Ask HN: What are the cons of DAGs as compared to blockchains?

32 pointsmgalka8y ago14 comments

Crypto currencies based on directed acyclic graphs (e.g. IOTA) seem to solve nearly all of the biggest problems with blockchains -- scalability, low or 0 transaction costs, no 51% attack, no need for miners.

DAGs cannot be used for generalized smart contracts, so they are not a competitor for Ethereum. But do they have any disadvantages as compared to Bitcoin and other pure cryptocurrencies?

I assume I'm missing something, because otherwise DAGs seem too good to be true. And so far there has been little attention paid to them by the crypto ecosystem.

14 comments

tromp8y ago

The main disadvantage is that their growth rate is uncontrollable. If it's possible to add billions of tx to the DAG in a day (possibly with some custom hardware for solving simple PoWs), and a node needs to process those billions of tx to have absolute confidence that some payment it receives is not double spent, then the system breaks down.

dustfinger8y ago

Would a tx rate limiter not help here?

tromp8y ago

How do you limit the rate? I only see the requirement of a low-difficulty PoW, which is easily subverted with some custom hardware.

throwaway30yo8y ago

A problem that the DAGs seem to face is that integration with an exchange is a much more difficult process. At least with XRB (Which I am most familiar with) each exchange has to set up multiple hot wallets and ensure they are kept in sync.

I think XRB is now exchanged on 3-4 sites and they all only have various levels of support.

I think the DAGs (XRB, Byteball, IOTA and likely many more I am unaware of) have the potential to be very useful for microtransactions.

francoisjammin8y ago

I am super interested in the answer to this as well. It seems too good to be true. That is why I have not dipped my toes in RaiBlocks (XRB).

ezekg8y ago

I sent a few test transactions with XRB and they appeared in my wallet within 10 seconds. Confirmed on the network within 1 minute. I read the whitepaper and it seems solid and seems like it will scale much better than other mined coins, not to mention it's more energy efficient because of the lack of mining. Excited to see where it goes, as I'd love to see a coin that is viable for real-world transactions (honestly, I just dislike carrying/receiving cash and dealing with banks to use my money). Looking to check out IOTA next.

quickthrower28y ago

I've been playing with a small amount of IOTA (<$0.01) and found yes txns get stuck waiting for confirmations for days and it is very hard to force them through.

Due to the one-use addresses your entire funds are stuck, because you can't resend from that address, so you can only wait, reattach and 'promote'. Promote means create a 0-value txn confirming your txn that ends up being more likely to confirmed itself. Nice in theory.

vjcucucurg8y ago

I own some byteball, rai, and iota. I like the high tps. But im not sure that no transaction cost is a positive. It makes it easy to spam the network to bring it down. Having some proof of work to guard against this seems inefficient and wasteful Why spend time burning electricity when you can just charge the same amount in transaction charges. A small but real transaction cost seems cleaner.

albertgoeswoof8y ago

I don’t think any DAGs are running in the real world without a central conductor for security. Which is a huge attack vector and downside

dustfinger8y ago

I see no reason why one could not develop a DAG based crypto-currency that is Turing complete to compete with Ethereum.

mgalkaOP8y ago

DAGs do not keep track of the order of transactions, so general purpose smart contracts are not possible

arisAlexis8y ago

byteball

flignats8y ago

They require mass adoption to work as intended.

arisAlexis8y ago

byteball is a DAG system with smart contracts I suggest people to look into it

j / k navigate · click thread line to collapse

14 comments

tromp8y ago

dustfinger8y ago

Would a tx rate limiter not help here?

tromp8y ago

How do you limit the rate? I only see the requirement of a low-difficulty PoW, which is easily subverted with some custom hardware.

throwaway30yo8y ago

I think XRB is now exchanged on 3-4 sites and they all only have various levels of support.

I think the DAGs (XRB, Byteball, IOTA and likely many more I am unaware of) have the potential to be very useful for microtransactions.

francoisjammin8y ago

I am super interested in the answer to this as well. It seems too good to be true. That is why I have not dipped my toes in RaiBlocks (XRB).

ezekg8y ago

quickthrower28y ago

I've been playing with a small amount of IOTA (<$0.01) and found yes txns get stuck waiting for confirmations for days and it is very hard to force them through.

vjcucucurg8y ago

albertgoeswoof8y ago

I don’t think any DAGs are running in the real world without a central conductor for security. Which is a huge attack vector and downside

dustfinger8y ago

I see no reason why one could not develop a DAG based crypto-currency that is Turing complete to compete with Ethereum.

mgalkaOP8y ago

DAGs do not keep track of the order of transactions, so general purpose smart contracts are not possible

arisAlexis8y ago

byteball

flignats8y ago

They require mass adoption to work as intended.

arisAlexis8y ago

byteball is a DAG system with smart contracts I suggest people to look into it

j / k navigate · click thread line to collapse