[font-licensing] GUST font license Question regarding embedding
Jerzy.Ludwichowski
Jerzy.Ludwichowski at gust.org.pl
Sat Jun 17 16:43:43 CEST 2023
Hi Max,
what follows is Karl Berry's reaction to your concerns, very slightly
edited for the benefit f the list. Karl was/is GUST's e-foundry's
principal advisor in GFL matters so no wonder that I wholly agree with
his thoughts.
[Maximilian Bloch via font-licensing] :
>
> As a distributor of free fonts and also as an avid Linux user for over
> 20 years I am in the process of writing an FAQ regarding the GUST font
> license <https://www.gust.org.pl/projects/e-foundry/licenses> and I am
> hoping someone here could give me feedback by pointing out possible
> mistakes or help formulate a better more precise answer?
>
> Q: does the GUST font license (GFL) allow embedding in a portable
> digital document like PDF or an ebook?
>
> A: Most likely yes, but not 100% reliably. Even if possibly all
> authors of GFL licensed fonts may have intended to allow embedding
> in portable digital documents, the GUST font license is
> unfortunately a bit ambiguous. For sure, fonts licensed under the
> GUST font license (GFL) can be used freely and restrictions only
> apply tomodification and distribution
> <https://www.latex-project.org/lppl/lppl-1-3c/#conditions-on-distribution-and-modification>of
> the font files themselves.
>
> The actual question then would be: "Is embedding a font into a PDF
> or ebook considered a modification of the font?" While technically
> this can be the case, as some bits and bytes may be altered, the
> intended visual outcome of using the font remains the same. Also
> considering that the font is not intended for use outside of the
> created document, distribution of the document does not have to be
> viewed as a distribution of the font.
>
> The GUST font license (GFL) does not explicitly depict embedding,
> so the license could raise questions among technical savvy users
> and interpretation of the license becomes a matter of opinion.
> This is the reason why thefont exception
> <https://spdx.org/licenses/Font-exception-2.0.html>clause has been
> added to the similar GNU GPL license and also why the similarSIL
> OFL license
> <https://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&id=OFL>was
> born, where embedding isexplicitly allowed
> <https://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&id=ofl-faq_web#dc00d574>.
>
[Karl Berry, private communication]:
> 1. We certainly intend to allow embedding. A font license that does not
> permit embedding, both "whole font" and subsetted, seems next to useless
> to me, nowadays.
>
> 2. Embedding a whole font seems non-problematic to me. I'm skeptical of
> "bits and bytes may be altered". Bits and bytes may be altered by any
> copying method for any material -- consider DOS CRLF vs. Unix LF, for
> instance. Processing a PDF through Ghostscript, which everyone does,
> will generally change the exact bytes of included images (possibly
> problematically, as Jacko knows, but that's a different topic 😄, but no
> one thinks that changes the license of the document, or is not
> allowed. Etc.
>
> 3. Embedding a subset of a font, which is the interesting case, is more
> problematic. I think that has to be considered a modification of the
> original. Fortunately the LPPL (of course) allows modification, or it
> would not be a free software license. Furthermore, the fonts themselves
> will (should) contain metadata that gives the original source (url) of the
> font, which to my mind satisfies the crucial LPPL condition 6.d.2:
> 2. Information that is sufficient to obtain a complete,
> unmodified copy of the Work.
>
> Ergo, it seems to me that there is a reasonable argument that GFL'd
> fonts can be partially (or wholly) embedded in documents, without the
> user needing to do anything explicit. Which is the outcome we
> want. There are certainly counterarguments and alternate interpretations
> which would forbid embedding, but we don't need to endorse or address
> them, IMHO.
>
> 4. I'm skeptical of inducing FUD because of "interpretation". Every word
> of every license is open to interpretation. Linus interprets the GPL
> differently than the FSF. If a user wishes to shoot themselves in the
> foot by refusing to embed LPPL-d fonts, fine, that's their choice.
>
> 5. We can't "fix" it by adding an exception (as you somewhat imply that
> we should), or doing anything else that would make the GFL no longer the
> same as the LPPL, legally speaking. That would be a far worse outcome
> than anything else, triggering a full license review by everyone, to the
> detriment of all concerned.
>
> 6. Thus my suggestion would be to add an item to
> https://www.gust.org.pl/projects/e-foundry/licenses
> stating that we believe the LPPL/GFL, as written, permits embedding,
> either in whole or in part. With a link to the font-licensing archived
> message, as you suggest. And that that is our desired outcome.
As to the last point (6): this has yet to be done and will, but don't
hold your breath.
Best,
-- Jerzy
Jerzy Ludwichowski
GUST, Vice President
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/font-licensing/attachments/20230617/72a9f404/attachment.htm>
More information about the font-licensing
mailing list.