# Chat logs about sig format

Kyle Bryant, [20.09.18 06:47]
[In reply to Kyle Bryant]
Well, not just on board, but to actually use the dapp without needing ether too

George Spasov, [20.09.18 06:49]
Which standard do you use for your contract?

Kyle Bryant, [20.09.18 06:54]
Not sure what standards there are, we were plugging V, R, and S into every function and had a rudimentary relayer, until I went to ETHBerlin and discovered @austingriffith contracts, now we're tearing out signatures from our contracts methods and building on top of EthWorks Identity contracts which does a delegate execution call

George Spasov, [20.09.18 06:56]
I seem to have POC on what you did, maybe you can reuse it

George Spasov, [20.09.18 06:56]
https://github.com/LimeChain/IdentityProxy

Kyle Bryant, [20.09.18 06:57]
https://github.com/BoomerangProject/boomerang-universal-login/blob/master/relayer/contracts/KeyHolder.sol

Sina Mahmoodi, [20.09.18 10:15]
How about we try to find differences between projects (from low-level stuff like wire format, to higher-level like rewards), come up with arguments for each approach, and try to see if most projects can agree on an approach and try to standardize as much as possible.

To start this, I created an issue for one of the differences I could see at first glance among projects, which is how they handle signatures. How do you do signatures, and why?

https://github.com/austintgriffith/meta-transaction-format-share/issues/6

Austin Griffith, [20.09.18 16:11]
So the hot question of the day is... do you split your v, r, s up on-chain or off? 🙂

Piotr, [20.09.18 16:11]
[In reply to Austin Griffith]
On-chain

Kyle Bryant, [20.09.18 16:12]
We were doing it off-chain because we assumed the gas cost would be lower

Radek Ostrowski, [20.09.18 16:12]
yeah, same here

Piotr, [20.09.18 16:13]
Sure. Off-chain is cheaper. But if you want to make it a open marketplace then you need to have p2p consensus.

Austin Griffith, [20.09.18 16:13]
It is slighthly lower off-chain but you trade that for # of arguments and vars on the stack I think. Would love to hear more discussion. Posting on the github now:

Austin Griffith, [20.09.18 16:13]
https://github.com/austintgriffith/meta-transaction-format-share/issues/6

Austin Griffith, [20.09.18 16:15]
Piort, I don't think I understand the argument. what does the p2p consensus have to do with the signature format? Time for Austin to get educated.

Piotr, [20.09.18 16:16]
Hold on. It doesn’t have anything to do :) I think I answered 2 questions in 1

Austin Griffith, [20.09.18 16:17]
aaaaah rad

Piotr, [20.09.18 16:17]
p2p consensus is needed if you want to pick who is going to deploy the transaction (sig) on the network