Re: BYTEBALL: Totally new consensus algorithm + private untraceable payments
Well, it's hard not to see DAG in Byteball design.
It's also seen in Ethereum design, any idea why they call themselves "chain", not "DAG"?
Ask them. Perhaps they started from a chain and allowed for a DAG in some cases to accommodate for some need.
- GHOST protocol gives a similar topology of blocks and if we have only 1 transaction per block then there is no difference in first order approximation.
And what? Of course, there are edge cases. And I didn't invent DAG if you want me to admit that 
- The whitepaper tells about ordering and the main chain.
And?
I just mentioned things which point toward "chain", not "DAG".
They are not mutually exclusive.
If you want to make some point, make it.
- There is no publicly available information on transactions topology in high TPS mode (this would show if Byteball utilizes high scalability granted by a DAG).
Any reasons to believe the topology will be in some way "wrong" in high TPS mode?
Yes. I suspect the ordering will slow everything down significantly. Is it hard to run a test?
It might slow down or might not. What I know for sure, the current implementation leaves a lot of room for optimization. But it is not my primary concern at the moment, all these TPS should be useful for something, and I pay most attention to the demand side rather than the supply.