should dataset be baked into provider?

[goastler]

Essentially, I believe dataset should be baked into provider for a couple of reasons:

• a provider should never be able to run more than 1 dataset from their provider entity, as this would spread the stake
• another provider can be registered for another dataset, etc
• it's easier to handle
• we have no functionality / plans to have a 1:many mapping for provider to dataset
• the provider could mash multiple datasets into one dataset, thus rendering the 1:many dataset functionality obsolete

[goastler]

after some discussion on slack, yes because of the above reasons and the only time we'd need to look at historic dataset id's would be in an audit setting with an archive node

[forgetso]

Yes