Bill Katz

This is related to issue #117 and should be used to note divergence of denormalizations if one type of denormalization has an error.

Also needs to support restriction to ROIs. Plan is to have one labelindex data instance per ROI rather than have on data instance that can quickly return top N given...

Partially implemented via commit bb863eb1f6 that supports annotation counting. Ranking by number of voxels will be added by October.

Very basic authentication was added recently and for the upcoming v1.0 dvid that will be expanded to include authorization. When that happens, we'll be able to add endpoints for deletion....

@jingpengw This issue is about a repo-level request deleting a data instance, and that may involve multiple types of data associated with that single data instance. In your case, you...

Assume this isn't urgent since this is for `/split` which is fairly unused vs `/cleave` and `/split-supervoxel`

Do you think this should be a multipart MIME type like how file uploads are handled, or an ad-hoc binary scheme, something like the following? ``` ... ... ```

@stuarteberg In the case of a kafka queue overload, there could be a situation where a `split-complete` message is successful but the actual `split` is not. Given that we have...

The above was based on incorrect view of events missing from kafka log not from the internal representation. Just to verify, the only missing supervoxel splits were from the broad...

On the other hand, this would slow down all ingestion and mutation of labels since we have to check each label voxel for possible error.