All of those "proof of work" type schemes eventually will dwindle down to the same centralized network they are aiming to replace.
With the amount of compute power required to perform any operation mining will end up being handled by a couple of big players which the rest will use a proxy.
BitCoin already is pretty much centralized with mining becoming an operation which requires you to setup a bloody data center to get into the game, yes allot of those are some weird complexes in the middle of no-where China, but they are still bloody huge and expensive.
And there isn't much you can do about it, if you build a network which establishes "trust" based on "effort" the effort cannot be trivial, or to be more exact the amount of trust one will have in such network is tied directly to the amount of effort one has to put into it to perform a given task.
> All of those "proof of work" type schemes eventually will dwindle down to the same centralized network they are aiming to replace.
And a payment processor in Bitcoin is closer to a company that runs cash registers than a provider like VISA. The worry is about the companies that want to control both ends, like Coinbase, not payment processors in general that simply take care of accepting BTC on behalf of the payee.
With ever increasing mining complexity coupled with ever growing demand for more and more mining since the rate of transactions in the system is directly tied to its mining rate when the block rewards will become rarer and rare till they eventually halt you'll pretty much will end up only with the "Banks" doing or controlling the mining since they'll have to ensure at least a minimal mining rate which covers their transaction volume.
How so? That's not my understanding at all - you could have a rate of transaction at the level of VISA with just a dozen computers or so. The mining difficulty is a consequence of having many miners (which makes the software auto-increase the difficulty of the problem), not of the increased transaction rate.
the block rewards will become rarer and rare till they eventually halt
That's why there are transaction fees, which pay the miners in lieu of the block rewards.
Ever increasing difficulty raises the bar so high that it centralizes a system due to the hardware requirements (there is no way to effectively mine BC today on standard computing hardware sorry) and their costs.
Look at the BC network now it's either huge centrally controlled mining pools or multi-million dollar mining farms which are controlled by individuals, this isn't decentralization of anything but computing assets as far the the pools are concerned.
For example, in 2014 a pool named GHash.io was close to obtaining 50% hashpower. It is customary for pools to stop accepting new signups when they are close to 50%, however GHash.io refused to do so. The miners organised an exodus and GHash lost half of their miners in a month.
How do you really know all those pool operators are actually different people?
For a 'trustless' system, it seems like you need to place an awful lot of to trust in this small, secretive cabal of mining operators.
At the moment (some) miners are motivated to avoid 51% attack because they gain more from protecting Bitcoin's value than they do from profiteering with a 51% attack.
If that stops being the case, even for a moment, then the 51% attack will happen.
http://www.coindesk.com/eba-51-attack-remains-bitcoins-bigge... [2 weeks ago]
> The core of Ethereum will be released under the most liberal of licences.
Then why aren't they considering MIT? That's about as liberal as you can get.
> it will come with an amendment allowing it to be linked to be statically linked to software for which source code is not available.
NO NO NO NO STOP NO. Do not attempt to modify open source licenses without consulting with a lawyer. It seems innocent enough - but it could cause problems down the road. There is also some rules that say you cannot call it GPL anymore if you modify it [1].
"In this way, while we have not arrived at a final licence, we expect to select one of the MIT licence, the MPL licence or the LGPL licence."
They are definitely considering the MIT licence. Did they do a ninja edit?
I definitely agree with you about the amendment thing.
Perhaps my head was spinning from the vagueness of their license decision and I just overlooked it.
My reason is not that profiteering is bad, but in this particular case, I just don't see how you can compromise anonymity with copyright and a business around it.
For this to be truly anonymous, I just don't see how a business can be attached but cause at the end of the day we are still coupled to the original writers.
What's I'd love to see is something like ethereum but without a business or copyright attached to it.
I hope this makes sense, to truly build a decentralized and anonymous networ, can you try to also build a profitable business on top of it? Maybe I've misunderstood.
I'd love to start working with it, however, I realize how his is still have he early day. The future is a completely decentralized, anonymous p2p network that has no overwhelming force from a single agent to censor or corrupt information.
Imagine if facebook, Google, all of it could run on our phones, tsblets, desktop. You can't shut it down. By the n I assume we will have some sort of mesh networking where we don't even need to go through an isp, just a cluster of trusted and self healing network nodes.
Anonymity and decentralization make advertising-based revenue difficult, true, but there are lots of other open source business models; consulting for example.
However, in this case they're clear: they'll be selling etherium. That's completely compatible with even the most strict definition of Open Source (or Free Software, for that matter).
You can also find companies that don't. I am under the impression that nprobe source code is under the GPL [1] but the author sends DMCA requests to those who want to use the source code under the terms of the GPL [2] [3]. Then there was the company who created the entourage edge (an Android powered device) - for the longest time they refused to release the modified kernel source....they eventually did then they went under.
[1] http://linux.exosec.net/mirrors/www.ntop.org/nProbe.html
[2] https://github.com/github/dmca/commit/859890b8aaf4db678dce96...
[3] https://github.com/github/dmca/commit/85c03a289fe56c67636ed6...