Mining subprotocol is ambiguous about `job_id` namespace

[plebhash]

the Mining subprotocol spec is ambiguous about the job_id namespace

so the reader might ask: under which context does a job_id have to be unique?

• universally? (e.g.: pool-wide)
• per-client?
• per channel?

[plebhash]

my subjective interpretation is per channel, and on private discussions with @GitGab19 and @TheBlueMatt they stated the same

@Fi3 @jakubtrnka could you provide your input so we can reach consensus and proceed with an update to remove this ambiguity from the spec?

[Fi3]

I agree