First of all take everything I say with a grain of salt. Ethereums DAO-bailout hardfork, was in my opinion the worst thing that happened in the history of the cryptocurrency world. I have no doubt that my disgust influences my view on it. That being said, we can continue with regular programming.

I was actually referring to even bitcoin having way less of a scaling problem than ethereum, despite just about all public statements from apparent ethereum proponents claiming the opposite (which in my view is more than a little disingenuous). I do believe bitcoins current roadmap looks to be in way better shape than ethereum, as in having an as small and efficient layer 1 as possible and then do layer 2 stuff on top of it like lighting (which is looking to be a more decentralized and anonymous payment network than ripple but without the IOU crap).

While ethereum might be able to also add a lighting layer, it suffers from it's layer 1 being way more inefficient and bloated in comparison. If smart contracts can be added on layer 2 similar to payment channels like lighting, then that would make ethereums layer 1 even more of a waste and exercise in futility. When the expert advise is for buying dedicated data centre level hardware as the way to keep your local node synced, you'd expect to see more people to start realizing that scaling on ethereum is a bigger issue then for btc with it's raspberry pi full nodes on spinning rust keeping up just fine.

Now that's just comparing ethereum to bitcoin. The next problematic design decision of ethereum is how limited it's smart contracts are. As Larimer has pointed out several times, things like bitshares' (and therefore steemits) smart contracts responsible for order matching are not possible on ethereum. The options of contracts interacting with other contracts are very limited. Which appallingly enough is conveniently disregarded by just about all the pundits hailing the world computer, but kinda makes it even less impressive than it already is.

Btw to put ethereums "world computer" in to perspective, people seem to forget that all the nodes are not adding any cpu processing power to the network they are just repeating it. They are all running the exact same operations. If someone would take the time to work out a comparison of the actual processing operations the network is capable of with a modern cpu/gpu you'd start to get the picture of just how disappointingly inefficient it really is.

What I appreciate about the design of bts (and as a result it's derivatives like steemit and eos) is that it always acknowledged that blockchains are just about the most inefficient way to do things, so the idea is to only do on the blockchain what you cannot do any other way. It also accepted that centralization happens and built in mechanics to not only deal with it, but to use it to increase efficiency but also to make it easier for the network users to keep track of it and tools to deal with it (limited number of witnesses and inbuilt voting mechanics).

Ethereum started out in complete opposition to bts take on pos (maybe also because Larimer publicly calling out Vitalik about ethereums scaling issues which are now very obvious) and lauding pow. While they now seem to have come around somewhat, their attemts feel weirdly half-assed to me. Admittedly I can't help looking at their move towards POS in the shadow of the DAO-sceme and its bailout. So I keep seeing it as another attempt at a powergrab or consolidation of their control.

I hope to be rid of the effect of the Rothschilds, Saouds and Vatican cum suis. Which is why I'm more than a little skeptical about ethereum. Sorry to say it the same reason I'm not without reservations about EOS. Block.one having so much influence is bothering me more than a little, but I can't argue with the design of and concepts behind EOS making far more sense to me than ethereum ever did.