NumFOCUS

can you describe the sorts of improvements you are looking for? please also feel free to open issues on github.

Can we open this under a open discussion for and not an issue. Since this, to be a bit oversimplified, it's none astro format vs astro formats there is no docs about those astro formats and it makes yt unusable for us non astros

We're getting together an advisory board for a domain expansion project to address exactly this type of thing. More details, including proposal, here: http://lists.spacepope.org/pipermail/yt-dev-spacepope.org/2017-August/006933.html … . Info on participating, etc will be coming in early Jan.

Because @yt_astro has similar overall architecture (when viewed by an end user) to @HoloViews it would advantages to the num/vis python community if @HoloViews would participate at this meeting.

We haven't done the best job of reaching out to HoloViews, but certainly will try. But at the same time, I disagree about the position that yt has in the community and that of HoloViews. They are *considerably* more complementary than you are describing.

That 'zero-sum' deal is from a comment in a fight I was in earlier this year. It's not at @yt_astro . Just that it is important in how developers need to view end users. Sepcilify end user we haven't liberated from closed source.

I'm pretty sure I'm the person that said it's not a zero-sum game, when we engaged on twitter some months ago. I don't understand what actions you are advocating, and it's not clear to me that there is a prescriptive authority that would be able to execute those actions.

@powersoffour lets pick that up. In @HoloViews the interactive tools for Bokeh are amazing. But Bokeh last I checked does not have LaTex support for in graph text https://github.com/bokeh/bokeh/issues/6031 …. So and for that one feature that @matplotlib has I now have to decide which do I use.