Questions about soft fork

[pooya87]

When you are thinking of BIP148 don't think about WHAT it activates, but think about HOW it activates it (or rather wants it activated) and maybe you understand my view better.

But we can't, or shouldn't, speak for all of them, no? The UASF/BIP-148 was merely a proposal on how to have Segwit activated, an upgrade that many people in the Bitcoin community truly wanted. Plus if it's your opinion that 5-10% couldn't force the miners to activate an upgrade, OK. But it did with Segwit, because there was unquestionably more than 5-10% of the Economic Majority that actually wanted Segwit.

You are confusing two separate matters. There is a difference between "wanting SegWit" and "wanting SegWit at any costs". Majority of people wanted SegWit but only a handful wanted it at any cost. And that is what BIP148 is, going against what has worked in all bitcoin soft-forks (eg reaching 95% majority) and splitting the network threatening bitcoin's security, reliability and blockchain's immutability just to activate SegWit.
In fact if you paid attention in those days you would have seen a lot of users state that they do not want SegWit IF it leads to a chain split.

It wasn't like the laughable BCash.

It is like that more than you think. There is no clause in the proposal to check what the network thinks (ie. miner's vote), it just dictates that anybody who wants the change (whatever it is, whether SegWit or bcash or can be anything else in the future) can reject any block that doesn't activate it and split the chain!!!
That goes against everything bitcoin stands for and it is a malicious attack.

[garlonicon]

Also, we had the same kind of attacks for Taproot, but again, it was a minority: https://web.archive.org/web/20210416030248/https://bitcointaproot.cc/#faq

Is this a User-Activated SoftFork (UASF)?

No. This activation uses BIP 8 to ensure a safe and clean activation coordinated by miners. It is therefore a Miner-Activated SoftFork (MASF). However, it does not give miners the additional power to veto Taproot, and should miners neglect to coordinate an early activation, will still activate Taproot during late 2022. In that fallback scenario, it is accurate to consider it to have become a User-Activated SoftFork (UASF). Miners have publicly indicated that they also support Taproot themselves, so it is expected that activation under the normal MASF routine should proceed smoothly, and no UASF fallback will be needed.

So, it was not UASF only because things were activated by miners. But in case of not activating that, those users would cause a chain split.

[Wind_FURY]

When you are thinking of BIP148 don't think about WHAT it activates, but think about HOW it activates it (or rather wants it activated) and maybe you understand my view better.

To put everything in context we are talking about Segwit = WHAT, and what happened during 2017 = HOW.

But we can't, or shouldn't, speak for all of them, no? The UASF/BIP-148 was merely a proposal on how to have Segwit activated, an upgrade that many people in the Bitcoin community truly wanted. Plus if it's your opinion that 5-10% couldn't force the miners to activate an upgrade, OK. But it did with Segwit, because there was unquestionably more than 5-10% of the Economic Majority that actually wanted Segwit.

You are confusing two separate matters. There is a difference between "wanting SegWit" and "wanting SegWit at any costs". Majority of people wanted SegWit but only a handful wanted it at any cost. And that is what BIP148 is, going against what has worked in all bitcoin soft-forks (eg reaching 95% majority) and splitting the network threatening bitcoin's security, reliability and blockchain's immutability just to activate SegWit.
In fact if you paid attention in those days you would have seen a lot of users state that they do not want SegWit IF it leads to a chain split.

Because why? Because Jihan Wu and his friends from the mining cartel were delaying, and politicizing the miner activation process instead of what truly its purpose is, which is just a signal to let everyone know that they are ready for an upgrade.

Shaolinfry made his proposal as a response, and "the rest is history", https://bitcointalk.org/index.php?topic=1805060.0

Every newbie who wants to learn more about Bitcoin should read that important proposal in that topic.

It wasn't like the laughable BCash.

It is like that more than you think. There is no clause in the proposal to check what the network thinks (ie. miner's vote), it just dictates that anybody who wants the change (whatever it is, whether SegWit or bcash or can be anything else in the future) can reject any block that doesn't activate it and split the chain!!!

That goes against everything bitcoin stands for and it is a malicious attack.

The miners don't speak for the whole network. If it did, then the network is centralized towards the Mining Cartel.

[DooMAD]

The UASF was actually gaining support further towards "Independence Day," with Eric Lombrozo and other Core developers starting to be louder in their support. Plus if it took not more than 5-10% of the network, the intolerant minority, to make the miners notice/listen, then I believe it's a success. It's not just the miners who can enforce the rules.

I didn't get the sense that many users were actually running a UASF-enabled client.  There was definitely a lot of noise, but I'm not sure how much of it was backed up by actions.  

Some of the UASF supporters were also engaged in this weird double-standard where they claimed that bcash was "stealing Core's code", yet they had to fork the GitHub themselves in order to create a UASF-enabled client.  Not to mention that you can't exactly "steal" open-source software which is shared freely.  

[pooya87]

Jihan Wu and his friends from the mining cartel

They actually created the shitcoin called bcash using the same principles of BIP148, it was even called MASF (mocking the UASF thing). That is a minority group creating a fork disregarding the rest of the network (including miners' votes).

The miners don't speak for the whole network. If it did, then the network is centralized towards the Mining Cartel.

I never said they do anywhere! But you can't deny that miners are an important part of the network and attacks like BIP148 are completely ignoring/eliminating miners.

[BlackHatCoiner]

I sometimes hear the phrase "it breaks consensus" from hard fork critics. Did Bitcoin Cash break consensus? It surely upsets the network for a while, but if such upset were to break consensus, it isn't a strong consensus to begin with. Breaking backwards compatibility, technically speaking, does mean breaking consensus, but economically speaking, you are free to vote against the hard fork by selling the post-hardfork coins for the pre-hardfork.

[tromp]

I sometimes hear the phrase "it breaks consensus" from hard fork critics. Did Bitcoin Cash break consensus?

Yes, they did, not just by allowing larger blocks, but by in fact *requiring* a block larger than 1MB in size.

As a result, BTC block 478559 was rejected by Bitcoin Cash for not exceeding 1MB [1]:

Code:

ERROR: AcceptBlock: bad-blk-too-small, size limits failed (code 16) (block 00000000000000000019f112ec0a9982926f1258cdcc558dd7c3b7e5dc7fa148)

It took a while longer for Bitcoin Cash to reach height 478559.

[1] https://connortumbleson.com/2017/08/02/bitcoin-cash-bcc-is-born/

[wajik-tempe]

Question: is the invalidation of an old, valid rule considered part of soft fork? Yes, according to the wiki. So, you can mine an invalid block, that is valid in old clients terms, and broadcast it in the old client network. Does that encourage old clients to switch to new version, since they might hear on blocks that are likely to reorg? What could be the excuse of a non-Segwit node to stay in non-Segwit?

if a non-SegWit node wants to stay on the old version of the protocol, they can do so, but they will not be able to validate transactions that use the new SegWit rules. This means that they will not be able to participate in transactions that require SegWit, and they may miss out on certain benefits such as lower fees. A soft fork adds stricter rules, but they are still compatible with the old rules. Invalidating old, valid rules is not part of a soft fork

[BlackHatCoiner]

This means that they will not be able to participate in transactions that require SegWit, and they may miss out on certain benefits such as lower fees.

Actually, they can. Constructing a SegWit transaction is possible regardless of the node you're running, even though I'm not sure if Core would allow you to do that.

A soft fork adds stricter rules, but they are still compatible with the old rules. Invalidating old, valid rules is not part of a soft fork

You've got it wrong. Invalidating the old rules is what's the soft fork is all about. These "stricter rules" you mention in the former sentence is the restriction of the old rules.

[Wind_FURY]

Jihan Wu and his friends from the mining cartel

They actually created the shitcoin called bcash using the same principles of BIP148, it was even called MASF (mocking the UASF thing). That is a minority group creating a fork disregarding the rest of the network (including miners' votes).

The miners don't speak for the whole network. If it did, then the network is centralized towards the Mining Cartel.

I never said they do anywhere! But you can't deny that miners are an important part of the network and attacks like BIP148 are completely ignoring/eliminating miners.

I never denied the importance of miners, but focusing the discussion during SegWit's activation, they did use miner-signalling as a political tool to delay, or even as an attempt to stop the soft fork. The UASF/BIP-148 was a necessary move to distribute power throughout the network.

Plus saying that, BIP-148 specifically, is an attack against the network is wrong. There were Core Developers who supported it. Some even preferred it.

https://en.bitcoin.it/wiki/Segwit_support

I'm sorry for the late reply, I didn't see your post.

[pooya87]

There were Core Developers who supported it.

That's never a valid argument if you ask me. For one, we have had Gavin Andresen and Mike Hearn were also core developers once and they both went nutz and made questionable moves and statements. One supporting faketoshi and the other spreading FUD about bitcoin...

[Wind_FURY]

There were Core Developers who supported it.

That's never a valid argument if you ask me. For one, we have had Gavin Andresen and Mike Hearn were also core developers once and they both went nutz and made questionable moves and statements. One supporting faketoshi and the other spreading FUD about bitcoin...

Valid argument or not, it's still not an attack. It was merely one of the mechanisms to have SegWit activated. When the miners were acting against the interest of the network, through BIP-148 it was shown that there are checks and balances, which prevents the centralization of power towards the mining cartel. It was a learning experience for those who were directly involved, and for the other participants of the network.

[DooMAD]

There were Core Developers who supported it.

That's never a valid argument if you ask me. For one, we have had Gavin Andresen and Mike Hearn were also core developers once and they both went nutz and made questionable moves and statements. One supporting faketoshi and the other spreading FUD about bitcoin...

Ideas are best judged on merit and not by which individuals support them.  It's not a popularity contest and we don't need to introduce a cult-of-personality element.  Appeals to authority aren't the correct method for network governance.

As for what's considered an "attack", that's always going to be subjective.  Some ideas are more reckless or dangerous than others, but ultimately it's all just people doing what they want to do. 

[Wind_FURY]

There were Core Developers who supported it.

That's never a valid argument if you ask me. For one, we have had Gavin Andresen and Mike Hearn were also core developers once and they both went nutz and made questionable moves and statements. One supporting faketoshi and the other spreading FUD about bitcoin...

Ideas are best judged on merit and not by which individuals support them.  It's not a popularity contest and we don't need to introduce a cult-of-personality element.  Appeals to authority aren't the correct method for network governance.

As for what's considered an "attack", that's always going to be subjective.  Some ideas are more reckless or dangerous than others, but ultimately it's all just people doing what they want to do. 

Taken in the right context, the reason why I emphasized that the Core Developers supported an idea is precisely because of the merit behind the idea - at that time. The Scaling Debate was a unique period in Bitcoin's history, which probably no one in the community knew what move was truly going to be the right move. I could only imagine how enlightening it was for everyone directly involved.

[pooya87]

Taken in the right context, the reason why I emphasized that the Core Developers supported an idea is precisely because of the merit behind the idea - at that time.

The abstract idea of nodes voicing their preference is a good one with merit but BIP148 is not and it is instead a dangerous proposal because of a simple fact that it disregards majority. It simply has no way of checking or caring what the majority says, whether it is 1% or 99% it will enforce the change!

In any change, regardless of what it is, we must reach consensus amongst network participants otherwise it goes against the idea of being decentralized.

[BlackHatCoiner]

The abstract idea of nodes voting for a change by just running a full node software is just terrible, in my opinion. The majority isn't determined by the number of Bitcoin clients, for if it was, we wouldn't have mining; we'd select the order of the transactions by voting as in Proof-of-Stake, but with instances of Bitcoin clients instead of money.

There has to be a way of collectively voting for a change without giving absolutely every voting power to the miners, isn't there?

[HmmMAA]

There has to be a way of collectively voting for a change without giving absolutely every voting power to the miners, isn't there?

Nope , there isn't . Imagine voting in a country that has no electoral catalogues and anyone could vote as many times he want , could you ever have a valid election ?
That's a problem that anonymity causes , to have a valid/honest result you would need known entities . That's why miners are the best choice . They are known and you can see if you agree or disagree with their choice . The economic incentive will drive them to decide the best for the network or they will lose clients and profit .

[Wind_FURY]

Taken in the right context, the reason why I emphasized that the Core Developers supported an idea is precisely because of the merit behind the idea - at that time.

The abstract idea of nodes voicing their preference is a good one with merit but BIP148 is not and it is instead a dangerous proposal because of a simple fact that it disregards majority. It simply has no way of checking or caring what the majority says, whether it is 1% or 99% it will enforce the change!

In any change, regardless of what it is, we must reach consensus amongst network participants otherwise it goes against the idea of being decentralized.

Disregards the majority? Although BIP-148 was controversial, SegWit was still wanted by the majority. It only became a necessary "movement", or a kind of "mechanism" for SegWit's activation, because Jihan Wu and the mining cartel were playing political games by using miner signalling as a political tool which delayed activation.

[BlackHatCoiner]

Nope , there isn't . Imagine voting in a country that has no electoral catalogues and anyone could vote as many times he want , could you ever have a valid election ?

Absolutely not, and this is fundamentally how a UASF works.

That's why miners are the best choice . They are known and you can see if you agree or disagree with their choice

First of all, miners don't have to be known. There is already a fair percentage of miners who are anonymous. We only know that the majority of the hash rate comes from known pools. Secondly, if I can't vote, how can the miner know if he's made the right choice?

[DooMAD]

vote

Before we get too in depth, I'd pause to question this particular aspect.  I believe when satoshi used the word "vote" in the whitepaper, it was in order for people to latch on to an existing concept that they understand.  I don't think it was intended to imply that consensus is akin to voting, though. 

Voting requires permission.  Running code to enforce network rules is an enactment of will.  The consensus mechanism then allows those who agree on a common ruleset to build a chain together.  In practice, it's a very different concept to voting, in the traditional sense.