Topic: The Barry Silbert segwit2x agreement with >80% miner support. - page 69. (Read 120032 times)

I'd rather have a half assed or joke proposal than just sitting stubbornly on our asses like Core as bitcoins capacity issues become more critical by the day.

Correct and it's been sitting at 38-41% for ages. You showed the pic showing 48% daily rate making it look like it was climbing. The daily rate has even peaked over 50% in the past.

One-week variance dropped from 42.1% to 41.3. Insignificant change confirmed.

this silbert thing is just a joke. there is nothing written in code so far. where is the BIP?

Why not take another look right now and see for yourself?

Variance? The one week figure has significant variance? Really?

That much is true, and I'll bet money that it will not be the new, ASIC-free, non-Bitcoin altcoin that you're wanting to turn Bitcoin into. 

The very same thing which makes Ver&WU to think that people gonna name their alt as "bitcoin".

That's for people to decide and we will never know unless it happens and trust me it will happen. There will be many forms of bitcoin and only one of them will bear the "bitcoin" name.

Barry should try to get Blockstream and Core into the agreement discussion. If they refuse then perhaps we will know first hand who is truly at the root of the attack on bitcoin.

I actually would propose a fork with PoS for the consensus, while keeping PoW for the coin creation.  The PoS would not be remunerated, it would just be a "free service to the community" in the same way that people are now running (powerless) full nodes.

But let us stick to your idea to change the PoW scheme.  For instance, scrypt instead of sha-256.  This has many facets to it.

First, I'm always puzzled why bitcoiners want to have something in bitcoin, that is already up and running elsewhere: if you change PoW into scrypt, you have Litecoin.  Litecoin is essentially bitcoin, but with PoW being scrypt.  Litecoin was invented, exactly, because already in 2011, people understood that asic owners were the deciders of the coin - it failed however, because there are also scrypt asics.  Note that DASH invented X11, a joyful mix of 11 different hash algorithms, "to resist ASICS".  Useless:

https://asicminermarket.com/product-tag/x11/

For the moment, cryptonight is still standing.  But with monero's increasing price, I wonder whether it will stand a long time.

There are serious efforts to make essentially ASIC resistant algorithms, and I think that burstcoin is using a so big memory needing PoW algorithm that you need disk space of TB for it ; it becomes "proof of capacity".  If I were a hard disk vendor, I would hype such coins.

But what I want to say, why stick with bitcoin if you want to use the system of another coin.  Simply sell your bitcoin and buy the other coin !

But let us consider that, nevertheless, you try to change the PoW function in bitcoin. Let us say for the moment that we take scrypt.  So the new bitcoin is simply a version of litecoin, but with a different initial chain. What happens ?  This is in fact a totally new coin you create, with not even the option for the actual miners to "switch camp".  So "old bitcoin" will simply continue to function, with all its stuff.  The only thing you did, was create a totally new coin, that happens to have the same coin possession as the bitcoin owners at the point of forking.   But it is, say, litecoin, but with the former balances of bitcoin.

People will now own two coins: the new alt coin you just made, and their original bitcoins of course.  In order for the new coin to take all the value of the old bitcoin, people need to sell all their old (original) bitcoins, and buy more of the new lite-bitcoin.  What makes you think that people are going to dump their *original bitcoins* to buy the new altcoin, and they didn't do that with the existing litecoin ?

After all, you have to sell your *original bitcoins* to buy more of a new coin that is in all respects like Litecoin.  

Note that deciding on which coin to sell, the "new free money you got" (the lite-bitcoin) or your "original bitcoins", you have to be careful.  If you make a mistake, you lose everything !  It is not a matter of *your own choice* but it is a matter of your own prediction of what others will do, while others will try to make you dump your "valuable" coin, and buy the shit coin instead.  Which one will it be ?

I'm not sure which I find more humorous, your unfounded delusions about ASICBOOST or your delusion that Bitcoin will ever be ASIC-free again.

I understand you invested heavily in Bitmain toys lately and you are having nightmares when you hear UASF and PoW change or anything which is going to make your hardware useless (or your precious ASICBOOST) but you should understand, this is for your own good, for all miners out there.

You'll take some losses but no big deal. In the long run you'll be the one who's laughing.

We know that.

That's why we propose not UASF only but UASF + PoW change, what did you think... Current situation with the miners is problematic. Their power levels need to be re-adjusted.

Can we also dispense with spewing out some random false dichotomy like this?


It is actually possible for a rational human to believe some of the things BU believes in (or in this instance, one of the things BU believes) while not believing in BU.  

Edit:mindrust, if you think Core incapable of creating a node client capable of handling the same amount of information that a 1990s era IRC server could, perhaps it's time to stop drinking the Core-ade.

I think the first requirement of a user is to be able to do transactions in a cheap and permissionless way.  It is very funny to run your own node in your basement if you're a geek, but you don't need that to do transactions.  If you ask a user whether he wants to be able to transact cheaply and without any hassle, but he has to invest in a few TB hard disk and maybe upgrade his internet link if he wants to be a geek in his basement ; or whether he wants to keep using his old core-duo PC with his 300 GB disk for his geeky full node in his basement, but with the current transaction difficulties, I think the answer will be very fast ; but it is sufficient to make a hard fork and ask the market to vote over the two models: that's how the free market is supposed to work:
do you want a big car that can run over all kinds of terrain but can only reach 100 mph, or do you want a sports car that can do 200 mph but only on good roads ?  Sell both !  The customer will chose.  There's maybe a public for both.

By far most users of bitcoin have never run a full node.  They don't need it.  The full node copies simply what the miners made, and the miners are the entities that keep the miners in check.  The wallet is what downloads the right data from a full node, that doesn't need to be trusted because the wallet can verify the cryptographic authenticity, namely that the data it received, was the data that miners made.

But this is already the case.  And solely due to bitcoin's design, which determines consensus as by hash power consensus, where miners are put in a game that leads to a Nash equilibrium of all of them playing by the same rules, which we call "consensus".  What changed from the original conception, although highly predictable, was the clustering of mining power into pools which are not very high in number: 5 for majority, 20 for essentially 99% majority.  
Bitcoin's design, on top of that, makes it extremely difficult for a miner pool to not be in the majority consensus with the very slow difficulty adaptation, which kills all small minority miners deviating from the consensus.

What does this bring us ?
a) Even if there are only 20 mining pools, if they remain decentralized, meaning, they don't collude, meaning, they don't "sit in a room and agree on something", then bitcoin's protocol remains what it is.
b) If a serious hash majority of these mining pools (say, 8 of the big ones) colludes, meaning, they sit in a room, and decide to take action together, then that action will be what bitcoin's new protocol will become.
c) If two comparable camps appear in these mining pools, then they can remain locked up in a prisoner's dilemma like in (a), or they can "pull the trigger" and break miner consensus.  ONLY AT THAT POINT, users can say something (in the market).  The only thing a UASF could obtain, is that the trigger is pulled somewhat faster by making believe at least one camp that they might win.  But the nodes not necessarily being representative for the market, that's a risky bet.

Users can vote in the market if miners break miner consensus, and fork into two coins.  As long as there is miner consensus, users can just decide to accept that, or to leave bitcoin and their holdings behind them.

This was not the case as long as miners were manifold, and dispersed throughout the P2P network, because the P2P network FILTERED blocks that came from about anywhere.  But this is not the case any more.  Miners don't need the P2P network with Joe's node in his basement to get the blocks from their peers.  It would make them lose too many blocks.  You can estimate the connectivity between miners by the orphan rate.

https://blockchain.info/charts/n-orphaned-blocks?timespan=60days

There have been 3 orphaned blocks more than a month ago, and nothing since !!

You would almost think that there is only ONE MINER POOL, so good are their connections.  Joe's node in his basement is totally out of this game.

3 blocks in 2 months, that's essentially 3 collisions when there have been 10000 blocks produced.  In other words, miner pools know each-other's blocks within a time lapse of the order of 10 minutes * 3 / 10 000 = 180 ms !!

Joe's node in his basement has filtered ZILCH between miners.  In 180 ms, they know one-another's blocks (otherwise, purely statistically, they would have published many more orphaned blocks).

Taking the ability to run full nodes from the users and giving it to the elite mining companies who have access to resources will disturb the balance. Users will have no say  in bitcoin's future. Miners will be the only deciding factor.

Too powerful miners will be a too ez target for governments. Goverment killz minerz, bitcoin dies.

That's what you all BU shills do not understand from the beginning. Please Fork the fuck away asap, take your BUcoin and let us be.

Just to interject

From what I have read on this dispute it appears very clear that regardless of their technical merits neither

1) Variable blocksize
or
2) SegWit alone without a block size increase

can achieve consensus at this moment in bitcoin history. Thus neither are viable pathways forward right now.

Some compromise position must be adopted. It does not matter if it is a perfect solution. All it needs to be is safe and good enough to achieve widespread if grudging consensus.

If there were no safeguards in place, then it would be easily gamed, which is why several safeguards have been introduced and more are under review.  I want to take every reasonable precaution to prevent gaming the system.



Yes, it's fair to say I wasn't anticipating you completely contradicting yourself by saying the the proposal was too conservative and the adjustments were too small.  How large would you like the potential increases to be?  



Has anyone actually done any research into how many transactions drop out of the mempool because they took too long to confirm?  All of the spam transactions people link to when they post threads about spam attacks have already been confirmed into a block.  So see if this can penetrate the thickness of your skull, the 1MB limit didn't prevent the spam from increasing the total size of the blockchain.  The spam was confirmed into the blockchain.  All the 1MB cap achieved was to make it take longer to confirm.  

Spam.  Still.  Got.  In.

While I don't doubt you have the right intentions, the fact is you don't seem to understand what it is you're even arguing for.  The scaling proposal you support does not prevent motivated attackers from increasing the total size of the blockchain.  There isn't a single proposal that does.  Which is why I want to look at things like increased fees for repeated transactions from the same address over a short period.  Something that would actually help deter spam, more than any arbitrary cap ever could.