Well expressed and I agree that indeed it is these core differences that are the "key things" that sets Nxt apart (and therefore where the major development efforts should be focused).

Ah yes - of course - thanks (should have realised this) so each and every (full) node will constantly be doing its own updating of a ledger and getting ready to "wipe out" the old history.

I guess the problem comes though in what do you send a brand new node who has no block chain (or do they need to download one from the last "checkpoint")?

Agreed - so a new "genesis" block would have to be the balance of all accounts from before the last possible "re-org" point.

BTW - said "genesis" block is going to become "huge" in the future (if we had one million accounts then surely we are talking 200-300 MB which you are not very likely to be able to be sending to nodes very quickly - or are we working on some sort of special format to "compress" a genesis block?).

Although it could be divisive it might also be worth "pruning" tiny balance accounts (ones with less than or equal to the minimum fee say) and having those as "fee rewards" for the construction of the new genesis block (this gives some added incentive to create it rather than just to "skip your turn" because of the "work involved").

I guess that the block hash itself could be used as a "random" piece of data if required - am still not convinced of how this is going to work but obviously it is early days yet (looking forward to reading more about the technical details).

Well perhaps there are some examples that the person "running" the script doesn't care - but if it was *my* script being run then I would want to be 100% certain of the state after it was run (before it was run) if at all possible.

I would also be worried more about potential forks if the state could not be predicted in advance.

Which is why you wouldn't be able to "run" said script until is *has* been included in a block (and I don't think it should rely on data in the most recent block either).

I think it is *mandatory* that the state needs to be verified (otherwise how to ensure correct operation of the script?).

Okay - so it *is* being interpreted (and presumably by each node that sees the new block also otherwise it could have been interpreted incorrectly).

I think you'll need a hash of the memory "state" after execution which should probably actually be provided by the person wanting to "execute" the script (presumably through a second tx).

To my way of thinking it would be:

1) Create the "script tx" and pay a fee just based upon its size (am not sure where we are storing its state - perhaps that is going to be in some sort of AM).

2) Send a "run script" tx later with a fee and an "expected state hash" (if the nodes run the script and fail to get the same state hash then it would be an invalid tx).

Or are you thinking along the lines of something more "automated" than this approach?

I guess I am missing something - how then does it fly? Surely it needs to be validated as to what it does by all (or at least the majority) of nodes otherwise how can we know it has been processed correctly?

I don't quite get this part (the other statements all look reasonable) - what exactly do you mean by "pushing the block" vs. "forging" and which nodes are "checking" that the script operation has been performed correctly (and exactly how)?

From what I understood the Account Control (is that the right name?) feature will allow you to allocate your "forging" rights to another account (although I don't think that the specifics have been disclosed yet).

So I don't think you would need to be "logged in" at all.

Why not be a bit "next generation" and have the minimum fee worked out according to tx history (so it could be adjusted up or down automatically based upon the bandwidth consumption)?

Indeed - these seem like pretty good features which is why I question (although have also posted ideas about) the need for things like "Turing complete" contracts (or whatever they should be called).

It appears that this is a bit of a problem for the Nxt community - it doesn't seem to know exactly who it is trying to appeal to (apart from *everyone*).

Understand that *because* of their design our "Turing complete" competitor will *never* be able to do as many TPS (they will probably end up unable to even do as many as Bitcoin).

I don't think we should be trying to compete in that space - personally I think some sort of separate blockchain for such "luxuries" would probably be a better move.

Also as much as I admire Alan Turing's achievements I don't think his name is going to be much in the way of a great marketing angle.

I think CfB had said something along the lines of "voting should not be via forum polls but instead via posts" presumably made into this topic.

Actually - I don't see it (and I knew about that project before most others). Average people don't even know who Alan Turing is and I think would understand something like "can process transactions as fast as VISA can" much better than something that is "Turing complete" (they would probably say "oh - so did this Turing guy forget to complete something and why is it that we need to do that now?").

On the fees - I'd go with divide it by 10 (so 0.1) and see how that goes first before perhaps dividing by 10 again (maybe that won't be necessary for some months).

I think it would be better not to be marketing about 1000+ TPS and CfB seemed to agree that "even 100 TPS would be very good".

Bitcoin can only "theoretically" handle 7 TPS *now* so if Nxt is able to do more than 10x that I think it will be "good enough" for at least a year (and would still put it well in front of the competition).

Maybe it is or maybe it isn't (still not quite clear to me) but do we really *need* to be marketing this angle *now*?

Not disagreeing with the realities about how many TPS are likely to actually be "executed" but it is still not a good idea to be marketing something that can't actually (at this stage) be delivered.

Again - it just doesn't seem very clear to me how we can do all these things - "forge with a solar powered cubie or the like", handle 1000+ TPS and also be able to handle a "Turing complete" transaction language (oh - and yes - be able to do it all on your average "smart phone").

Maybe so - but 300GB of storage?