6

It seems there is a reasonable amount of overhead (and some risk) for participating in the baking process on the 48 hour test chain during the voting period.

As far as I can tell there is no incentive for a baker to actually participate in the test chain baking outside of simply "it is good for the network", and there is also no downside, so long as a critical number of bakers participate.

I guess I see it as similar to the choice of whether to run the accuser or not, for an individual baker there is little need or incentive.

2

It is a good question. The answer relies on I believe the mistake you made in asserting that there is "no downside" to not participate in the test net.

For example: if a baker has rights just after the protocol updates then if the baker did not prepare his setup to adjust for the new protocol he has a risk of missing his first bakes under the new protocol.

It gives an incentive to "get prepared" and the testnet is the best way to do the transition smoothly

So in conclusion i would say in particular for professional bakers who have public performance metrics and delegators to answer to, that it is important for them to make sure their setup is prepared for the protocol update during the testnet phase

4
  • 1
    Ensuring you have up to date baking software is absolutely necessary for your baking operation, however that does not mean you need to participate in the test chain, by design the baker should flip to the new protocol for you after the successful test. I'll grant the "be prepared" incentive is there, however it seems equally as weak as the "good of the network" incentive considering you run an increased risk of losing rewards by participating in the test chain – xtzbaker Mar 2 '19 at 15:57
  • 1
    @xtzbaker i disagree. If you as a baker wait until the testing phase is finished to change your config you will have operational risk. If you discount this risk as 0 it is your choice of course. But as any risk it has a monetary cost. – Ezy Mar 2 '19 at 16:08
  • The question isn't about preparing your node for the new protocol, a baker can just as easily prepare themselves using zeronet or alphanet, and simply have the appropriate versions of the software running on mainnet prior to the new protocol being adopted. The question is specifically about taking part in the baking process on the test chain. There appears to be very weak incentives for bakers to participate in the baking process on the test chain. – xtzbaker Mar 2 '19 at 20:54
  • @xtzbaker i respectfully disagree one more time: it has everything to do with preparing for the new protocol. zeronet is not guaranteed to be the exact same thing as the mainnet. Testnet is. There is 0 downside in testing in testnet instead of zeronet if the baker is willing to do some testing at all. – Ezy Mar 2 '19 at 20:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.