Open Letter: Dear C3.js & the community from billboard.js

[netil]

Dear @kt3k, @masayuki0812 and C3.js community:

The year 2021, billboard.js project turns its 4yrs of journey after the initial fork from C3.js v0.4.11 Jun, 2017.

The initial motivation and reason of the new project were arose from the doubt of the continuity of C3.js.

Why we decided to start billboard.js?

C3.js has been contributed and helped facilitating the ease way to visualize data on the web, and it should be respected for the works done.

How to make these efforts done continues without the worries and concerns of the discontinuity?

The big and hard part of the Open Source project faces, are the "sustainability" and the "imbalance" between producers and consumers.

How to solve them? How to make worth OSS project continue its way?

We're not writing to give or propose any solution, but we're strongly believe with the help & contribution from the community can over come(or at least make better) this.

From the billboard.js project, we showed and proved the willing and demonstrated its stable continuity of the project by adding & enhancing the functionalities such as:

• New chart types
  • Radar, Area range, Bubble, Muti Gauge
• Introduced Plugin Interface & its plugins
  • Bubble Compare
  • Stanford Diagram
  • Text Overlap
• Performance improvements
  • billboard.js 2.1.0 release: Another big step for performance improvement & more!
• Make able to be tree-shaken to cut down final bundle size
• Themes
• Ensured the stability by increasing the Test Coverage up to 90%
• Transition of codebase to TypeScript
• Various new enhancements options
  • Pie's inner/outerRadius, Multi Axes, Legend Template, Custom Data Points, etc.

TBH, we don't know the current status of the C3.js project and its continuity, but as we expressed the worries as 2017,

If this status continues, as your worries and concerns, is quite reasonable to predict the death of the project.
#1965 (comment)

it seems similar concerns are expressed by the users again. (correct us if is wrong)

• Dead ?
• C3 HAS BEEN FORKED BY BILLBOARD.JS WHICH IS ACTIVELY MAINTAINED

Very carefully, we want to propose the idea of "unifying the efforts" (like inviting all core committers, etc.) to billboard.js.

Providing an unified source, surely will make beneficial to all.
And will make stronger the legacy and its works done from the C3.js.

Hopefully the proposal idea to be understood in a good way and to be considered positively.
(Also welcoming any other improved ideas/opinions as well)

Sincerely,
from the billboard.js team.

[GDFaber]

Hi, I don't have much time at hand right now, but want to say I'm glad that a conversation between the projects has started.

[GDFaber]

I first found out about C3.js in 2017 when I was looking for a charting library based on D3. I have been using C3 in my projects since then, because it hides away the rather difficult D3 parts while offering a well thought-out API with many options, yet it is possible to use D3 for customization.

That said, my shoutouts clearly go to the people who built this beautiful charting library. This is also why I asked to become part of the project team, to give something back in return, and to help maintaining the project.

However, it turned out that my spare time is less than I expected, and many of the issues require more code insight than I can offer at the moment. The core maintainers also seem to be pretty busy for what I can say, as there is hardly any commit apart from Dependabot merges.

Is was not until the beginning of this year that I found out about billboard.js and took a look at it. What I can say at a first glimpse is that it looks and feels as good as C3, with a few minor drawbacks from my personal point of view (for example, I like the C3 tooltips better because they move with the cursor). But that's just details, and there are even new chart types. Obviously billboard.js gets better maintenance and development on a frequent basis, while the C3 project has its problems there.

I'd like to say that I'd appreciate if the billboard.js website gave more credit to C3, though. I mean, it absolutely does, but you have to go to the API page to see it, and to be honest it does not really feel like shoutouts to me, but maybe that's just me :-)
#2829 (comment)

I would be more than happy if the core maintainers of C3 and billboard would find a way to get both C3 and billboard back into one project again, as it had been before the fork in 2017. But I understand that this is not an easy task, and not only because of differences in the code. And I also understand that this may only be my own point of view and my two cents I'm throwing in.

Finally, here are some wild guesses and questions for a start (I don't mean to provoke or insult anyone with this, forgive me if I do):
Will billboard.js agree to become a part of C3.js like considered back in 2017?
Will C3.js agree to join forces with billboard.js to become even more efficient?
Will there be a way to work in parallel with both of the projects so we can both prosper?

What do you think? Where should we go from here? I'm excited for your opinions on this topic.

Have a great day

[netil]

"Forking" and starting new project is quite frequent in a OSS world(well known ex. like KHTML → WebKit(WebCore) → Blink) and we think "billboard.js" is taking similar path as other forked project, being totally different & independent.

So, from this basis, is quite difficult to take options like, "work in parallel", nor the codebase merge.
(not just because of the criteria only, realistically there's no benefit for both)

Our option goes for this.

Will C3.js agree to join forces with billboard.js to become even more efficient?

[netil]

@kt3k what's your thought?

[kt3k]

@netil Hmm.. thank you for your message. I have to admit billboard is better maintained and developed lately, and it still keeps the sprit of c3 very well. Thank your for your effort on this.

I'm going to talk with @masayuki0812 in person about whether we can do anything about your suggestion. Thank you for your suggestion anyway.