Distributing work of adding icons via PR?

[jonathan-g]

Right now, the process for adding new icons is to create an issue and submit an SVG of the requested glyph and wait to see whether @jpswalsh wants to add it when he has time in his busy life.

Is there a way this project could be adapted to allow users to do more of the work ourselves: convert the SVG into the appropriate format to add to the .eot, .ttf, .svg, and .woff font files, and then submit a PR, as opposed to needing one person to do all the work alone?

For instance, I'd be willing to do work to speed up adding the SSRN logo (#42) and The Conversation (#56) to academicicons if there were a clear process for doing so.

Part of the challenge is that git is not good at merging different versions of binary files like academicicons.ttf, so a process like this would only work well if there were a toolchain to automate building the main font files (academicicons.ttf, .eot, .svg, and .woff) from .svg files for the individual glyphs (or from a master .svg for the whole font, since that's a textual XML-based format), and adapting this project to that way of working might be a lot more trouble than it's worth.

[jpswalsh]

I agree 100% with this suggestion, and have reached the same conclusion myself. However, I don't know how to go down that path. If there were some automated way to do the multiple steps I go through in the IcoMoon online tool, with only a folder of SVGs as input, then there wouldn't be the bottleneck that there is now.

Having said that, I think there's an argument to be made that we can move away from the font file approach entirely, which is actually a bit of a hack. I'd much rather Academicons became first and foremost a curated library of text-based vector icons, where the output is free to change with the trends of the web design world. There has been talk for a couple of years on the merits of using the .svg format for icons rather than font files. I'd be happy to support any work in this direction.

[Bertbk]

I also totally agree. I would also love to help but I'm clearly not expert in "icon process". Have a few questions, though:

1. Wouldn't fontello and it's API be a usefull friend for us? They provide ready to use Makefile. Users would thus "only" have to add their SVG to the JSON config file, which is git-friendly and we could build a CI / Test. I tried it locally with a single homemade svg-icon (HAL, issue #99 ) and it worked like a charm...

2. @jpswalsh Could you please elaborate about keeping only .svg file and leaving the font files away? In particular, would that breaks the CTAN LaTeX package you provide?

Thanks

Edit : Maybe, a good idea would be to merge academics icon into ForkAwesome? Moreover, it seems to be simple to do a PR in their project, see for example this one