Topic: Soft Fork to Increase the 21M Limit? (Read 3939 times)

A miner majority (i.e. like a softfork) can force changes to the Bitcoin protocol using the method described here: https://bitcointalksearch.org/topic/increasing-the-blocksize-as-a-generalized-softfork-1296628

The above proposal was for blocksize limit increases (mildly controversial), but the same idea could in principle be used for other changes that would otherwise require a hardfork.

This 95% rule is also a rule that can be changed at will, nothing enforce it. If I'm a large mining pool, I will trigger it with even one block and start to produce new blocks immediately

It seems the last defense is exchanges, if exchanges do not upgrade to new software, then all the coins from those new blocks will worth nothing and become useless. You can setup a new exchange for exchanging those new coins, then most possibly they will become some kind of alt-coin exchange with neglectable USD value/volume

I think this is the biggest weakness of bitcoin, e.g. no systematic way to secure the rules from being modified, thus make the software itself a single point of failure

The only measurement we have today is a FOMC like commiter's group, and there is no clear guidance in their decision making, and Gmaxwell has already resigned

Yes there is.. Soft-forks only reach super-majority (and the feature switched on) if the block versions produced by miners reaches reach a number indicating the change. 95% of 1250 concurrent blocks is the exact mechanism - miners have full choice over what software they run.

Soft forks are a feature of Bitcoin. I don't see any way we can technically prohibit them within Bitcoin's architecture. Moreover, soft forks can be enforced by miners unilaterally, without others consent. That's because soft forks make rules more strict/add new rules, and these rules are enforced by miners.

Noway this will make it an alt-coin

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

In fact I'm surprised when Pieter promote this soft fork trick. It is cunning and stealthy, not a formal way to implement change, it is very like trying to find the security hole in core software and exploit to reach hacker's desire: Let's see is there any more things that we can do without being discovered by old core clients

In fact if you are a honest dev, you should immediately take measures to patch such kind of security weakness, not exploit it

It's just really stupid if you think about it. Who the hell wouldn't sell all of it if Bitcoin is about to do something diametrically opposed to what it was designed to do? (to not fall under the same tricks and traps as central banking). Everyone would sell since you know the supply is going to get increased therefore your money is guaranteed to devalue, so as the price goes down, miners would ultimately end up with less revenue. Im just hoping in the future Bitcoin mining is more decentralized, we can't be praying forever that miners will not do something stupid.

I think increasing/removing the 21M Limit would result in panic reactions. Adding extra decimals places is a much safer way to achieve the same thing

I would rather see the bitcoin code put a cap on max hash per connection to break up the mining monopolies or something of that nature..

Any change that would increase the 21M limit would create an alt-coin, that is not bitcoin.  If the question is something akin to namecoin running in parallel to bitcoin, then I think at best you'd get a similar valuation to namecoin, probably less since at least NC has an additional function.

These types of arguments were made before the last halving about miners shutting down etc.  Sure, there may be some miners who shut down really non-profitable rigs (perhaps old Avalons for example if they are still running), but if they are covering power costs alone, they'll keep them running.  The sunk cost from the purchase price is just that, a sunk cost.  In November 2012, difficulty dropped slightly after the halving (the first and second readjustments were down slightly around 1-2% and 9-10% iirc) but within about 8-10 weeks was up.  Given the number of new ASICs that will be coming online in the next 8 months until the halving, I would expect something similar.  There may be a slight drop in July/August, then difficulty will recover.  Time will tell.

I completely agree with your statement and stand by your side because I don't see any value in increasing the limit of Bitcoin beyond 21 million it will just break the Bitcoin market and make it worthless.

Why the hell would we want to increase the 21M limit?

1 bitcoin is dividable into god knows how many decimals. Scarcity of coins is key otherwise the price would plummet. 

I disagree. The "increase blocksize" part is actually not the most important. It just happens so that we hardly have time for a smooth hard fork (IMO it should take 6-12 months at least for a hard fork preparation, plus many performance issues need to be addressed beforehand), hence a time-buying soft fork is preferred.
More important are: malleability "fix", easier script upgrades, fraud proof enhancements.

But deploying SW with a hard fork might be better design-wise IMO.

New minting of the original _bitcoin-1_ would clash with the non-upgraded full nodes.
A new soft-fork-currency is like bitcoin-2, non-upgraders would not notice it has emerged, they will happily accept them (blocks which contain bitcoin-2) as valid blocks.
Soft forks would not violate the former validation rules, it just makes them more narrow (such as, former validation rules _AND_ blocks must contain bitcoin-2 information).

Actual implementation may not be so beautiful one, depending on the available features/space.

Depending on the miners' demographic distribution, it probably works in the same way as the original bitcoin issuance (hashing competition).
Launching a soft-fork-currency will stimulate the new mining competition, it is different from the random coin-creation on the Bitcoin blockchain. It creates scarcity because it will be directly backed by the world's biggest mining facilities (only one), and difficulty to make a soft-fork will ensure it will be done only in a really needed situation.

Miners also can promote the soft-fork-currency by giving some discount for their tx-inclusion service.

To increase the limit of 21M is not necessary because you also can increase the decimal places after the point without increasing the total amount of Bitcoins mined.

Dogecoin introduced a inflation rate but I think this is not necessary.

Um no.. I am not for any increase of maximum coins over 21 million. I much prefer true scarcity over moderate scarcity any day. An increase of total coins in my eye is a betrayal and unneeded. If this were to happen I would probably get out, and I bet many many others would rush out too.

And for what? To pay the miners? Why would we need to subsidize the miners?
The miners made an investment, a bet, that they could ROI/profit whatever. If they bet wrong why should we give up the value of our coins to help them out?

Sorry you guys bought up all these miners and created a mining bubble that may be about to burst, but tough shit. I don't see anybody subsidizing me when I make a bad trade.

When the halving comes if BTC doesn't rise enough for all the miners to be profitable then I guess half of them will have to shut down before the remaining half will be operating in the green. It will still surely be plenty of hash to keep the network secure no?

"Lets increase the amount of total bitcoins to get these people out of a bad investment"... I think not..

When you invest in miners or anything you risk taking a loss VS the chance to profit. Yeah it sucks when you make a bad investment but that's the way it is and you should EAT IT just like everyone else.. I eat em all the time, why shouldn't the miners?

It is my understanding that SW is not a bad thing, however, as a soft fork shoved into the coinbase, it is messy.

To have my support it would have to be a hard forking implementation, and if it received consensus, why can't we pass an increase in block size along with it?

This is adding complexity to bitcoin, and overhead for developers.

Actually soft fork often means this:
1. A super majority of miners start to ENFORCE the change.
2. Not all full nodes enforce the change, but allow it.

New minting would clash with any of the 6000 full nodes that did not upgrade.
SegWit would not from my understanding because of "hacks".

Failure to attain super majority from miners would mean the change could not be enforced unless everyone upgraded (unlikely if you can't even convince 5 mining pools to begin with).

Quorum? There is no voting for a soft fork. Miners either upgrade their client or do not. In a soft fork, all old nodes can see the new blocks, can still have transactions sent to addresses that they recognize, and can spend their existing coins.

The only way to ensure a soft-forks success is to backport the change, as some miners use older versions of bitcoind. They can continue to use the older version until quorum is met. If not, they might produce invalid blocks (according to the 95% of the network prepared to adopt the change)

It's hardly forcing miners, when they're responsible for signalling support.