Bitcoin’s Segregated Witness: More Than Just Malleability Fixes and Scaling

[AngryDwarf]

From "Do you support Segwit" https://bitcointalk.org/index.php?topic=1864477.msg18541674#msg18541674

Not in its current form as a soft fork.

Implemented as a soft fork, it creates a two tier network where non-segwit nodes are fed a kludge of data which means they cannot validate transactions fully. Also it creates a 'two bucket economy' situation where segwit transaction types are heavily discounted. It also requires two merkle trees instead of one. Without miner majority, the implementation could create a hard fork leading to a bilateral split, as non-segwit nodes would see a longer chain, but segwit nodes would invalidate that chain as soon as segwit block is mined and built on top of it. That is also why UASF without miner support is a terrible idea. Sorry if my technical analysis here is misinformed, please correct it.

As a non contentious hard fork with miner majority support (one chain left standing as the longest active chain, the other killed off), these issues could be eradicated. It could be implemented alongside a blocksize increase and the segwit data space and been given a 1:1 weighting. That would likely have received miner consensus too.

I'm certainly not technically clued up enough to state with certainty that perhaps a better technical implementation could have been achieved had it been rolled out as a hard fork, but it would have reduced the code complexity and technical debt trying to achieve what is a very poor definition of 'backwards compatibility'. It is possible that as a hard fork, some of the soft fork kludges would not have been needed, and I wonder how much more could have been achieved for such a major protocol upgrade.

So I suppose the best available option in this poll, is 'don't care'!

All I ask is for people to clear up any misunderstandings or misinformation I have been fooled into.

[1714619007]

1714619007

[1714619007]

1714619007

[1714619007]

1714619007

[1714619007]

1714619007

[franky1]

Crayzians (crazy asians) are driving the price up faster than I've ever seen before

1. look at the unconfirmd spam
https://blockchain.info/charts/mempool-size?timespan=1year
june/july spike -hmm i wonder which team needed something to be implemented so needed to create drama that month.. oh blockstream(core) CSV
october -> spikes -hmm i wonder which team needed something to be implemented so needed to create drama that month.. oh blockstream(core) segwit

2. look at the code rules that allowed fee increases
hmm i wonder which team removed the fee priority - oh yea blockstream(core)
hmm i wonder which team removed reactive pricing when demand was low to replace it with average fee to keep prices up - oh yea blockstream(core)
hmm i wonder which teams says "just pay more" is best economics. rather then make code rules that actually work - oh yea blockstream(core)

3. dont blame "asians" when its the coders decisions at play

[franky1]

OP, you made a grave mistake by not self-moderating this thread. Anything that is even remotely positive about Segwit or Bitcoin Core, even when we are talking about *facts*, will get smashed by the group of hired goons. You can find them in all related threads, see: franky, jonald, Alex, kiklo. There are some that occasionally come and go, see: zimmah.
This thread is already flooded with such responses, even though the article mentions *some factual improvements*.

factual improvements?
lol
which are where..
oh im guessing "if" 100% move funds to segwit keys and the utopian unicorn stops native spammers.
i say utopian unicorn, because segwit does not stop native spammers. it actually helps native spammers

EG
v0.12 - 4k txsigoplimit 20k blocksigoplimit (under 10 second validation time)
v0.14 - 16k txsigoplimit 80k blocksigoplimit (under 8 minute validation time)
plus other new spam attacks that even carlton banks revealed.

lauda.. you loved having the "quadratics is bad" arguments for a whole year... even you should understand quadratics worse with segwit code limits if used by native keypair users.
wake up, spammers wont use segwit keypairs

[ebliever]

OP, you made a grave mistake by not self-moderating this thread. Anything that is even remotely positive about Segwit or Bitcoin Core, even when we are talking about *facts*, will get smashed by the group of hired goons. You can find them in all related threads, see: franky, jonald, Alex, kiklo. There are some that occasionally come and go, see: zimmah.
This thread is already flooded with such responses, even though the article mentions *some factual improvements*.

I'll keep that in mind. I've never done a moderated thread as I do have an antipathy towards censorship, but it may be time to learn. (Even if I were to do so, I'd only censor rude/vulgar posts and invite them to try again, not ones I disagree with.)

As it stands, I think most people just check the OP and link if there is one, and don't wade into all the comments most of the time. But if they do, I would be stunned if the average reader just wandering into this debate found themselves persuaded by the over-the-top, often rude and vulgar (not all, but a few), and kooky conspiracy-minded posting of the anti-SW crowd.

The average person, for example, is going to note that most Bitcoin organizations and developers favor Segwit. If it's such a terrible idea as the more obnoxious ranters say yelling specious claims about technical risks and so forth, then why are so many people supporting it? Why do we keep hearing about BU's code failing but not SW crashing and burning? There is a huge discordance between anti-SW rhetoric and reality, and they aren't making themselves persuasive with their wild rhetoric. (Insofar as a couple of them are more restrained and trying to be more measured in their responses, I thank them for that. But they are not representative.)

All in all, I thank them for bumping the thread, and driving people away from their viewpoint. It helps people understand the benefits and need for Segwit, and why we likely need to push through a UASF despite the headaches. I'd remind everyone that the clock is ticking, and every day BitMain mines with covert ASICBOOST the more mining become centralized under their domination. (Please don't insult my intelligence by saying that just because they admitted they spent the time and effort to design ASICBOOST into their chips doesn't mean they are using it.)

[AngryDwarf]

EG
v0.12 - 4k txsigoplimit 20k blocksigoplimit (under 10 second validation time)
v0.14 - 16k txsigoplimit 80k blocksigoplimit (under 8 minute validation time)
plus other new spam attacks that even carlton banks revealed.

And don't forget the tragedy of the commons:

Removal of Priority Estimation
- ------------------------------

- - Support for "priority" (coin age) transaction sorting for mining is
  considered deprecated in Core and will be removed in the next major version.
  This is not to be confused with the `prioritisetransaction` RPC which will remain
  supported by Core for adding fee deltas to transactions.

Removing coin age priority at this stage will push fees even higher, by causing long term coin holders to compete with the spammers. Perhaps it could be removed in the future in the name of fungibility, but I don't consider it a good idea with the current capacity issues.

[QuestionAuthority]

Crayzians (crazy asians) are driving the price up faster than I've ever seen before

1. look at the unconfirmd spam
https://blockchain.info/charts/mempool-size?timespan=1year
june/july spike -hmm i wonder which team needed something to be implemented so needed to create drama that month.. oh blockstream(core) CSV
october -> spikes -hmm i wonder which team needed something to be implemented so needed to create drama that month.. oh blockstream(core) segwit

2. look at the code rules that allowed fee increases
hmm i wonder which team removed the fee priority - oh yea blockstream(core)
hmm i wonder which team removed reactive pricing when demand was low to replace it with average fee to keep prices up - oh yea blockstream(core)
hmm i wonder which teams says "just pay more" is best economics. rather then make code rules that actually work - oh yea blockstream(core)

3. dont blame "asians" when its the coders decisions at play

I'm not blaming Asian buyers for anything. I love them for it. They're buying and holding fast and massively driving the price up. What's not to love?

I'm not pro-anything. As I said before, I don't care how the idiots in power change Bitcoin. Soft fork it, hard fork it, SegWit the shit out of it, roll it up in a blunt wrapper and smoke it, I just don't want anything done until all the crazy buying action stops and it peaks. Why spook the crayzians while they're buying like mad. Leave them alone.

[Alex.BTC]

As it stands, I think most people just check the OP and link if there is one, and don't wade into all the comments most of the time. But if they do, I would be stunned if the average reader just wandering into this debate found themselves persuaded by the over-the-top, often rude and vulgar (not all, but a few), and kooky conspiracy-minded posting of the anti-SW crowd.

The average person, for example, is going to note that most Bitcoin organizations and developers favor Segwit. If it's such a terrible idea as the more obnoxious ranters say yelling specious claims about technical risks and so forth, then why are so many people supporting it? Why do we keep hearing about BU's code failing but not SW crashing and burning? There is a huge discordance between anti-SW rhetoric and reality, and they aren't making themselves persuasive with their wild rhetoric. (Insofar as a couple of them are more restrained and trying to be more measured in their responses, I thank them for that. But they are not representative.)

My ass is jealous of the amount of shit that just came out of your mouth.