rep by get_account_reputations() and rep in the post info by e.g., get_discussions_by_blog() are inconsistent.
Initially I thought this may be due to the sync delay of hivemind, but interestingly sometimes hivemind rep was higher (i.e., lower makes sense if it's due to the sync delay).
I agree that log_rep is useful for the front-end, but all front-ends already expect the raw reputation, so for backward compatibility, the raw reputation should be provided anyway. This extra conversion (which cannot be exact) seems unnecessary to me.
Is there any reason behind this design? e.g., dapps already using log_rep from hivemind?
How can I access log_rep directly? The existing APIs seem returning raw rep only.
In any case, I think it's better to have the extra field of log_rep in addition to the raw rep.
Initially I thought this may be due to the sync delay of hivemind, but interestingly sometimes hivemind rep was higher (i.e., lower makes sense if it's due to the sync delay).
Finally, I realized that it's due to the conversion: https://github.com/steemit/hivemind/blob/dcb177937578b59bd6471e654654532e58adc170/hive/server/condenser_api/cursor.py#L109
That is, hivemind seems not storing the raw reputation value at all in account table: https://github.com/steemit/hivemind/blob/f7a467921678d928a0d94928c811442b8ab80bce/hive/db/schema.py#L40
I agree that log_rep is useful for the front-end, but all front-ends already expect the raw reputation, so for backward compatibility, the raw reputation should be provided anyway. This extra conversion (which cannot be exact) seems unnecessary to me.
In any case, I think it's better to have the extra field of log_rep in addition to the raw rep.