Bitcoin Forum

Hardfork to make it bloat more! Centralization, you unable to store the complete blockchain. 20MB blocks. Hardfork without consensus on it. Risk a split of the network. What's going to happen next?

https://bitcointalk.org/index.php?topic=919629.0

Oh, stop. Read Gavin's actual post: http://gavintech.blogspot.com/2015/01/looking-before-scaling-up-leap.html

In the comments he notes that current thinking is to propose that blocksize doubles every 2 years for the next 20 years. Moore's law (compute power) and Nielsen's law (bandwidth) should outrun it, so cheap/home machines in the future will be MORE able to process bitcoin blocks than they are now.

Also, Satoshi intended for the blocksize to be raised, for those of you who consider the ethos of the early days to be inviolate:

Yes, remain calm, don't listen to all the FUD.

Ohhhh no.. Bitcoin is gonna be worthless.

I gotta run and sell all my coins now!

Bye!

 ;D

Thanks for letting me know it was Satoshi's idea and explaining it.

May I remind (or teach) you that as long as there aren't more transactions trying to get into the blockchain than it can accommodate, the size of the blockchain will NOT be affected by the max. block size? i.e. it's the total number of tx that matters, and not how they are spread across the blocks?

Did you understand that 20MB is a maximum and not a nominal size for each block?


So what exactly is Gavin trying to do if not trying to reach consensus with his blog post?

Did you read it? Did you fully understand it?

i think it's a pretty bad move to start such a discussion in this current market. Hardfork bitcoin? For what again?

That's a mighty big assumption considering moores law is running into issues now.
Than the blocksize should have been coded the way he intended originally instead of what we have now.
hard fork for just this; is stupid.

Gavin started the conversation months ago.

https://blog.bitcoinfoundation.org/a-scalability-roadmap/

Moore's law has been "running into issues" for decades. Also, the more relevant "law" is Nielsen's.



I wasn't around at the time, but my understanding is that the 1MB limit was put in place as a quick-fix to thwart DOS issues at the time. As should be clear from Satoshi's comment above, the intent was to up to the limit in not too long. So it "should've" been fixed in early 2011.



No, fixing this is critical. This is native money for the internet. 7 TPS is unacceptable, and bitcoin runs the risk of being relegated to toy status longterm if it can't scale. If there's no strong technical reason not to allow the scale-up (and there isn't), we should do it, and do it soon.

trollogic:

WAH! bitcoin is fail because block size too small!

..time passes, block size increases...

WAH! bitcoin is fail because block size too big!

No.  Will fail because seven transactions per second is laughable for a real currency.  Will fail because the alternative is hardfork & blockchain bloat.
Why does such trivial shit need to be explained?

Doesn't surprise me, as based on historical transaction sizes Bitcoin can't even handle 5 transactions/sec.

The dumb part about it, after over a year of time from where the issue became apparent, this is the laziest and most unimaginative solution possible. If there is a hardfork anyway there could be done something smarter about it, like that variable block size proposal, or perhaps addressing the cause.
Every node storing every transaction forever is incredibly wasteful. Partial storage with parity would be one approach, colluding smaller transactions into larger one would be another, different kind of nodes or even changing the way the Blockchain stores transactions are all directions to research.

But that would require the will to innovate from the Bitcoin development team, not changing a few hardcoded values.

Because cognitive dissonance is a bitch.

That's stupid idea, it's almost impossible to do an hard fork that will make 51%+ move to it.

These are the assumptions some people have problems with.

The thing is, people have been predicting the end of Moore's law since at least the mid-1980s, always "within 3-5 years". Eventually they will be right, but there is a long way to go.

There were similar predictions about the max speed a modem could do over POTS, until people got over 300 baud, then 9600...

By the way That Mircea Popescu dude is a schizo clown.

A greedy sociopath who is gonna see his holdings in BTCeanie BTCabies bitcoin suddenly vanish when this thing finally dies.

Considering we are talking about bitcoin, I'd say it's appropriate.

It's a circus.

Yes the cognitive dissonance of being sure that you are right that nothing will usurp your real money. That you worked so hard for, that you wholly believe to be the one true money. The absolutely overwhelming terror of potentially being wrong. Continuing to see daily evidence that shows everything you believe in may be a lie. Desperately seeking ways to discredit the alternatives. Cognitive dissonance is indeed a bitch.

The real truth is it could go either way. I am comfortable with whatever the outcome. I'm not wed to bitcoins colossal success or its utter failure, or anything in between. You both seem to be trapped in a corner of your own making. Fortunately for both of you, you have the chance to escape! All you need to do is not change any of your real money for bitcoin and to assure your continued happiness, delete your accounts and never visit this forum again.

If you carry on hanging round here, reminding yourself of the alternative, you'll just make yourself more and more depressed. Do yourselves a favour end the suffering.

DDes are increasing the block size upper limit, not the minimum size. Pools can continue mining 200k blocks as they do now. IMO, it is not a major concern.

More like egomaniac.  He's using this issue to try and draw attention to himself.  The block limits were put there to prevent spam as a short term bandaid.  There's obviously a problem with making block sizes too large, but it would also require an enormous Bitcoin economy to make it a problem, or...lots of people spamming the chain with gambling site nonsense.  As long as you can prevent spam, increasing the upper limit isn't that big of a deal.

I guess it comes down to an issue of, would you rather have fractional reserve bitcoin where everything is off-chain, or non-fractional reserve where everything is on-chain with more bloat.  Some people might ask, if you're doing everything off-chain, then how is it beneficial over the systems we already have?  There's clearly some point there since the Keynesians could still be running wild in an off-chain system.

Well you can still use light wallets that don't store whole blockchain.

https://bitcointalk.org/index.php?topic=271711.0

 ;D

Ah, I see pirateat40 in there.


Bitcoin: getting the worst scumbags on the planet together since 2010.

calm down it's for testing only, and only the longest blockchain count, this is why a strong miner support is good , otherwise anyone can just run his own chain and fork bitcoin

Bullish news for Litecoin  8)

Just put a post on reddit and gets your mates to thumbs it up, pretty easy ;D

Litecoin suffers from the same principle, the increased block rate makes it an issue somewhat later but it doesn't address the root of the problem.

If you give a notice of a year, then it is possible.

As usual, EM is both right and sort of nasty about it at the same time :D

Playing devils advocate its a less complex change, easier to test, and more YAGNI.

Having said that I agree with EM. Necessity is the mother of invention though, and right now bloat is a red herring imho, disks are big enough space isn't an issue (for some time yet). A quick change to buy more time whilst the more creative solutions are worked through doesn't strike me as unreasonable.

Think about fossil fuel, government debt & QE, meat farming. All totally unsustainable, all just carry on.

Its human nature to kick the can down the road. Why would BTC be any different, especially when thats how everything has always worked since forever. Forecasting that bigger blocks=> unsustainable bloat => 'the end of BTC' is the same mindset the peak oil folks, the stackers and the doomsday preppers.

I'm not saying any of these people are wrong, I'm saying that pragmatism generally wins out.

20MB blocks maxed out is roughly a terabyte a year. 3TB drives are already down to $100. 6TB is top end 8-10TB are already in the pipeline. its going to be several years at least before blockchain size outstrips storage capacities. So lets put this 'bloat' myth to bed.

who the hell wants to dedicate an entire hard drive to a blockchain? Oh wait who is rational in the digital currency world.

more mindfuck and buttrape, that's all.

But since we are heading to 50cents-prices we shouldn't care that much about it. Let Gavin play in his sandbox.

Let's be rational indeed.

If Average Joe ever uses a bitcoin wallet, it will be SPV at best, possibly just a client connected to his bank, using some yet-to-be defined upper layers on top of it (or something Ripple-ish) to make the UX smooth and 'secure'.

The point being, Average Joe won't store the blockchain at all. Other big players upstream (something Google-ish) will provide the service and find a way to monetize them.

Sources: my own experience of the internet evolving from 1994 to 2015 ::)

So are we trying to centralize bitcoin than to just a few big players? or is this supposed to be
a currency away from the major banks. Because the big players who run a 6tb blockchain will be the new banks.
as an avg joe. I don't trust any of the wallets yet. I prefer core/wallet currently and 30gb block chain is stupid now.
Great we both have the same experience big woop :P
except I cut my teeth on a trs-80 so I guess I could count the 80's too

Who's "we"?

You want to run a node? go right ahead. As I said, for the foreseeable future - even with a bigger block size - it only takes commodity hardware.

Mundane truth vs hyperbolic speculation. On BTC the latter seems to count more than the former :/

we=community "face palm" ppl act all fucking smart on here than play stupid with their responses.

I know what you mean, but I had to learn the hard way to dissociate between my ideals and reality.

The internet was supposed to be peer-to-peer, and for a long time, I was putting the blame on NAT, providers, IPv4 address space depletion, failure to embrace IPv6 correctly etc. That was before I realized that Average Joe didn't care all that much about the peer-to-peer internet. Same goes for net neutrality.

I'm still using my own mail server on my own domain name, and for many years I was maintaining a dedicated network infrastructure for the company I was working for. Then came the cloud computing era, and it became increasingly difficult to justify maintaining our own mail servers, the even the web servers were virtualized into some cloud services, etc. Eventually, as a sysadmin, I was fired because I wasn't useful anymore, all the services that I did provide internally were redundant with Google, Amazon, Rackspace etc.

So be it, I've learnt it the hard way. You can have a dream, ideals, and tons of reason to believe in it, but the reality check is coming sooner and later to sweep it right in front of your eyes. Better learn to adapt and foresee, that was my lesson.

What would be stupid is to think that what *you* think is entirely representative of what everyone else thinks.

I run a node. It costs me virtually nothing. Adding more storage isn't going to change that materially.

I'm not being smart. I don't need to be smart, this is *really simple*.

sanity. so refreshing!

The original protocol had no limit on block size.   The 1MB limit was simply added as a temporary measure to prevent denial of service attack.  At the time the average block was <50KB so it was considered a pretty huge limit.   It was never intended to be permanent until the end of time.

Once again mundane truth scoffs in the face of speculative hyperbole!

agree seems the stability of the market plus tings of this nature are hurting the overall view of BTC lately

In my earlier post, I talked about how there was an issue of small block sizes leading to an off-chain, possibly fiat or fractional reserve Bitcoin environment.  Now that I've thought about the issues more, I'm not quite sure raising the TPS to Gavin's intended levels does anything to solve that.  I would like to hear other people's thoughts on this aspect of block sizes.  If his "fix" does nothing to solve this possible problem in the future of Bitcoin, why even bother raising it if we're going into an off-chain environment anyway?

The only purpose in raising it would be if he forsees a failure of block reward subsidy to secure the network.  If the marginal TPS upgrade isn't enough to stop an off-chain environment, the block reward subsidy problem will still exist unless Bitcoin is run as a non-profit utility by some random organizations, payment processors, or governments.  Overall, it's a very complex scenario that requires people to make assumptions about future transaction levels, both on and off-chain...

Las time we started to butt up against the block size limit, it only required a 'soft' fork to change the block size. If we start to butt up against it this time it takes a 'hard' fork to change the block size. (my understanding at least).

As it requires a hard fork, its probably wise to get that sorted *before* we start butting up against the limit.

I'd agree it only buys time though. Better to have more time than less.

Testing with 20mb blocks might not mean that we go straight to 20mb blocks, it might be that the hard fork implements the 20mb as the hard limit, but we only go to 2 or 5 meg blocks with a soft cap (like it used to be before 1mb blocks). Just speculatin'

I don't know if the outcome will be net good or bad, but I think the intention (buy time) is good.

Is something still a 'hard' fork when 99% agree just to follow the proposal. It inherently becomes the 'real' chain, anyways. I think the difference between smoothly done integrations of new things and amendments and a hard fork is very small.