Tom Morris

> OpenRefine extensions can be styled however they wish by the extension author While technically true, I'd argue that there's a strong case to be made for extensions to conform...

I wonder if there are any other data dependent weaknesses in our `Change` serialization. We may want to do a general review to see if we need a more global...

Was this done with the help of an IDE's refactoring tools? Trying to figure out how much attention I need to pay to the voluminous changes...

A general menu description facility / language would both allow us to offer different layouts (e.g. "classic" & "new and improved") as well as allow the user to to customize/extend...

> If you oppose it I can try to work without. Not opposed, just trying to understand it. I am, however, very reluctant to introduce new APIs using the Properties...

Sorry for the delay in replying. This is probably a conversation that's important enough to be worth having outside of the context of a single PR review. I think that...

I didn't look in detail, but I thought this could be done in a compatible way, which was why I asked: > Assuming that there will be additional methods at...

> The usual approach row.record.cells[columnName].value works only if columns are fully aligned. That sounds like a bug -- or at least a highly unexpected behavior. OpenRefine optimizes memory usage by...

I fixed #3290 recently, but the problematic behavior with `row.record.cells` eliding null cells still exists. If I change the example above to: ```json { "key": 123, "foo": [ 4, 5...

This was likely broken by the attempt to fix #6772