With all the Ethereum news around Turing complete languages and NXT also chipping in (https://nxtforum.org/general-discussion/price-speculation/msg71681/ (scroll down a few posts) I wondered what is required to built something similar for Bitcoin/Peercoin and what is already available in the Bitcoin ecosystem.
Relative to Peercoin, I believe that the Turing-complete blockchains will be sidechains which Sunny is working on. This will allow Peercoin’s main blockchain to remain small and capable of high security and functionality far into the future.
Ethereum is pioneering three different languages (LLL, Mutan and Serpent) for advanced intelligent blockchains which can be learned about at forum.ethereum.org. These types of advances I believe will be implemented into future Peercoin sidechains.
[quote=“NewMoneyEra, post:2, topic:2728”]Relative to Peercoin, I believe that the Turing-complete blockchains will be sidechains which Sunny is working on. This will allow Peercoin’s main blockchain to remain small and capable of high security and functionality far into the future.
Ethereum is pioneering three different languages (LLL, Mutan and Serpent) for advanced intelligent blockchains which can be learned about at forum.ethereum.org. These types of advances I believe will be implemented into future Peercoin sidechains.[/quote]
Assuming it is still important to keep the PPC blockchain as small as possible, it makes sense to use sidechains (or even the less secure OT) to implement a Turing complete language.
However when a sidechains gain more importance you will have to sync with them too, increasing you sync time anyway. So for certain applications likely to be used by the majority of users in the future it may make sense to built the functionality straight into the main chain. I think a Turing complete language is one of those to consider building into the main chain as I think this will be a standard integral feature in blockchains in a few years.
The only reasons I see to use a sidechain for, would be if you would want to see competition in solutions (e.g. different Turing complete languages) and standardise on API level or for features the current blockchain is inadequate for (e.g. high transaction volumes) or features of use for a small number of users requiring relatively high resources on main chain (data storage). But maybe we should have that discussion in another thread.
I’ll be surprised if Turing-complete chains become the norm. I expect that huge blockchain bloat and miner processing power consumption will make this undesirable. I could be wrong.
I also see some issues here. Actually I see a lot of issues and not only technical ones.
If all programs in the world run on top of it, the blockchain will be huge. If programs have to compete for processing time, when things get crowded the cost of running programs will be to expensive and for many programs the cost will be to high to justify the profit of running the program. Since the value of etherium is decided by the market, you could even design and start to run a proram that after a while perhaps will not be able to run anymore because the cost of running it gets to high. If enough money is at stake, the network could even be attacked by financially driving up the cost of etherium. Also the more programs running on the chain, the more incentive to attack it, which means that the higher it scales the greater the incentive to crash the network and the bigger the consequences if it fails.
I could go on and on about this, but I won’t.
Oh and no I’ve not studied the design enough to really have an opinion here. I’m just presuming how they have designed it and criticizing my own assumptions. I mean, if I can come up with this kind of criticism without even giving it any thought really, they sure must have too and hence also solved these issues. Right?!
You both have a point. I do think that blockchain bloat and high CPU power will get less of a problem with time assuming Moore’s law still applies and Bitcoin and the likes will end up splitting chains to reduce bloat. This may technically be very similar to a kind of sidechain, but just one which isn’t creating blocks any longer and just exists as a historical ledger. The coin itself lives on a new ledger cryptographically tied to the frozen sidechain. Won’t be Bitcoin taking such a step as the first one, it might be Dogecoin or the likes.
It may take a while before this technology becomes feasible and stable though.
It shouldn’t stop us from making the steps in the right direction in the mean time, that’s why I thought it might be something for Fazon.