IMO, a decentralized cryptocurrency can not be a security, since the issuer
is the blockchain system itself. Its direction and control is absolute. If you
would argue that ETH is no longer a decentralized system, as evident by
their ability to undo the DAO failure incident (violation of blockchain purpose),
and thus transforms into a centralized system and device, than not only is
their network in a state of "unrealized capturability", but it could be said to
be an "illegal security". The issuer transfered from the blockchain, to humans.
Absolutely! Right to the spot. And changing the way how ETH is being made (from POW to POS) is also the decision that made by humans, not the blockchain itself.
The more ETH is being "human manipulated" it will also drive ETH closer and closer to "(illegal)security" status. And such tendency will surely affect the whole ICO market (and I'm not talking about lot more other related things).
I wish to elaborate.
There are three important forms of Consensus that exists in properly
functioning cryptocurrencies.
The first, is obviously "Miner Consensus", which is what most people
call "block building" and is how the single chain is formed and it's
security is ensured. This is outlined in the Bitcoin whitepaper and
what solved the Byzantine Generals problem. This is an active chain
form of Consensus.
The second, is "Validator Consensus" or "Network Consensus". This
type includes everyone who is maintaining the blockchain on their
own independent and diverse systems around the world. Whether
an incentivized business or an altruistic individual, they both monitor
the network and the miner's block work, and can alert, and ultimately
prevent actions that violate the rules we all agree to follow on this
chain. (This type also contains Miners who are "honest". All honest
miners should distrust other miners by their nature, thus watching
and verifying other miners actions.) This is a passive chain form of
Consensus, with some outside active ability during violations.
The third, is "Community Consensus", which does not have a direct
mechanism within the system (currently), but is intended to slowly
gather support of ideas on changes to the single chain, for a future
date. This type includes all other participants that do not fall within
the first and second type of Consensus. This is what binds us all
together as one and bring us the security that ensures overall
ecosystem and network survival. This is an active user form of
Consensus which can determine active and passive chain
Consensus types for the benefit of all.
In the third, some ideas are easy to implement since they do not
actually change any major systems, but just add to them. On the
other hand, there are some ideas that are hard, because they directly
impact certain systems that could change the current known dynamic
which is currently stable. The "blocksize debate" was controversial
and thus didn't receive high community consensus because some
members of the community would be significantly disenfranchised.
(Normally, to get others to join your position, you would help build
up their systems so that they are made secure, and thus will join
you in the near future. But that is not what has occurred since they
do not want to perform the heavy work and create, but take the
easy path and conquer. But, wide is the road and narrow is the gate.)
So, my point for writing all this is to address your comment:
"And changing the way how ETH is being made (from POW to POS)
is also the decision that made by humans, not the blockchain itself. "Your comment in the context of my statements above, ignores
Consensus. IMO, if ETH goes PoS through Bitcoin's third type of
Consensus, the blockchain system will have still issued the token.
IMO, the only time that the blockchain does not issue the token is
when the system can be directly or indirectly steered by an entity,
like the ETH developers with the DAO incident, or the token is
created by a known entity, like a legal corporation making an ICO.
But, if the ETH devs want everyone to move to PoS, because "that was
the plan", it could be argued that since a plan exists prior to the
creation of the Ethereum platform, it can not be a real cryptocurrency
since the whole system is actually directed and thus controlled. If they
allow their community consensus to deny the change to PoS, you could
say they are a cryptocurrency, but if the devs overrule that "community
contention", that new chain is a security.
So, in this context, you may be correct in that ETH was a security before
it was ever built, because it violates multiple forms of Consensus, since
their developers have the ability to steer the Miner Consensus, the
Network Consensus, and the Community Consensus.
This is interesting and I need to think about this more in depth later.
I believe I agree with you now that ETH is more of a security than a
properly functioning cryptocurrency.
Edit: After some thought, it might be that all cryptocurrencies are just
illegal securities and they each are in their own state of capturability.
So, when considering something to be a security, it might really be
"If a government took enforcement measures against a certain
cryptocurrency now, what is the probability of their success?". Thus,
the labels we give within our ecosystem is irrelevant and illusions.
We may be all security sinners in the eyes of our governments. Some
will be judge and other will not, all depending on our ability to remain
decentralized and anticipable. In this viewpoint, the three types of
Consensus are barriers to prevent or delay that required judgment.