Matt Thompson

I'm working on a set of regression tests in [another repo](https://github.com/mattwthompson/interchange-regression-testing), which is a part of the Interchange adoption [plan](https://openforcefield.atlassian.net/wiki/spaces/IN/pages/2225111061/Project+Plan%3A+Adoption+of+Interchange+backend+as+export+machinery+for+ForceField.create_openmm_system?atlOrigin=eyJpIjoiMTJhY2RkNjA0ODBhNGI4ZWI2ZmM3ZjMzNzgzMDZhNjEiLCJwIjoiY29uZmx1ZW5jZS1jaGF0cy1pbnQifQ). Despite the naming, it's toolkit-centric and meant to test results...

It might be useful to put some bounds on what behavior we really must avoid regressions with, or at least split things up. If it's a matter of the charges/conformers/etc....

Closing as we're moving away from the ParmEd route.

I've also been bit by this - not resulting in broken production, but resulting in me looking goofy passing mapped SMILES to `Molecule.from_smiles`. I didn't know `from_mapped_smiles` exists and wouldn't...

I'm pretty confused by this. I can't argue that it makes sense to set velocities before letting OpenMM do the virtual site math - but I also don't think it's...

This is a blocker for some large-scale testing I - and probably others - would like to do; library charge assignment being significant slower than AM1BCC (https://github.com/openforcefield/openff-toolkit/pull/954#issuecomment-857068752) largely defeats the...

Thinking out loud, and I'm probably wrong - the original goal here is to make `LibraryChargeHandler.find_matches` more performant when passed a parameter generated from `LibraryChargeType.from_molecule()` more performant. Is there any...

On the most recent commit of #1036 (8af8f5ecb1020aed76a20ea67267ac2a493072bd ), I can get the first example to run with OpenEye in about 13 seconds, but it hangs with RDKit

Summary of #1036 is that `unique` was added and turned on by default for `LibraryChargeHandler.find_matches` (only) and no `match_heavy_first` was added to the public API or publicly-accessible behavior. It would...

This is failing because my patch did not fully fix the issue; I'll fix it over there and we'll have to wait for another release. Presumably a version 0.11.2 of...