Re: Added RFN-violation bug template to wiki

On Sat, Mar 17, 2018 at 9:53 AM, Dave Crossland <dave@lab6.com> wrote:

I think for those who want to use RFNs that desire is perfectly valid.

Therefore I think rather than asking for removal, the recommendation could either be to use (a) a codename for the development of the family that is not reserved (and that Debian will use) or (b) drop the RFN and use trademark assertions/registrations instead.

Yeah, I think the "advise designers on their trademark options" approach has been vastly underexplored. I attempted, briefly, to draft a couple of boilerplate font-name-trademark templates; perhaps it's time to resurrect that project.

IMO, the underlying issue is that RFN clause implements a trademark-like protection within a copyright license; keeping the two separate might make for fewer instances of confusion. But who knows; I'm speculating.

On Fri, Mar 16, 2018 at 10:44 PM, Paul Wise <pabs@debian.org> wrote:

I wonder if we should also add a sentence suggesting to ask upstream
to remove the RFN. It could also provide upstreams with the less
desired alternative of softening the RFN it to allow building from
unmodified source and encourage upstream to provide a deterministic
automated build process.

Pushing for a change seems slightly too aggressive, to me; maybe we could start by just explaining the impact that the RFN has and offering more than one alternative? I would agree with Dave that there are perfectly valid uses for RFNs — a prime example being the SIL fonts, which claim an RFN on the "SIL" portion of the font name, not on the entirety. Unscientific surveying suggests that many type designers aren't aware that you can even use RFNs in a piecemeal fashion like that; a little more public educating could go a long way toward decreasing unintentional licensing conflicts.

Nate

nathan.p.willis
nwillis@glyphography.com