Topic: 1MB vs Gavincoin video explained in 3 minutes (Read 2841 times)

Full blocks are no problem at all. At least no urgent one. Microtxs go offchain for example. Gambling for 20 cents a bet or shit like that. Recently someone sent me 10k satoshi with a message attached for coin double scam. So as long as scammers can still afford to spam the chain with their "service" there isn't a problem.
I think people underestimate a lot how many of the txs are acutally unimportant dust txs and txs under 3$.

Depends which store.

I think as blocks become  fuller we'll see more services for payment offchain. But as long as there is no demand because fees are minimal there won't be any. When fees rise and people want to save on fees such services will quickly pop up i could imagine.

Nobody opens a business for which there is no demand. If fees rise, demand for cheaper txs will also rise and with it offchain tx processors. I could imagine blockchain.info doing such a thing for example. For them it should be easy to hack together things like that as they could serve as trusted party.

There should be quite a few smart solutions for offchain txs that haven't even been developed yet because as pointed out already: there is currently no demand for it because fees are so low.

Once fees are higher and there is some real demand for that, soon businesses for that will open doors for customers.
Gavincoin wants socialist central planning when in reality the free market effects sort it all out by themselves when left alone.

i think thats the key, we should be ready for that problem before it happend, i still cant understand why they didnt reach a concensus and thats the part i dont like, if "we" are not together we will fail.

i believe they were just testing it on test net, for all this time, or running it locally for further testing

it's better to prepare for a problem long before it happens

Can I use changetip to pay in a store? As far as I know, it's not designed for that.
I really would like to see a video of someone doing that.

I think changetip is such a thing?


-----



Gavincointards are going strong on this thread again ...

I haven't read the complete thread yet, but I have one question:
If this video was made 2 years ago, why aren't there any off-chain payment processors?
It's easy to talk about all this stuff in an academical way, but is anybody actually doing the things, they are talking about(obviously for at least 2 years)?

As far as I can remember from that video that I've seen, it feel like he's trying to 'force' it. I just don't think that jumping onto XT was the right move, and it might have been planned for a long time. If the consensus is reached for XT, Mike Hearn will be in control of the development on github, right?
I'm not sure who currently in control of QT. As far as I know some commits that Gavin has proposed were denied and they are implemented in XT (what he refers to 'minor changes').
The real question is what Mike Hearn is trying to do? He wasn't really active with development AFAIK.

So there isn't a really good solution for this. Are you sure that we are going to be reaching consensus this time? From what I can understand, if we don't have reach consensus (90%) for XT they are going to introduce "checkpoint blocks" to ignore the longest chain.
https://www.youtube.com/watch?v=DB9goUDBAR0

Aren't both proposed solutions non scalable?
How long could we go with 1MB + LN solution without a hard fork (which is what is tried to avoid at all costs) vs Gavin's blocksize increase?
Do we just accept with 1MB + LN situation that fees will keep increasing indefinitely?

Why do you disagree?  Is there any other way to do it?

There is nobody "in charge" that can make a decision for everybody.  There isn't any kind of "vote" that would satisfy everybody.

Bitcoin requires consensus, and in this situation consensus is impossible.  The only option remaining is for there to be multiple separate "consensus" each with their own fork of the blockchain.  Then some measure of chaos begins while the competing ideas play out in the real world and we see what happens.

With Gavin's current "way he is trying to solve things", the larger blocksize isn't being forced on most people.  There is an option available to anybody that wants it to support larger block sizes.  Once the miners (and merchants) exceed 90% on the matter the decision won't need to be made, effectively consensus will exist. If the miners never reach 90% on the matter, then everything stays the way it is today and others are free to try whatever solution they prefer.


There is no way "out of the mess", by its very nature all protocol changing decisions in bitcoin will always be messy.  The only thing that can be done is to accept the messy nature of it, provide software that makes clear what is supported, and hold off on activating any forking changes until the system has arrived at near consensus on the matter.

The consensus requirement has always been (and always will be) one of bitcoin's biggest weaknesses.  It's a great thing for stability in design (making it nearly impossible for any single entity or group to force their own changes on the majority), but it makes advancement and improvement extremely difficult to accomplish.

Even when the increase in block size was first proposed (months back) it was a mess, but now it is much worse. If I recall correctly in a recent statement Gavin (video) was sitting next to Mike and he said that a lot of people have been telling him to take the lead on the matter. However I wonder who these "people" are, since nobody likes when things are being forced on them.
Disregarding whether I agree with his solution or not, I definitely disagree with the way that he is trying to solve things.

My question would be: What would be the best way to get out of the mess that we're currently in and make a decision, even if it isn't the right one?

Only among the uneducated and easily manipulated.  Those that take the time to learn about the details are aware that in Satoshi's original software the block size limit was 32 MB. It wasn't until 2010 that Satoshi decided that he'd temporarily add a 1 MB limit.


You might think that you "know" that, but many people would disagree with you.

Yes.  Consensus is difficult.  Nobody gets to "decide" for everybody else, and everybody gets to decide together.  That's great when everybody agrees, but when there are multiple competing ideas and there is no agreement then it all becomes a HUGE mess.  In the end, several people will put forth their own preferences in software.  Some of those may "catch on" and the system could fork into competing currencies.  Some may die as the developers come to realize that they don't have as much support for their idea as they thought they did.

I'd like to think that the concepts that have the best analysis and research behind them will be the most likely to succeed, but in the short run it will probably be the ideas that have the most persuasive personalities involved, and in the long run it's likely to be "survival of the fittest" with many random and poorly conceived ideas suffering painful and horrible deaths along the way.

That is actually what I wanted to point out somewhere. Who gets to decide what is right and how?
I guess that is one of the problems. Everyone is trying to pursue their own agenda/way in regards to the problem even if that 'solution' has significant cons. Even though I'm in for gradual increase, I'm not so sure about the exact numbers.
Personally I think that 4 or 8 MB would be better to start rather than 20 MB.

Nice theory, but sidechains were not seriously talked about until about a year or so after the vid.

I'm not one of the devs, but the idea of scaling through subordinate chains which derive their value from a single primary chain was about the first thing which occurred to me since it was fairly obvious that a flooding network would not be especially practical if it were to remain a true peer-2-peer solution.  Not only is it kind of a no-brainer, but it opens the door for a variety of more specialized niche tuning, and some potential needs for a currency are mutually exclusive which makes a one-size-fits-all solution less desirable even if it were practical.

With a robust 2-way-peg (which is something I did not think of) a sidechain is a near perfect proxy for Bitcoin which doesn't bloat the infrastructure needs and doesn't impact the inflation/deflation characteristics.  Bitcoin as a backing or in native form is a sigle-point-of-failure which should be protected against all attacks.  If bloated and popularized it is susceptible to fungibility attacks through whitelisting and blacklisting which is something that certain people have advocated for a long time.  When used primary as a high-powered and highly distributed backing system it would be more difficult to attack successfully in this manner.

I loled at that video, it is clearly fud.  It is pretty liberal with the facts, and outright lies in places.  I am thinking this video was funded by those devs who have a vested interest in the sidechains, and making a lot of money from keeping the block size low.

Msg to ~retep:  I'll make a contribution again if need be.  I don't use reddit because the 'tard ratio on Bitcointalk is already galling enough and reddit is fast becoming as slimy as Wikipedia and Snoops before it.

On the new vid, try to work in the fact that those of us who thought along these lines struggled for a way to tie subordinate chains (or more generally, 'stores of truth') to the Bitcoin blockchain all along (me since 2011) but certain ideas about how to do it are more recent.

IIRC, the debate around and push toward bloat was as fierce back when this vid was made a year and a half ago as it is now.  Scare stories about impending disaster were a similar part of the mix and they came from the same corners.  Somehow Mike and his clan can easily create super-node clusters to handle 'visa level' transaction rates or whatever, but they cannot see a possible way to deal with UTXO's cramping a memory cache.  Funny that.

So Mr. Todd just added his comment on this:


https://www.reddit.com/r/Bitcoin/comments/39po7q/blocksize_youtube_animation_favoring_1mb_blocks/cs5cv1x

And there's maybe going to be a continuation/update of this video.

https://www.reddit.com/r/Bitcoin/comments/39po7q/blocksize_youtube_animation_favoring_1mb_blocks/cs5hzfu

What scientific analysis have you done to reach this conclusion?  Are you just pulling numbers out of your @$$ because you like the sound of them?  Do you really want to risk destroying the balance between incentive and cost on your "gut feel"?  How much do you care if your "guess" is correct?  If you're wrong and bitcoin is destroyed because of it, do you really care?

I really had the impression that the 20Mb shit, XT or Asshole-client, call it what you want, was going to be an official client which you can download at bitcoin.org. Is it really some rouge project? how the hell can a rouge project get so much hype?