Carter Francis

> Pitching in as an extension developer on some points raised by @CSSFrancis. (Feel free to move this comment to another issue/discussion.) > > > > > Maybe we can...

> Thinking about it, I backtrack from my suggestion to use a `dict` to store axis labels. As you mention, that would provide an alternative way to annotate data, but...

@francisco-dlp I've come back to this a couple of times and it appears that there isn't an easy way to define the `is_array_increasing` function in a consistent way. The problem...

@jlaehne Good to keep track of these things. I need to look into the PR again but I'd love to have your opinion on some of the proposed changes as...

This seems like it was pretty out of date so I had to pull some of the changes into a new branch and rebase on that. If someone wants to...

> @CSSFrancis, I have the feeling that this will be quite a bit of work to finish this PR and most likely too much for the 2.0 release, what do...

Circling back to this just so this doesn't get dropped. I think that this should be in pretty good shape, but we will need to handle the case where you...

@jlaehne even though I'm usually in the several small PRs camp this is about the minimum size to get the logic correct for the Axes and AxesManager classes. I'll come...

> The problem with returning a new object is that the events connected to the old one will be lost. @ericpre I'll make sure that this isn't the case... Seems...

@sivborg would this not be fixed by just having the compute navigator function call `nansum`? Or are all of the values nan in the middle chunk? Using the center chunk...