizards
luke-jr has quit [Excess Flood]
luke-jr has joined #bitcoin-wizards
anon616 has quit [Remote host closed the connection]
goksinen has joined #bitcoin-wizards
anon616 has joined #bitcoin-wizards
Aranjedeath has quit [Quit: Three sheets to the wind]
goksinen has quit [Remote host closed the connection]
harrymm has joined #bitcoin-wizards
Ylbam has joined #bitcoin-wizards
skeuomorf has joined #bitcoin-wizards
BashCo has quit [Ping timeout: 268 seconds]
Dyaheon has quit [Ping timeout: 268 seconds]
Dyaheon has joined #bitcoin-wizards
goksinen has joined #bitcoin-wizards
zaus has joined #bitcoin-wizards
<zaus>
How does Bitcoin overcome the DLS88 impossibility result?
<zaus>
(DLS88 = consensus in the presence of partial synchrony)
goksinen has quit [Ping timeout: 260 seconds]
<gmaxwell>
zaus: bitcoin doesn't achieve consensus, it just becomes less likely to reorganize over time (given certian more or less plausable assumptions)
<gmaxwell>
it's always possible that at any point martians show up with a more work chain which we just never heard about because we haven't had an internet connection to them... and they blow away the bitcoin history.
<gmaxwell>
(at least from a pure analysis of how the system works perspective)
<zaus>
(if martians have that much computational power, they would also break any authenticated byzantine agreement)
<zaus>
thanks gmaxwell
<gmaxwell>
they don't need that much processing power, certantly not enough to break any ordinary digital signature.
<gmaxwell>
they could have just recieved a copy of bitcoin back when it started and been mining it along just slightly faster than all of us; but partitioned.
<zaus>
oh.. I misunderstood you - that makes sense!
<zaus>
gmaxwell: would bitcoin fit in the DLS partially sync network?
<gmaxwell>
zaus: also wrt byzantine consensus-- basically any scheme that assumes even weak synchronicity of distributed elements is not very physically or pratically realistic... since just achieving synchronious operation itself basically requires byzantine agreement, or equipment that doesn't really look much like anything any of us use for computing. :P (e.g. whats the worst case bound on the time for
<gmaxwell>
a message between you and I?)
<gmaxwell>
No, Bitcoin 'works' (does what it does, which is not consensus in the formal sense normally considered in the lit) with very very low assumptions about any kind of synchronicity at all. Which is good because those assumptions don't map really well to actual distributed systems.
<gmaxwell>
(at least not ones that span multiple administrative domains)
<gmaxwell>
e.g. you imagine bitcoin implemented before electronic communications with messages being carried very unreliably by ships at sea ... it would still 'work'-- though you would have to wait an awful long time to be reasonably confident your transactions were settled. :)
<zaus>
ha! that makes sense
<gmaxwell>
you can think of bitcoin as taking all the hard parts like knowing how long a partition might last, and then dumping them on the user (who then promptly ignores them) by never making any promise to not go back an arbritary amount.
<gmaxwell>
in practice, so far, this seems to work out more or less okay. it helps that in a lot of domains Bitcoin's non-cryptographic competition is really slow to achieve any kind of irreversability. (e.g. international wire transfers take days, things like credit card payments aren't functionally irreversable for months)
BashCo has joined #bitcoin-wizards
<zaus>
hey gmaxwell, thanks a lot for the explanation
<zaus>
I think there is an implicit assumption of time in Bitcoin which is basically: on expectation, proof of work takes 10 minutes
BashCo has quit [Ping timeout: 240 seconds]
<gmaxwell>
zaus: kinda but it's very very very vague.
<gmaxwell>
it doesn't "take 10 minutes" -- it has some computational difficulty which is set by the consensus rules, and expressed in the block header. Each block gives its own idea of the time and there is very little constraint on these time values.
<gmaxwell>
every 2016 blocks (two weeks) the difficulty is adjusted to fix the rate based on the times in the blocks.
BashCo has joined #bitcoin-wizards
<gmaxwell>
But the only forced relationship that value has with "actual time" is that each node will (temporarily) reject a block if its claimed time is more than two hours in the future from the nodes own personal ideal of the time.
<gmaxwell>
s/ideal/idea/
<gmaxwell>
Which means that nodes could pretty much set their clocks via sundial and a alamanic and a vague idea of their location.
<zaus>
(would this count as synchronized clocks?)
<zaus>
(weakly)
<gmaxwell>
The purpose of this limit is not really related to the consensus algorithim, but rather to remove an incentive to misbehave on the part of miners.
MoALTz has joined #bitcoin-wizards
<gmaxwell>
e.g. you could eliminate it completely and the system would work fine even the miners clocks were not synchronized at all... so long as the miners were honest enough to not purposefully advance the their clocks forward in order to play out the system's inflation faster.
<gmaxwell>
the interblock interval would not be ten minutes but roughly the hashpower weighed average of the '10 minutes' per the miners clock frequency.
<gmaxwell>
but it doesn't need to be 10 minutes to work-- if the interblock interval starts getting close to the communication delays the system becomes much more inefficient and starts taking (potentially much) longer to converge, but it keeps on working
<gmaxwell>
zaus: some people (e.g. amiller) have proposed ways of controlling the difficulty for systems like this that make no reference to time at all.-- e.g. you can use the existance of forks in the chain to tell you to slow down blocks. (though this kind of design seems like it would result in the greatest geographic concentration of participants ramping up the speed to the point where outsiders cannot
<gmaxwell>
participate equally--- the system becomes unfair when the interval is not insigificant compared to communication delays).
skeuomorf has quit [Ping timeout: 260 seconds]
<gmaxwell>
So I think fundimentally in the bitcoin-like consensus there is no requirement for synchronicity of any form. ... though the pratical system has a very very loose time connection which is really just there to control inflation.
<gmaxwell>
and mostly miners are just trusted to do that right, but the the rules are strong enough that they can't profit greatly from cheating it... but not so strong that an actual block has ever been rejected due to that 2hr rule.
<gmaxwell>
(and does a rule which never executes exist? :P)
* gmaxwell
goodnight
<zaus>
have a good night gmaxwell, I really appreciate your time :)
NewLiberty_ has quit [Read error: Connection reset by peer]
Joseph__ has joined #bitcoin-wizards
NewLiberty has quit [Ping timeout: 240 seconds]
jewsus has joined #bitcoin-wizards
MaxSan has joined #bitcoin-wizards
MaxSan has quit [Ping timeout: 255 seconds]
TheSeven has quit [Ping timeout: 240 seconds]
TheSeven has joined #bitcoin-wizards
UnrealLife has quit [Quit: Leaving]
laurentmt has quit [Ping timeout: 260 seconds]
laurentmt has joined #bitcoin-wizards
laurentmt has quit [Client Quit]
mountaingoat has quit [Ping timeout: 260 seconds]
mountaingoat has joined #bitcoin-wizards
<andytoshi>
lol, oh, i didn't see that. if he mentions it again i'll complain at him
Guyver2 has joined #bitcoin-wizards
<bsm1175321>
Hmmm maybe I should throw my hat in the ring on the topic of retargeting algos. Just about everything ever done on this topic is downright stupid.
<sipa>
I think Bitcoin's retarget algorithm is decent. If you really want to adjust every block, multiply with the 2016th root of the ratio between 2 weeks and the actual time the last 2016 blocks took
<bsm1175321>
Bitcoin's retargeting algorithm only works because hashrate has been monotonically increasing.
<sipa>
?
<sipa>
it has gone down on multiple occasions
<bsm1175321>
Hasn't gone down enough to enable hashrate attacks that altcoins have seen...
<bsm1175321>
And the timescale of adjustments is long enough to be annoying to short attention span attackers.
<sipa>
if your cryptocurrency doesn't have nearly all hashrate available for that PoW function, it is vulnerable regardless
<sipa>
PoW algorithms don't change that
<sipa>
s/PoW/retarget/
<andytoshi>
also, "it's only worked because it's disincentivized attackers" means "it's working" :)
<bsm1175321>
Why not retarget once a year then?
<bsm1175321>
Inserting random time constants and arguing over them is not good design.
<sipa>
because of annoying variance in block rate
<sipa>
in an economically relevant setting, your cryptocurrency has all the PoW it can get, and there won't be sharp changes in hashrate
<andytoshi>
bsm1175321: if the interval is too wide then you're not reacting to hashpower changes quickly enough and your blockrate goes all over the place. if it's too small then you're going to get bad measurements because of the uncertainty principle
<andytoshi>
the "random number" is empirically in a good spot, i'm sure it's not optimal, but i see no pressing reason anyone would tweak it
<sipa>
trying to accomodate for fast change is opening up other attacks
<andytoshi>
if you're complaining about the linear feedback, bitcoin's system is very simple and obviously not going to overshoot (though it might not converge as fast as is theoretically possible)
<sipa>
i've simulated a few of the "fancy" retarget algorithms, and they're almost all easily brought into a regime where they oscillate across multiple orders of magnitude
<bsm1175321>
It's a reality that any PoW function that MW chooses will be initially mined using GPUs, and therefore hashrate slewing and 51% attacks will be possible due to the existence of GPU farms. The reason Bitcoin avoids this fate is that by the time anyone really noticed it, it had already moved to ASICs.
<bsm1175321>
Can Lord Voldemort give us ASICs on day 1? ;-)
<sipa>
s/MW/grin/
<bsm1175321>
In the absence of ASICs, MW absolutely needs a more sophisticated retargeting algo.
<sipa>
s/MW/grin/
<bsm1175321>
Sure :-P
<andytoshi>
thanx sipa
<andytoshi>
bsm1175321: MW is an abstract technology that also includes elements (which has no PoW :)), it has nothing to do with PoW
<bsm1175321>
Is there to be a MW/grin fork?
<sipa>
?
<sipa>
MW is just technology
<bsm1175321>
sipa: you're differentiating... why?
<andytoshi>
bsm1175321: there are already multiple implementations of MW that work in completely different ways
<bsm1175321>
andytoshi: what are the others?
<sipa>
bsm1175321: is there to be a blockchain/bitcoin fork?
<bsm1175321>
sipa: the banks are trying real hard.
<andytoshi>
bsm1175321: elements supports MW in the current codebase (tho there isn't a chain running with it yet afaik, i suppose anyone who wants to stand up a functionary can start one..)
<sipa>
bsm1175321: ok, better example
<sipa>
bsm1175321: is there to be a (merkle tree commiting to transactions)/bitcoin fork?
<bsm1175321>
sipa: I see what you're getting at, no explanation needed
<sipa>
ok!
<andytoshi>
you just create normal CT transactions with empty scriptpubkeys and use a 1-bit rangeproof on a OP_RETURN as the "kernel"
<sipa>
andytoshi: that sounds like just slightly less terrible than counterparty and bitcoin
<sipa>
:p
<andytoshi>
sipa: lol, i specifically added support to elements for this
<bsm1175321>
andytoshi: neat, is it public yet?
<andytoshi>
maybe we can do 0-bit rangeproofs?
<andytoshi>
bsm1175321: yes, i'm checking if it's on the master branch or what..
<andytoshi>
bsm1175321: well, nobody has written a wallet that produces these transactions
<andytoshi>
but the blockchain support for it is public
<sipa>
andytoshi: and no support in the node software to do the fancy pruning
<andytoshi>
sipa: ah, yeah, that too
RubenSomsen has joined #bitcoin-wizards
<sipa>
"details"
<andytoshi>
:)
<andytoshi>
i hope to get some of the wallet stuff done this summer
mountaingoat has quit [Ping timeout: 245 seconds]
pro has joined #bitcoin-wizards
n1ce has joined #bitcoin-wizards
talmai has joined #bitcoin-wizards
mountaingoat has joined #bitcoin-wizards
talmai has quit [Quit: mining]
talmai has joined #bitcoin-wizards
talmai has quit [Quit: mining]
RubenSomsen has quit [Ping timeout: 252 seconds]
Guest12838 has quit [Remote host closed the connection]
Guest12838 has joined #bitcoin-wizards
<luke-jr>
[12:35:19] <andytoshi> bsm1175321: elements supports MW in the current codebase <-- it does? :O
<andytoshi>
luke-jr: not elements alpha, which would unfortunately require a hf to support it. but yes, elements very nearly supported MW-style transactions from day one
<andytoshi>
see above, you can just use scriptless outputs and they'll behave the same as MW outputs. problem was that without scriptSigs it was possible to malleate the outputs to add scripts (e.g. add OP_RETURN and burn somebody's coins)
<andytoshi>
the fix for that was dead simple, have the rangeproofs sign the scriptpubkeys
<andytoshi>
had there not been this problem, i (or somebody) probably would've hammered out a simple wallet to demo this, so that we could do a "lol we supported mimblewimble since before it was invented" blog post. alas :)
JackH has quit [Quit: Leaving]
kmels has joined #bitcoin-wizards
<luke-jr>
XD
<luke-jr>
andytoshi: but surely not the "prune forever, don't even download" style MW?
laurentmt has quit [Quit: laurentmt]
<andytoshi>
luke-jr: well, if you recognize MW transactions, and you're fed the IBD data in the right way (there are "pegs" into and out of MW-style outputs that you have to be aware of) you can do this
<andytoshi>
it would require some fairly serious node support
<luke-jr>
does this mean MW would have no value in an ext.block?
<andytoshi>
well, CT without MW is a huge pain in the ass, and you need CT to do the conversions so transparently
<andytoshi>
s/pain in the ass/resource hog/
<andytoshi>
what i described basically is an ext block, but the ext block contents are mixed in with the main block ones
thrmo has joined #bitcoin-wizards
goksinen has joined #bitcoin-wizards
goksinen has quit [Ping timeout: 268 seconds]
Giszmo has joined #bitcoin-wizards
skeuomorf has quit [Read error: Connection reset by peer]
jtimon has quit [Ping timeout: 240 seconds]
CheckDavid has joined #bitcoin-wizards
talmai has joined #bitcoin-wizards
Guest44006 is now known as Dav2
goksinen has joined #bitcoin-wizards
goksinen has quit [Ping timeout: 260 seconds]
kristofferR has joined #bitcoin-wizards
harrymm has joined #bitcoin-wizards
harrymm has quit [Max SendQ exceeded]
harrymm has joined #bitcoin-wizards
MaxSan has joined #bitcoin-wizards
talmai has quit [Quit: mining]
talmai has joined #bitcoin-wizards
gm2051 has joined #bitcoin-wizards
jcluck has joined #bitcoin-wizards
jcluck has quit [Remote host closed the connection]
Giszmo has quit [Quit: Leaving.]
talmai has quit [Ping timeout: 240 seconds]
abpa has joined #bitcoin-wizards
gm2052 has joined #bitcoin-wizards
gm2051 has quit [Ping timeout: 245 seconds]
Giszmo has joined #bitcoin-wizards
jtimon has joined #bitcoin-wizards
n1ce has quit [Read error: Connection reset by peer]
n1ce has joined #bitcoin-wizards
talmai has joined #bitcoin-wizards
vega4 has joined #bitcoin-wizards
oleganza has joined #bitcoin-wizards
BashCo has quit [Ping timeout: 255 seconds]
Sosumi has joined #bitcoin-wizards
MaxSan has quit [Ping timeout: 260 seconds]
BashCo has joined #bitcoin-wizards
talmai has quit [Quit: mining]
talmai has joined #bitcoin-wizards
thrmo has quit [Quit: Waiting for .007]
gm2053 has joined #bitcoin-wizards
talmai has quit [Ping timeout: 255 seconds]
gm2052 has quit [Ping timeout: 252 seconds]
Guyver2 has quit [Remote host closed the connection]
cluckj has quit [Read error: Connection reset by peer]
goksinen has joined #bitcoin-wizards
mn3monic has joined #bitcoin-wizards
goksinen has quit [Ping timeout: 258 seconds]
MaxSan has joined #bitcoin-wizards
MaxSan has quit [Ping timeout: 245 seconds]
priidu has joined #bitcoin-wizards
hongkonger has joined #bitcoin-wizards
<hongkonger>
Hello
<hongkonger>
Anyone here?
<sipa>
never
<kanzure>
agreed
Joseph__ has quit [Ping timeout: 240 seconds]
<hongkonger>
So I was having a concept on bitcoin sub-units for mass adoption in East Asia
<hongkonger>
0.1 mBTC will be called Victoria, 0.01mBTC will be called Meiji, 0.001mBTC will be called GuangMui
<hongkonger>
Current Price: 1 Victoria is ~1HKD, 1 Meiji ~1.4 JPY, 1 GuangMui ~1.4 KRW
<hongkonger>
By doing this, it is easy for non-tech-savy people to understand the approximate price of Bitcoin
<sipa>
#bitcoin
<hongkonger>
sipa: are you a bot?
<sipa>
no
<arubi>
some say...
<kanzure>
i might be.
<sipa>
i have circumstancial evidence that kanzure is a bot
<bsm117532>
I've met him in person. He's definitely a bot.
<hongkonger>
So is this a good idea? If it passes, East Asia would gain mass adoption
<bsm117532>
hongkonger: the #bitcoin channel is a better place for that topic.
<hongkonger>
got it
dnaleor has quit [Ping timeout: 240 seconds]
kmels has quit [Ping timeout: 260 seconds]
dnaleor has joined #bitcoin-wizards
hongkonger has left #bitcoin-wizards [#bitcoin-wizards]
MaxSan has joined #bitcoin-wizards
Giszmo has quit [Ping timeout: 260 seconds]
belcher has joined #bitcoin-wizards
arubi_ has joined #bitcoin-wizards
arubi has quit [Ping timeout: 248 seconds]
kmels has joined #bitcoin-wizards
Dyaheon has quit [Ping timeout: 255 seconds]
Dyaheon has joined #bitcoin-wizards
jewsus is now known as slimeball
UnrealLife has joined #bitcoin-wizards
Dav2 is now known as Davasny
Davasny has quit [Remote host closed the connection]
d9b4bef9 has quit [Remote host closed the connection]
d9b4bef9 has joined #bitcoin-wizards
Sosumi has quit [Quit: Bye]
CheckDavid has quit [Quit: Connection closed for inactivity]
Aranjedeath has quit [Quit: Three sheets to the wind]
kristofferR has joined #bitcoin-wizards
kmels has quit [Ping timeout: 240 seconds]
dgenr8 has joined #bitcoin-wizards
gigq has quit [Ping timeout: 252 seconds]
gigq has joined #bitcoin-wizards
Giszmo has quit [Quit: Leaving.]
thrmo has joined #bitcoin-wizards
oleganza has quit [Ping timeout: 240 seconds]
Giszmo has joined #bitcoin-wizards
laurentmt has joined #bitcoin-wizards
laurentmt has quit [Client Quit]
Belkaar has quit [Ping timeout: 268 seconds]
Belkaar has joined #bitcoin-wizards
Belkaar has joined #bitcoin-wizards
Belkaar has quit [Changing host]
UnrealLife has quit [Quit: Leaving]
AaronvanW has quit [Remote host closed the connection]
AaronvanW has joined #bitcoin-wizards
oleganza has joined #bitcoin-wizards
jouke has quit [Ping timeout: 260 seconds]
AaronvanW has quit [Ping timeout: 260 seconds]
jouke has joined #bitcoin-wizards
stonecoldpat has joined #bitcoin-wizards
<stonecoldpat>
I've been reading https://github.com/bitcoin/bips/blob/master/bip-0050.mediawiki and looking at the resolution which "kicked unpatched nodes off the network". I'm tempted to call this a hard-fork - i'm wondering if someone can tell me why it might not be?
<stonecoldpat>
(im not trying to troll - i'm looking into past soft forks and this caught my eye)
<gmaxwell>
stonecoldpat: it's a hardfork but BIP50 is relally inaccurate.
<gmaxwell>
stonecoldpat: nodes prior to 0.8 were all inconsistent with themselves, so it was a latent hardfork that existed since day 1.
<gmaxwell>
stonecoldpat: softforks are not forks at all (normally) everything if fully compatible, no alternatives chains are created, etc.
<gmaxwell>
I think yout graph is shit.
<stonecoldpat>
gmaxwell: Yeah I read about the inconsistency. The graph is a work in progress... lol
<gmaxwell>
how many times per _day_ does ethereum have a 5 block reorg? but you list "accidental split" on bitcoin for a 5 block reorg? then draw one year of ethereum history against almost three years of bitcoin? lame sauce.
<gmaxwell>
then don't even point out where eth splitting created an irreconcilable chain fork which persists to this day which caused well over 100 grand in losses for users (Exchanges that lost funds due to replay)-- something which as of yet has never happened in bitcoin.
<stonecoldpat>
I've split the graph into two rows because it wouldn't fit into a page - I can see how it might make it look like a direct comparison. In the text I've explained the emergency of classic and the need for replay.
<stonecoldpat>
emergence*
AaronvanW has joined #bitcoin-wizards
<MaxSan>
lol
<MaxSan>
it is misleading from the way the graphics are laid out
sipa has left #bitcoin-wizards [#bitcoin-wizards]
<MaxSan>
the time frames scopes at first glance appear to be over the same time frame
<MaxSan>
like the width of the boxes
<MaxSan>
i aint sure what to call it
<MaxSan>
you cant stretch a time because you have no room for text in an infographic its misleading as hell
<MaxSan>
again, soft forks are not the same as chain splits, unless they split the chain...
Giszmo has joined #bitcoin-wizards
<MaxSan>
stonecoldpat: ^^
<MaxSan>
same applies for hardforks actually but by nature, that is what they do.
<stonecoldpat>
At least I've learnt this evening i'm not good at diagrams. I'll get that fixed :)
<gmaxwell>
:P
<MaxSan>
:D
<andytoshi>
ignoring the spacing things that others have pointed out, i think it needs to be clearer that a hardfork is a "flag day, you must upgrade!" whereas the other things are not .. so you could run a bitcoin node from 8 years ago (assuming you had a lot of patience and the 0.8 leveldb bug didn't nail you) and still come up to speed
smk has joined #bitcoin-wizards
<andytoshi>
also the lengths of these forks ought to be mentioned somehow (good luck :) that's even harder to diagram), stuff like the DAO where there has been a permanent split in which $50M was directly re-assigned ownership by the network and tons of people lost money trying to adapt to
<andytoshi>
vs, say, all the listed softforks which afaik were 0-length "forks"
<MaxSan>
that would be an interesting metric to visualise actually
<MaxSan>
even for temporary chain splits of a few blocks with BAU with orphaning blocks