Last week, I wrote about the subtitles of what the SIL Open Font License (OFL) can be applied to.

When people say a font is open source, they might really mean:

  • The distributed font files (ex. the .otf, the .woff2) are released under a permissive license, typically the Open Font License
  • The entire project is distributed under a permissive license, which includes source files (ex. the .ufo or .glyphs file), and associated documentation and build scripts

My feeling is that there are more examples of the latter, but perhaps that is because of the amount of time I spend on GitHub, and the visibility these projects get via Google Fonts.

Here’s the relevant part of the the full text of the license OFL:

“Font Software” refers to the set of files released by the Copyright Holder(s) under this license and clearly marked as such. This may include source files, build scripts and documentation.

Emphasis mine. (Conversion to curly quotes, also mine.) This is not legal advice, but in my reading, this indicates the “Font Software” does not necessarily need to include the “source files.”

In many other software licensing situations, you wouldn’t be able to make this distinction, since the software source code is what you are licensing.

Maybe a better metaphor in this case is music. You’re probably familiar with Nick Sherman’s typeface versus font metaphor:

The way I relate the difference between typeface and font to my students is by comparing them to songs and MP3s, respectively (or songs and CDs, if you prefer a physical metaphor).

You can take this idea pretty far, even in this context: if the typeface is the song, the font is the .mp3, then the source files are mixed tracks of the original recordings.

Having the individual recordings—the stems—of an audio track make it much easier to work with individual pieces of the song. For example, some producers go as far as using only the vocal track, and discarding everything else in favour of their own original work.

Say you have access (and permission) to make a remix of a song, but only have the mixed-down, distributed file that everyone is used to hearing. Are there tools and processes to extract the vocals out for your remix? Yes, but it will take more effort, and the quality will be reduced.

The same idea applies to OFL licensed fonts and source files.

If you are hoping to “remix” an existing typeface with a permissive license, you might be able to do what you need starting with the distributed .otf file as a base. It would be easier, and there would be less data lost along the way, if you were able to work with the complete source files.

If you are choosing to make a remix, rather than it being a commission, the availability of these source files might influence your decision about what song to choose.

For typefaces, I suspect you are much more likely to have the choice of a starting point, even if the end result is a commissioned project. With that in mind, it’s worth factoring in the availability and state of the source files in your selection.

Until next time,