Jeremy Leibs

@teh-cmc we were talking through this with respect to Transforms I realized this approach has the potential to cause some annoying performance issues for chunks. In particular, every unused optional...

Thanks for the write up! The motivation for `.update()` makes sense as a mechanism of controlling tag-application. The concept of "most-specific at timestamp wins" is already how I was imagining...

Here's a very simple example that makes the viewer totally unuable: ```python import rerun as rr from rerun.datatypes import RotationAxisAngle rr.init("rerun_example_many_entities", spawn=True) def one_strip(number_points: int): return rr.LineStrips3D( [[0.3 if i...

@nchodosh when we tried to do this before we ran into some issues with build environments that were old enough to include the older glibc not having access to a...

We actually made some good progress on this recently. The viewer is no longer linked into the wheel .so. Instead we bundle a separately built rerun executable which we launch...

Just to clarify a bit further, this would make it so that the **inner** values of arrow batches are never null. However the **outer** list-arrays itself would still have nullable...

Note some places in the code indicate we should change our defaults to BGR. I do NOT think we should do this. BGR is a bad, legacy choice from opencv....

Some frameworks have restrictions on what defines a valid column-name. For example: lance doesn't permit top-level field names to contain a ".". This is a good argument for Rerun being...

The introduction of dataframe APIs is likely to place a heavy documentation load that would be very well served by cook-book style documentation. This is particularly notable as likely motivating...

There is a very related problem to this: In many situations we directly send TextLog messages to the `/rerun` path if a user has logged something that threw a caught-and-handled...