Dan Allan

TIFF sequences are now supported out of the box, and all the limitations noted above were addressed in #511

Drive-by comment to say: * I saw @jacobtomlinson's SciPy Lightning on this project and finally got around to using it. It's great! * As you can imagine, I found this...

Thanks @t20100. I agree that the blocking I/O sounds like the hard part. We may have to live with a synchronous I/O for now and just make sure that timeouts...

I have been pondering the trade offs of making `msg.obj` a list (in some cases) or packing additional objects into `msg.args`. I think this approach is the less messy option...

Capturing notes from a chat this morning, remaining steps on this PR are: - [x] Multiple descriptors - [x] Refactor HDF5-specifics from TiledWriter to "Handlers" - [ ] Deal with...

Per discussion with @maffettone and @genematx: - The "handlers" in this PR are not doing I/O. - It seems worth exploring reusing the Adapters in `tiled.adapters` to do I/O, such...

This diagram shows how this new component (marked as "???") fits in to everything else:  From brainstorm with DLS on what to call this: 1. Compactor 2. Condenser 3....

To belatedly respond to @prjemian, I really like the word "Collector" for a callback that collects documents. This particular thing we are trying to name does something more specific. I...

Yes, using real RE with mock devices to provide data is the route we have taken up to this point. It would be interesting to learn more about the DLS...

Got it, thanks for expanding on that. I think the terminology ”mock RE" may be the only thing we are tripping on here. Having purpose-built functions that consume messages is...