Nobody said the growth would stop, what has been said is: hitting the blocklimit will not cause big trouble because the first thing that would happen would be microtransactions for tiny amounts of money happen somewhere else. So all your doom preaching and urgency is invalid.
Hitting the limit will not render btc unusable - in fact it will clean the chain from spammy microtransactions.
That's funny. So here you are entering a store, present your creditcard to pay for your coffee..... But guess what? I'm sorry sir, your order is below 5$, you will have to fill up this special "Starbucks-card" first with at least 50$, since we don't accept 5$ creditcard payments here.
Give me one reason we would not allow micropayments to happen on the blockchain? They're all fee-paying-payments just like the large(r) ones. If you want the blockchain to be supported by transaction fees at one point you're going to need a lot of transactions.
Besides that, why put a hard limit on an open-market-concept? Let the market decide. If miners don't want to mine what in your opinion are "spammy" transactions then they wont. For the rest of us, let us mine what we want to mine, in the end for the miners they are all equal in terms of transaction fees.
It would have space for them for a while but at some stage the network wouldn't have space for them and that's with clunky as fsck fiat payment system volume, that would likely go up a few orders of magnitude with programmable money. By the time micropayments are a common thing more efficient means of carrying the out will have evolved but there's not much point placing the barrier of entry artificially high even then.
Got to say, after reading up on all the reasons against, space and network limitations, miners rewards, propagation times, all the centralisation issues, etc. the most convincing reason I've seen yet to support a hard fork is all the FUD and empty arguments being put forward against it. If they're so eager to flood this thread with the same kind of crap as the speculation subforum there must be something really good about raising the limit.
Let's put a few of your arguments against the fork and line them out;
space
Currently the blockchain is ~34GB. If it will continue following the current ~0.35MB block average per block we'd be increasing the blockchain about ~30GB/year.
Block size average was around 0.17MB in 2013. More or less Moores law. If this keeps increasing at the current rate the cost of storage of the blockchain will remain the same over the years.
network limitations
20MB block / 10 min = 2MB/min = 0.033MB/s, common even on very very basic household connections.
miners rewards
Miners need either;
lots of transactions all paying a little fee
very little amount of transactions paying a large fee
Bitcoin needs prefers to be viable for as many people as possible, so a very little fee per transaction over a large amount of transactions has the preference.
propagation times
There are already techniques out there like Pieter Wuille's “headers first” approach downloading the longest chain of 80-byte block headers, which is only 25 megabytes of data. The headers are sufficient to know whether or not you have the best chain, and once your node has the headers it can “back fill” by requesting complete blocks from multiple peers in any order it wants, similar to how BitTorrent downloads chunks of a large file from many different peers at once. (source: Gavin's blog)
Pieter has also been working hard on ‘libsecp256k1′ — a highly optimized library for performing math on the elliptic curve used to secure Bitcoin transactions. It is undergoing extensive review, and will be rolled out when we’re convinced it is bug-free and completely compatible with the existing, OpenSSL-based code. (source: Gavin's blog)
rdponticelli has a pull request to run Bitcoin Core to run with a “pruned” block database. Once you have downloaded and indexed the full block chain, the only reason to store all of the old transaction data is to serve it to brand new peers who are performing the initial download. (source: Gavin's blog)
all the centralisation issues
Mining centralization? Node centralization? Currently it's already unviable to run a full node on your mobile or your occasionally-connected-to-the-internet-laptop. Full nodes in the future will stay completely possible for the hobbyist, for the non-hobbyist or people that simply don't want to store the full chain light clients will be a totally acceptable alternative.
