Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Better styling defaults #823

Open
philippjfr opened this issue Aug 18, 2016 · 9 comments
Open

Better styling defaults #823

philippjfr opened this issue Aug 18, 2016 · 9 comments

Comments

@philippjfr
Copy link
Member

Our default styling options were largely determined very early on when I fiddled with the matplotlib defaults a little bit and did not go through any detailed discussion. With matplotlib 2.0 (currently in beta) a lot of the default styling will change and I think that's a good time to also rethink some of our defaults. While matplotlib will be our main focus in this redesign, perhaps it's also worth thinking about whether we want our styling to be a bit more consistent between backends. I don't have any detailed suggestions yet but was hoping to kick off some discussion here, so please report what has always annoyed you about our defaults. I know the default matplotlib background color is something I definitely want to change.

@jlstevens
Copy link
Contributor

I agree matplotlib 2.0 is a great excuse for us to also improve our defaults. :-)

@jbednar
Copy link
Member

jbednar commented Aug 22, 2016

I agree as well; mpl 2.0 is the time, and improving consistency with Bokeh would also be great (and is also timely because Bokeh is about to reach 1.0). I'd personally be happy if all of our example notebooks came out reasonably similar between the two backends, after these updates.

For that to be meaningful I think we need to update our examples to use non-square aspects a lot more often, so that we notice behavior for long or tall plots better.

@philippjfr
Copy link
Member Author

The matplotlib team have just uploaded the default styling changes for 2.0. You can see them here.

@jlstevens
Copy link
Contributor

jlstevens commented May 30, 2017

I've assigned this to 1.8 and repurposed this issue to cover bokeh style changes as well.

My first proposed bokeh style change: make errorbars thicker by default:

@jlstevens jlstevens changed the title Better matplotlib styling defaults Better styling defaults May 30, 2017
@jlstevens
Copy link
Contributor

I also think we should start adding concrete suggestions for style changes to this issue. E.g making the default colormap 'fire'?

@jbednar
Copy link
Member

jbednar commented May 30, 2017

I think we should update to mpl 2.0 defaults in general; they look better to me, and they are what new users will get with mpl by default outside of hv.

@jbednar
Copy link
Member

jbednar commented May 31, 2017

My first proposed bokeh style change: make errorbars thicker by default:

I agree. I also favor removing the grid lines by default, and making the default sizes comparable between the two backends.

@philippjfr
Copy link
Member Author

I think we've addressed this for the most part. I'm sure there will be more tweaks we want to do so I'll reassign to 2.0.

@philippjfr philippjfr modified the milestones: v2.0, v1.8 Jun 27, 2017
@jlstevens
Copy link
Contributor

I think we've addressed this for the most part.

I agree! I just want to mention that #1610 was merged as well.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants