License choice?

[HughP]

What do we mean by the "Non-Commercial" clause in the Creative commons license? and is there a reason we are not using the font standard OFL? https://en.wikipedia.org/wiki/SIL_Open_Font_License

[cyphar]

FWIW, I agree with using a proper free font license as per GNU, but the CC NC clause is pretty well understood at this point. It's possible that Randall didn't provide them with the rights to use his handwriting under CC-BY or CC-BY-SA, so this font (as it is a derivative work) can't provide people with more rights than they were given.

[takluyver]

NC-BY is also the license for XKCD itself, so the font which is made from glyphs Randall drew probably inherited that license by default. I don't recall any discussions about licenses.

[pelson]

For the record, I'm happy to release all work currently available in xkcd-script (namely that originally documented in https://pelson.github.io/2017/xkcd_font/) in any way seen fit by the ipython/xkcd-font community - I don't think that covers the font itself though, as it could reasonably be considered a derivative work (as @takluyver said).

[HughP]

FWIW:

My understanding, as I have followed the development of this font, is that the font is actually derived from a hand writing sample. That sample has the same author as XKCD itself, but the sample is not part of XKCD. But this is only as I understand the facts, I could be wrong in my understanding of the facts. perhaps @randallmunroe could license the handwriting contribution CC-BY or CC0; if he did license it CC-BY we could name the font "RandallMonroe.ttf" and suggest that the terms of the CC-BY are met. If I am correct, that hand writing sample is not be subject to the same license as XKCD, we can be free from inheritance concerns raised by @takluyver and @pelson.

Choosing OFL 1.1 does give others the right to bundle the font with other software, that other software is allowed to have a separate terms of service and separate license. The GNU Font License does not allow for bundling with non-GNU licensed materials. So, in a sense, the OFL is more liberal in how products licensed with it can be used. While I do personally like GNU projects and the license and the social philosophy behind it, I think that a allowing for a wider and more liberal delivery of this font (independent of the software that created it) is more advantageous.

If @pelson 's work is licensed under the MIT license then derivative works can be free of the Share-Alike nature of the GNU type licenses. Freeing the font to be licensed OFL.

As I understand the data presented by Creative Commons in their report on the community understanding of the NC clause, it does not actually support @cyphar 's assessment that the the NC clause is well understood. The report from creative commons actually suggests that there are multiple understandings of what NC actually means and that people who use NC are prone to subjective interpretations depending on their opinion about the nature of the organization using the license. Additionally, the NC clause does not prevent the sell of access to NC licensed content, it only prevents the sell of the content itself (sometimes). The good news is that if this font was licensed with CC-NC that it might still be used in a Springer publication. <snark> . It is because of these these ambiguities around the NC clause that I do not recommend the NC clause - ever.

Actually, Creative Commons suggests that software should not be licensed under CC licenses. Fonts are one of those creations which are both creative expressions and software. This is one reason that the SIL guys decided to write the OFL.

I hope these thoughts helps us move in some direction more clearly.

[cyphar]

The GNU Font License does not allow for bundling with non-GNU licensed materials.

I wasn't saying the font should be GPL'd, I was just pointing to the list of free font licenses as collated by GNU / the FSF (the GPL is on there but so are several other licenses).

The good news is that if this font was licensed with CC-NC that it might still be used in a Springer publication.

I didn't know that. Given that Springer publications are ... sold (at quite a hefty price I might add, as a current university student) ... you would think that wouldn't be allowed. I stand corrected. Personally I've never advocated for the NC licenses because they stifle free culture.

[HughP]

@cyphar My understanding is that Springer only sells access. I could be wrong. But that is my understanding. I was also being a bit snarky about Springer. I know there is lots of love loss between the academic community and that particular publisher over their chosen business model.

[nicolasmaia]

Regarding the handwriting design, I think it pretty clearly falls into this category: https://commons.wikimedia.org/wiki/Template:PD-text

For that reason, I'd say it's safe to release this under a free license.

[takluyver]

I definitely do not think that applies. The linked wikimedia page is for text conveying facts or data in a generic typeface (e.g. a list of melting points of different metals). Because they are not creative works, they are not covered by copyright law. XKCD is clearly a creative work, and its specific lettering style is part of that. As such, it is covered by copyright, and the license Randall uses for it stands unless he decides to grant us a different license to the font samples in question.