[ft-devel] new CFF engine

[Dave Crossland]

:D

---------- Forwarded message ----------
From: Werner LEMBERG <wl-mXXj517/***@public.gmane.org>
Date: 1 May 2013 20:11
Subject: [ft-devel] new CFF engine
To: freetype-devel-qX2TKyscuCcdnm+***@public.gmane.org



Folks,


Adobe, in collaboration with Google, has contributed a new CFF parsing
and hinting engine. The last few months it was my job to fully
integrate the code into FreeType, and I'm very pleased with the
results. Here are links to the relevant blogs:

http://adobe.ly/12mJWGv
http://google-opensource.blogspot.com/2013/05/got-cff.html

Look at the comparison images :-)

Below are the relevant parts of the CHANGES file (which also give
details how to enable it); as you can see, there are also a bunch of
improvements for ftview and ftdiff.

I plan to release 2.4.12 in about a week.


Please test and enjoy!

Werner


======================================================================


- We have another CFF parsing and hinting engine! Written by Dave
Arnold <darnold-dv/***@public.gmane.org>, this work has been contributed by
Adobe in collaboration with Google. It is vastly superior to
the old CFF engine, and it will replace it in the next release.
Right now, it is still off by default, and you have to
explicitly select it using the new `hinting-engine' property of
the cff driver:

...
#include FT_CFF_DRIVER_H

FT_Library library;
int engine = FT_CFF_HINTING_ADOBE;


...
FT_Property_Set( library, "cff", "hinting-engine", &engine );

The code has a (mature) beta status; we encourage all users to
test it and report any problems.

[...]

- Using the `H' key, it is now possible to select the CFF engine
in both `ftview' and `ftdiff'.

- It is now possible to directly select the LCD rendering mode
with the keys `A'-`F' in `ftview'. The key mapping for cycling
through LCD modes has been changed from `K' and `L' to `k' and
`l', and toggling custom LCD filtering is no longer mapped to
key `F' but to key `L'.

- In `ftdiff', key `x' toggles between layout modes: Either use
the advance width (this is new and now the default) or the
bounding box information to determine line breaks.

_______________________________________________
Freetype-devel mailing list
Freetype-devel-qX2TKyscuCcdnm+***@public.gmane.org
https://lists.nongnu.org/mailman/listinfo/freetype-devel

[Elliot Polinsky]

Hi, could I unsubscribe from this list? Thanks.

Post by Dave Crossland
:D
---------- Forwarded message ----------
Date: 1 May 2013 20:11
Subject: [ft-devel] new CFF engine
Folks,
Adobe, in collaboration with Google, has contributed a new CFF parsing
and hinting engine. The last few months it was my job to fully
integrate the code into FreeType, and I'm very pleased with the
http://adobe.ly/12mJWGv
http://google-opensource.blogspot.com/2013/05/got-cff.html
Look at the comparison images :-)
Below are the relevant parts of the CHANGES file (which also give
details how to enable it); as you can see, there are also a bunch of
improvements for ftview and ftdiff.
I plan to release 2.4.12 in about a week.
Please test and enjoy!
Werner
======================================================================
- We have another CFF parsing and hinting engine! Written by Dave
Adobe in collaboration with Google. It is vastly superior to
the old CFF engine, and it will replace it in the next release.
Right now, it is still off by default, and you have to
explicitly select it using the new `hinting-engine' property of
...
#include FT_CFF_DRIVER_H
FT_Library library;
int engine = FT_CFF_HINTING_ADOBE;
...
FT_Property_Set( library, "cff", "hinting-engine", &engine );
The code has a (mature) beta status; we encourage all users to
test it and report any problems.
[...]
- Using the `H' key, it is now possible to select the CFF engine
in both `ftview' and `ftdiff'.
- It is now possible to directly select the LCD rendering mode
with the keys `A'-`F' in `ftview'. The key mapping for cycling
through LCD modes has been changed from `K' and `L' to `k' and
`l', and toggling custom LCD filtering is no longer mapped to
key `F' but to key `L'.
- In `ftdiff', key `x' toggles between layout modes: Either use
the advance width (this is new and now the default) or the
bounding box information to determine line breaks.
_______________________________________________
Freetype-devel mailing list
https://lists.nongnu.org/mailman/listinfo/freetype-devel
--
Cheers
Dave

[Dave Crossland]

Post by Elliot Polinsky
Hi, could I unsubscribe from this list? Thanks.

I have unsubscribed you :-)

[Victor Gaultney]

Dear OFLib folks,

Nicolas Spalinger and I, the maintainers of the SIL Open Font License, have posted a draft of an update to the OFL-FAQ (1.1-update3-draft). Although there are many small clarifications and refinements from version 1.1-update2, the main addition is a greatly expanded section related to web fonts. There is also a related discussion paper on Web Fonts and Reserved Font Names that deals with those issues in even more detail. Comments and feedback are welcome.

http://scripts.sil.org/OFL-FAQ_web_11update3draft
http://scripts.sil.org/OFL_web_fonts_and_RFNs

We hope this helps to bring clarity to some of the many difficult issues related to web fonts and Reserved Font Names.

Thanks,

Victor Gaultney

[Dave Crossland]

Hi Victor!

Awesome, thanks!

Could you provide a link to the txt version of the current FAQ (v2) so
I can make a diff ? :)

Cheers
Dave

Post by Victor Gaultney
Dear OFLib folks,
Nicolas Spalinger and I, the maintainers of the SIL Open Font License, have posted a draft of an update to the OFL-FAQ (1.1-update3-draft). Although there are many small clarifications and refinements from version 1.1-update2, the main addition is a greatly expanded section related to web fonts. There is also a related discussion paper on Web Fonts and Reserved Font Names that deals with those issues in even more detail. Comments and feedback are welcome.
http://scripts.sil.org/OFL-FAQ_web_11update3draft
http://scripts.sil.org/OFL_web_fonts_and_RFNs
We hope this helps to bring clarity to some of the many difficult issues related to web fonts and Reserved Font Names.
Thanks,
Victor Gaultney

[Dave Crossland]

Adobe’s patent had run out anyway?

What patent?

[vernon adams]

So (their) fonts would look better rendered on screens?

Great, but why was this work done?

[Barry Schwartz]

Great, but why was this work done?
So (their) fonts would look better rendered on screens?

I would assume they did it to advance the state of the art and support
human progress. For years I’ve been wanting them to do this for
exactly that reason. Also Google asked Adobe nicely (I assume),
because Google wanted to advance the state of the art, support human
progress, and make money doing so.

It is not good to withhold knowledge of how to render fonts well, when
that’s not even a source of income due to the withholding, and when
obviously most of the rest of the industry hasn’t the vaguest clue how
to render a font, and causes extensive damage by trying
(unsuccessfully) to foist that job onto font designers.

Unsurprisingly this also makes good business sense. Perhaps full
openness a lot sooner could have prevented the existence of TrueType
in the first place.

[Vernon Adams]

I might well do a little of that, not that i have any free time :) …but…
I'm very curious to see how desktop UI fonts render under this new freetype CFF autohinter.

One thing i have noticed when testing UI fonts on android and Chrome (both use a tweaked freetype) is that unhinted truetype fonts render better than ttfautohinted or manual hinted fonts. The freetype bytecode interpreter on those OS's (especially on a hi res screen) seems to result in too harsh rendering with hinted truetype fonts. I'm curious if the new CFF interpreter will bring opentype fonts to the first choice as a UI font.

-v

Somebody with way more free time than me could do some tests of autohinted TTF and autohinted CFF under the latest FreeType.

[Liam R E Quin]

Post by Vernon Adams
One thing i have noticed when testing UI fonts on android and Chrome
(both use a tweaked freetype) is that unhinted truetype fonts render
better than ttfautohinted or manual hinted fonts.

"better" is in many ways subjective. Distance from the viewer's eye to
the screen and the viewer's eyesight make a big difference.

People who wear glasses often prefer fonts that are not anti-aliased,
for example, especially at distances of under 18 inches/0.5 metres.

Ambient lighting can also make a difference.

Having said that, hinting for Type 1 / CFF fonts is very important, e.g.
to get the three vertical strokes of "m" to be the same thickness (if
that's desired by the design) at low resolutions.

Liam

[Vernon Adams]

Thomas (or others),
Could you recommned a few opentype fonts, with known good industry standard ps hints, that i could use for control tests of this new engine versus truetype on Freetype?

thanks

-v

Somebody with way more free time than me could do some tests of autohinted TTF and autohinted CFF under the latest FreeType. My suspicion is that on average the autohinted CFF will hold up better down to smaller sizes and look at least as good if not better overall.

[Liam R E Quin]

:D

Much awesomeness and kudos to Adobe!

Liam

[Vernon Adams]

I am starting to prefer that the default position of RFN's in the OFL should be preserved as it is now. Is the other option, to make the default 'no RFN'? IMO the default of the RFN keeps a 'positive balance' in the court of the original designer. Remember there is allways the choice of specifically not using an RFN.

In the event of another party wishing to retain the RFN after modifications have been made, the simple solution is for that party to obtain an agreement with the copyright holder of the font to retain the RFN.

Or, is there something i am not understanding in this? :)

-vernon

Dear Google Fonts folks,
Nicolas Spalinger and I, the maintainers of the SIL Open Font License, have posted a draft of an update to the OFL-FAQ (1.1-update3-draft). Although there are many small clarifications and refinements from version 1.1-update2, the main addition is a greatly expanded section related to web fonts. There is also a related discussion paper on Web Fonts and Reserved Font Names that deals with those issues in even more detail. Comments and feedback are welcome.
http://scripts.sil.org/OFL-FAQ_web_11update3draft
http://scripts.sil.org/OFL_web_fonts_and_RFNs
We hope this helps to bring clarity to some of the many difficult issues related to web fonts and Reserved Font Names.
Thanks,
Victor Gaultney
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups "Google Font Directory Discussions" group.
For more options, visit https://groups.google.com/groups/opt_out.

[Dave Crossland]

Hi,

Post by Vernon Adams
Or, is there something i am not understanding in this? :)

I think its unreasonable to expect every person publishing a blog who
makes their own subset to contact every copyright holder every time
they want to use a new OFL-RFN web font.

I think the desirable effect of RFNs can be preserved with a trademark
license, while alleviating the undesirable effect of prohibiting
subsetting and other trivial changes while retaining the name users
are familiar with.

I note that SIL itself also asserts both RFN and trademark rights on
its font names; so if you are keen to maximize control over your font
names, it may make sense to do what I'm suggestion in addition to
RFNs, rather than replacing them.

Cheers
Dave

[Dave Crossland]

Still not seeing why no RFN needs to be the default position

I want to see libre fonts as widely used as possible while remaining libre.

[Vernon Adams]

Post by Dave Crossland
I want to see libre fonts as widely used as possible while remaining libre.

Agreed. However, it's the 'remaining libre' aspect that is my concern. You know, we have the situation at present where at least one major corp is using my fonts in their services and not even i am 'free' to get the modified fonts they are using :) I realise that is only indirectly an RFN issue but for me it flags up the need to look at the altering of licensing for Libre fonts with caution.

-v

[Dave Crossland]

It's the 'remaining libre' aspect that is my concern

Replacing RFNs with trademarks is still the OFL and keeps all the
benefits the OFL has for remaining libre.

[Khaled Hosny]

Post by Vernon Adams

Post by Dave Crossland
I want to see libre fonts as widely used as possible while remaining libre.

Agreed. However, it's the 'remaining libre' aspect that is my concern.
You know, we have the situation at present where at least one major
corp is using my fonts in their services and not even i am 'free' to
get the modified fonts they are using :) I realise that is only
indirectly an RFN issue

It is not an RFN issue at all. The situation you describe is a violation
to any copyleft license (and OFL is a copyleft), with RFN or without.

Regards,
Khaled

[Vernon Adams]

I know. But. They are modifying and using the RFN without agreement from me.
The 'freedom' (or not) of the OFL files they are using, is another issue.
-vern

Post by Khaled Hosny
It is not an RFN issue at all. The situation you describe is a violation
to any copyleft license (and OFL is a copyleft), with RFN or without.

[Barry Schwartz]

Post by Vernon Adams

Post by Dave Crossland
I want to see libre fonts as widely used as possible while remaining libre.

Agreed. However, it's the 'remaining libre' aspect that is my
concern. You know, we have the situation at present where at least
one major corp is using my fonts in their services and not even i am
'free' to get the modified fonts they are using :) I realise that is
only indirectly an RFN issue but for me it flags up the need to look
at the altering of licensing for Libre fonts with caution.

Free software can be charged for; otherwise it is not free as in
freedom. Once you pay for it, if it is copylefted, _then_ you can
redistribute, etc.

So that’s not really a ‘situation’ unless they try to prevent the
second part.

[Vernon Adams]

Barry,
Not sure i understand 100% what you are saying :)
specifically - "Free software can be charged for; otherwise it is not free as in
freedom"
is there a 'not' missing?
-v

Post by Barry Schwartz

Post by Vernon Adams

Post by Dave Crossland
I want to see libre fonts as widely used as possible while remaining libre.

Agreed. However, it's the 'remaining libre' aspect that is my
concern. You know, we have the situation at present where at least
one major corp is using my fonts in their services and not even i am
'free' to get the modified fonts they are using :) I realise that is
only indirectly an RFN issue but for me it flags up the need to look
at the altering of licensing for Libre fonts with caution.

Free software can be charged for; otherwise it is not free as in
freedom. Once you pay for it, if it is copylefted, _then_ you can
redistribute, etc.
So that’s not really a ‘situation’ unless they try to prevent the
second part.

[Denis Jacquerye]

With some free software you have the freedom to sell it, to cover your cost
of distribution, support or for profit, as long as you provide the license
and offer the source allowing buyers to have the same freedoms you have.
The GPL gives that specific freedom:
https://www.gnu.org/licenses/gpl-faq.html#DoesTheGPLAllowMoney
The OFL doesn't give that specific freedom. If sold, it must be bundled
with other software, it cannot be sold on its own.
http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=OFL-FAQ_web#f9c4ccfe

Post by Vernon Adams
Barry,
Not sure i understand 100% what you are saying :)
specifically - "Free software can be charged for; otherwise it is not free as in
freedom"
is there a 'not' missing?
-v

Post by Barry Schwartz

Post by Vernon Adams

Post by Dave Crossland
I want to see libre fonts as widely used as possible while remaining

libre.

Post by Barry Schwartz

Post by Vernon Adams
Agreed. However, it's the 'remaining libre' aspect that is my
concern. You know, we have the situation at present where at least
one major corp is using my fonts in their services and not even i am
'free' to get the modified fonts they are using :) I realise that is
only indirectly an RFN issue but for me it flags up the need to look
at the altering of licensing for Libre fonts with caution.

Free software can be charged for; otherwise it is not free as in
freedom. Once you pay for it, if it is copylefted, _then_ you can
redistribute, etc.
So that’s not really a ‘situation’ unless they try to prevent the
second part.

[Pablo Impallari]

Post by Vernon Adams
You know, we have the situation at present where at least one major corp

is using my fonts in their services

Post by Vernon Adams
They are modifying and using the RFN without agreement from me.

I have the same concerns as Vernon.
I'm getting the feeling that removing the RFM will allow Adobe and MT to do
whatever they want without agreement from us.

Post by Vernon Adams
With some free software you have the freedom to sell it, to cover your
cost of distribution, support or for profit, as long as you provide the
license and offer the source allowing buyers to have the same freedoms you
have.
https://www.gnu.org/licenses/gpl-faq.html#DoesTheGPLAllowMoney
The OFL doesn't give that specific freedom. If sold, it must be bundled
with other software, it cannot be sold on its own.
http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=OFL-FAQ_web#f9c4ccfe

Post by Vernon Adams
Barry,
Not sure i understand 100% what you are saying :)
specifically - "Free software can be charged for; otherwise it is not free as in
freedom"
is there a 'not' missing?
-v

Post by Barry Schwartz

Post by Vernon Adams

Post by Dave Crossland
I want to see libre fonts as widely used as possible while remaining

libre.

Post by Barry Schwartz

Post by Vernon Adams
Agreed. However, it's the 'remaining libre' aspect that is my
concern. You know, we have the situation at present where at least
one major corp is using my fonts in their services and not even i am
'free' to get the modified fonts they are using :) I realise that is
only indirectly an RFN issue but for me it flags up the need to look
at the altering of licensing for Libre fonts with caution.

Free software can be charged for; otherwise it is not free as in
freedom. Once you pay for it, if it is copylefted, _then_ you can
redistribute, etc.
So that’s not really a ‘situation’ unless they try to prevent the
second part.

--
Denis Moyogo Jacquerye
African Network for Localisation http://www.africanlocalisation.net/
Nkótá ya Kongó míbalé --- http://info-langues-congo.1sd.org/
DejaVu fonts --- http://www.dejavu-fonts.org/

[Vernon Adams]

The RFN can have an integral role in how a designer can preserve or enhance a certain type of freedom for a font. Or it simply restrict a font's freedom. I'm still arguing with myself about it :)

Reserving the name of the font, sets down a licensing condition that must be met. If that condition is not met then the license is breached.
This is clear when dealing with 1 or 2 a large corporations (who may not be interested in preserving the font's freedom); as it gives 3 clear solutions for them to use their own modified version of a font;
1. Follow the OFL, use the font, change the font's name , and preserve the font's freedom.
2. Follow the OFL, use the font, get an agreement from me to use the RFN, and preserve the font's freedom.
3. Buy an embedding license from me, use the font non-free, and i preserve the font's freedom.

Now, with a few large corporations, this is highly manageable. But what happens with the mass of individual users and or small businesses, who maybe are also making modifications and serving the font ?
Of course, i can simply decide to ignore the potential mass of individual breaches of non-changing of RFN's, and instead simply focus on the few 'major' breaches. Bit of a license fudge tho imo.

The argument i have with myself is; why do i feel the use of RFN's is not necessary when dealing with masses of individual users, but i feel i want it there in case of corporate users?
It could be that i see that corporate users could easilly afford to buy (modestly priced) RFN agreements from me (if they need to use the fonts), thus 'giving something back' to the designer of the free fonts they are using, and funding future fonts.

-v

Post by Pablo Impallari

Post by Vernon Adams
You know, we have the situation at present where at least one major corp is using my fonts in their services
They are modifying and using the RFN without agreement from me.

I have the same concerns as Vernon.
I'm getting the feeling that removing the RFM will allow Adobe and MT to do whatever they want without agreement from us.

[Khaled Hosny]

Why do you think the font name is so important that to keep using it
those evil corporations will go out of their ways and sign deals with
you? (not to mention that it is the copyleft part of OFL that ensures
preservation of font freedom, not the RFN part).

Regards,
Khaled

Post by Vernon Adams
The RFN can have an integral role in how a designer can preserve or enhance a certain type of freedom for a font. Or it simply restrict a font's freedom. I'm still arguing with myself about it :)
Reserving the name of the font, sets down a licensing condition that must be met. If that condition is not met then the license is breached.
This is clear when dealing with 1 or 2 a large corporations (who may not be interested in preserving the font's freedom); as it gives 3 clear solutions for them to use their own modified version of a font;
1. Follow the OFL, use the font, change the font's name , and preserve the font's freedom.
2. Follow the OFL, use the font, get an agreement from me to use the RFN, and preserve the font's freedom.
3. Buy an embedding license from me, use the font non-free, and i preserve the font's freedom.
Now, with a few large corporations, this is highly manageable. But what happens with the mass of individual users and or small businesses, who maybe are also making modifications and serving the font ?
Of course, i can simply decide to ignore the potential mass of individual breaches of non-changing of RFN's, and instead simply focus on the few 'major' breaches. Bit of a license fudge tho imo.
The argument i have with myself is; why do i feel the use of RFN's is not necessary when dealing with masses of individual users, but i feel i want it there in case of corporate users?
It could be that i see that corporate users could easilly afford to buy (modestly priced) RFN agreements from me (if they need to use the fonts), thus 'giving something back' to the designer of the free fonts they are using, and funding future fonts.
-v

Post by Pablo Impallari

Post by Vernon Adams
You know, we have the situation at present where at least one major corp is using my fonts in their services
They are modifying and using the RFN without agreement from me.

I have the same concerns as Vernon.
I'm getting the feeling that removing the RFM will allow Adobe and MT to do whatever they want without agreement from us.

[Vernon Adams]

Erm…
I think it, because i see clear evidence of it :)

The RFN part was probably not conceived as a copyleft component of the OFL. My point though, is that it can (under certain circumstances) also be used to preserve certain freedoms, more than it may ever restrict freedoms.

Do you see that the RFN can restricts a font's freedom? I'm interested to hear thoughts on that, as i'm still grappling with all this.

-v

Post by Khaled Hosny
Why do you think the font name is so important that to keep using it
those evil corporations will go out of their ways and sign deals with
you? (not to mention that it is the copyleft part of OFL that ensures
preservation of font freedom, not the RFN part).

[Dave Crossland]

Post by Vernon Adams
The RFN part was probably not conceived as a copyleft
component of the OFL. My point though, is that it can (under
certain circumstances) also be used to preserve certain freedoms,
more than it may ever restrict freedoms.

What freedoms?

Post by Vernon Adams
Do you see that the RFN can restricts a font's freedom? I'm
interested to hear thoughts on that, as i'm still grappling with all this.

Yes, that is exactly what I am saying.

OFL-RFN fonts will simply not be used as widely on the web, because
the administrative burden of requesting, getting and tracking the
permission is too expensive - aside from any fees RFN holders might
try to charge.

[Vernon Adams]

Post by Dave Crossland

Post by Vernon Adams
The RFN part was probably not conceived as a copyleft
component of the OFL. My point though, is that it can (under
certain circumstances) also be used to preserve certain freedoms,
more than it may ever restrict freedoms.

What freedoms?

I'm playing devils advocat here a lot :) but i feel that if i use a RFN with particular font families it gives me at least some added control against their (mis)use. On a personal note, i am uncomfortable with this; i dislike benevolent dictators as much as any type of dictator!

Post by Dave Crossland

Post by Vernon Adams
Do you see that the RFN can restricts a font's freedom? I'm
interested to hear thoughts on that, as i'm still grappling with all this.

Yes, that is exactly what I am saying.
OFL-RFN fonts will simply not be used as widely on the web, because
the administrative burden of requesting, getting and tracking the
permission is too expensive - aside from any fees RFN holders might
try to charge.

Yes, i certainly agree that a non-RFN'd font is more mobile and more flexible than a RFN'd font.
I think designers will continue to have to make decisions about that, but then i see licensing as an integral part of a font's design anyway. The decisions you make about licensing will effect the mobility and use of the font, just as decisions on hinting, style, language support, etc will.

-v

[Khaled Hosny]

FRN, IMO, puts unnecessary restrictions on users of libre fonts to solve
and just complicates things (more so on the web) for no to very marginal
benefit. What issues FRN is trying to solve are half technical; change
in font metrics can cause text reflow, but technical issues ask for
technical not legal solutions (e.g. bundle the fonts with your document,
or even embed them which some thing Microsoft Office supported for years
and now Libre Office as well). The other half is the artistic integrity,
which I, obviously, find it all nonsense, fonts are functional utilities
not works of art, if I’m making a piece of art I’d use CC-By-NC-ND
or something like that.

Incidentally I don’t like the “not sold by itself” clause either, but
since FSF’s idea of bundling a simple hello world applications makes it
practically non-existent, I’m OK with it.

Basically the only reason I use OFL for my fonts is because it is the
most widely used copyleft license for fonts which allows for easy mixing
and sharing between fonts, but I have never been a big fan of it myself.

Regards,
Khaled

Post by Vernon Adams
Erm…
I think it, because i see clear evidence of it :)
The RFN part was probably not conceived as a copyleft component of the OFL. My point though, is that it can (under certain circumstances) also be used to preserve certain freedoms, more than it may ever restrict freedoms.
Do you see that the RFN can restricts a font's freedom? I'm interested to hear thoughts on that, as i'm still grappling with all this.
-v

Post by Khaled Hosny
Why do you think the font name is so important that to keep using it
those evil corporations will go out of their ways and sign deals with
you? (not to mention that it is the copyleft part of OFL that ensures
preservation of font freedom, not the RFN part).

[Dave Crossland]

Post by Khaled Hosny
The other half is the artistic integrity,
which I, obviously, find it all nonsense

Since many libre software projects use trademarks to maintain the
integrity of their name, I don't find the
artistic-work/functional-work distinction relevant here (although I
agree with you; I think that type is functional and this eclipses the
artistic aspects.)

[Khaled Hosny]

Post by Dave Crossland

Post by Khaled Hosny
The other half is the artistic integrity,
which I, obviously, find it all nonsense

Since many libre software projects use trademarks to maintain the
integrity of their name, I don't find the
artistic-work/functional-work distinction relevant here.

I’d not say many, only a few big ones who think there names are so
valuable, and that is without problems either, like the Debian/Mozilla
controversy over Firefox trademark.

Regards,
Khaled

[Pablo Impallari]

Post by Victor Gaultney
My personal opinion - not that of SIL International or an OFL maintainer,
- declare RFNs , both to protect our reputations and keep poor

derivatives from polluting the web fonts namespace, and

Post by Victor Gaultney
- be very generous with giving agreements to web font vendors that we

trust, and not try to turn it into an economic game where we try to seek
payment or other rewards in exchange for our RFNs.

Thanks Victor. I think that is exactly what I will do... keep using the
RFN's and being very generous giving agreements to the ones who ask for
them (Just for the record: Only Google, but no one else has yet asked).

Wishful thinking: I'm also hoping/expecting that Adobe or MT will be
generous too. In the same way as the small design agencies or the individual
users who donate everyday.

Cheers,
Pablo

Post by Victor Gaultney

Post by Dave Crossland

Post by Khaled Hosny
The other half is the artistic integrity,
which I, obviously, find it all nonsense

Since many libre software projects use trademarks to maintain the
integrity of their name, I don't find the
artistic-work/functional-work distinction relevant here.

I’d not say many, only a few big ones who think there names are so
valuable, and that is without problems either, like the Debian/Mozilla
controversy over Firefox trademark.
Regards,
Khaled

[Victor Gaultney]

Wishful thinking: I'm also hoping/expecting that Adobe or MT will be generous too. In the same way as the small design agencies or the individual users who donate everyday.

But keep in mind that you released your font freely, and so unless you would be providing further value-added enhancements yourself there is no reason for web fonts services to pay you simply for RFN use. I don't think that any service needs to do that.

Victor

[Vernon Adams]

I challenge Khaleds slightly 'straw man' comment :)

Fonts are functional utilities that include 'art work'. The fact that the 'art work' of fonts may not be part of your interest in them, does not separate function and art in fonts.
Ps - art work is functional utility too. It's a subset of the wider group of 'functional utilities'.

Having said that i probably agree with the general idea though; art works should be no more or less protected than any other utility.

Post by Khaled Hosny
The other half is the artistic integrity,
which I, obviously, find it all nonsense, fonts are functional utilities
not works of art, if I’m making a piece of art I’d use CC-By-NC-ND
or something like that.

[Khaled Hosny]

Post by Vernon Adams
I challenge Khaleds slightly 'straw man' comment :)
Fonts are functional utilities that include 'art work'. The fact that
the 'art work' of fonts may not be part of your interest in them, does
not separate function and art in fonts.

I did not say fonts have no artistic side (that would be a very silly
thing to say), but what I’m saying is that fonts are not made and
consumed purely for there artistic values; a font is not a painting of
Da Vinci or a calligraphy of al-Amidi that is possessed for being as
such and one need to be 100% sure what we has is an authentic
unadulterated, work.

Fonts are used for the functional value they provide and that value is
self evident; when I get my hands on a copy of, say, Amiri font what
value do I gain by knowing it is Khaled’s original or an modified
version by Vern? All I need to know is if the font I have in my hands
fulfils the purpose I need it for, and that I can tell from the font it
self, but knowing who designed it and whether it was modified or not
tells me nothing that I really need to know.

Regards,
Khaled

[Dave Crossland]

Hi!

Post by Vernon Adams
The RFN can have an integral role in how a designer can preserve
or enhance a certain type of freedom for a font. Or it simply restrict
font's freedom. I'm still arguing with myself about it :)

:)

Post by Vernon Adams
Reserving the name of the font, sets down a licensing condition
that must be met. If that condition is not met then the license is breached.
This is clear when dealing with 1 or 2 a large corporations (who may
not be interested in preserving the font's freedom); as it gives 3 clear
solutions for them to use their own modified version of a font;
1. Follow the OFL, use the font, change the font's name , and preserve the font's freedom.
2. Follow the OFL, use the font, get an agreement from me to use the
RFN, and preserve the font's freedom.
3. Buy an embedding license from me, use the font non-free, and i
preserve the font's freedom.

4. Don't use the fonts, because this is all too complicated.

Post by Vernon Adams
Now, with a few large corporations, this is highly manageable.

I am not sure about that.

Post by Vernon Adams
But what happens with the mass of individual users and or small
businesses, who maybe are also making modifications and serving the font ?
Of course, i can simply decide to ignore the potential mass of
individual breaches of non-changing of RFN's, and instead simply
focus on the few 'major' breaches. Bit of a license fudge tho imo.

Are you aware of http://en.wikipedia.org/wiki/Estoppel ? Allowing
infringement can make it hard to complain about infringement.

Post by Vernon Adams
The argument i have with myself is; why do i feel the use of RFN's
is not necessary when dealing with masses of individual users,
but i feel i want it there in case of corporate users?
It could be that i see that corporate users could easilly afford to
buy (modestly priced) RFN agreements from me (if they need to
use the fonts), thus 'giving something back' to the designer of
the free fonts they are using, and funding future fonts.

Asserting a trademark will carry the same requirement for corporate
users to license the trademark, and corporations already have well
established processes and budgets for licensing trademarks. OFL-RFN is
unusual, complex, and less likely to open that (meagre) jackpot.

Cheers
Dave

[Vernon Adams]

Post by Dave Crossland

Post by Vernon Adams
Reserving the name of the font, sets down a licensing condition
that must be met. If that condition is not met then the license is breached.
This is clear when dealing with 1 or 2 a large corporations (who may
not be interested in preserving the font's freedom); as it gives 3 clear
solutions for them to use their own modified version of a font;
1. Follow the OFL, use the font, change the font's name , and preserve
the font's freedom.
2. Follow the OFL, use the font, get an agreement from me to use the
RFN, and preserve the font's freedom.
3. Buy an embedding license from me, use the font non-free, and i
preserve the font's freedom.

4. Don't use the fonts, because this is all too complicated.

Yes, of course. "Don't use it" is assumed. I hope we would all prefer (4) to the font being used outside of the terms of the OFL?
Ps is it *really* too complicated?

Post by Dave Crossland

Post by Vernon Adams
Now, with a few large corporations, this is highly manageable.

I am not sure about that.

Neither am i :)

Post by Dave Crossland

Post by Vernon Adams
But what happens with the mass of individual users and or small
businesses, who maybe are also making modifications and serving the font ?
Of course, i can simply decide to ignore the potential mass of
individual breaches of non-changing of RFN's, and instead simply
focus on the few 'major' breaches. Bit of a license fudge tho imo.

Are you aware of http://en.wikipedia.org/wiki/Estoppel ? Allowing
infringement can make it hard to complain about infringement.

Yep

Post by Dave Crossland

Post by Vernon Adams
The argument i have with myself is; why do i feel the use of RFN's
is not necessary when dealing with masses of individual users,
but i feel i want it there in case of corporate users?
It could be that i see that corporate users could easilly afford to
buy (modestly priced) RFN agreements from me (if they need to
use the fonts), thus 'giving something back' to the designer of
the free fonts they are using, and funding future fonts.

Asserting a trademark will carry the same requirement for corporate
users to license the trademark, and corporations already have well
established processes and budgets for licensing trademarks. OFL-RFN is
unusual, complex, and less likely to open that (meagre) jackpot.

Yes. Like i said i'm still grappling and questioning myself about all this.

Post by Dave Crossland
Cheers
Dave

[Victor Gaultney]

Post by Pablo Impallari

Post by Vernon Adams
You know, we have the situation at present where at least one major corp is using my fonts in their services
They are modifying and using the RFN without agreement from me.

I have the same concerns as Vernon.
I'm getting the feeling that removing the RFM will allow Adobe and MT to do whatever they want without agreement from us.

Yes it would. And Joe's Web Fonts, and Jim's Free Fonts, and James' Funky Fonts and *anyone else*.

Keep in mind that we've already given all those people the freedom to do whatever they want. We have not hindered their freedom to modify, use, study, redistribute, etc… the font in any way. That is why the OFL is a valid and approved FOSS license. I'm proud to have fonts that I've designed released in this way.

If we declare RFNs we're saying only that people can't identify the result of their changes with the font names we've chosen and reserved - the ones that identify them with *me*. That is not unreasonable at all, and is certainly not a dilution of or an attack on freedom.

If you want to read the position of the OFL maintainers on RFNs please see the FAQ.

My personal opinion - not that of SIL International or an OFL maintainer, but as an independent font designer - is that we should:

- declare RFNs , both to protect our reputations and keep poor derivatives from polluting the web fonts namespace, and

- be very generous with giving agreements to web font vendors that we trust, and not try to turn it into an economic game where we try to seek payment or other rewards in exchange for our RFNs.

We chose to use the OFL so that our fonts could be freely available to everyone and through anyone, and we declare RFNs to make sure that users get - freely - what they expect from fonts that we create.

I hope we can be very friendly and helpful to all the web font services, but still hold them to some reasonable standards.

Thanks,

Victor

[Nicolas Mailhot]

Post by Vernon Adams

Post by Vernon Adams
You know, we have the situation at present where at least one major corp

is using my fonts in their services

Post by Vernon Adams
They are modifying and using the RFN without agreement from me.

I have the same concerns as Vernon.
I'm getting the feeling that removing the RFM will allow Adobe and MT to
do whatever they want without agreement from us.

The problem here is that the OFL is a weak free-software-like license

In a strong free software licensing model, Adobe and MT could do whatever
they want without agreement, but you'd always have the possibility to take
back what they did and do whatever you want with their work too without
their agreement. The OFL removed many of the classic clauses that made it
easy to take back a modified work.

Quite honestly, I think any legal clause aiming to prevent a bigcorp from
doing whatever they want with your work is an illusion. Bigcorp can always
pay more lawyers than you. It's much more realistic to make sure you get
something out of the deal to.

[Dave Crossland]

Post by Denis Jacquerye
With some free software you have the freedom to sell it, to

Generally FOSS means freedom to sell it for any purpose. That the OFL
restricts selling and that the FSF and OSI have approved it as a free
software/open source license is surprising, and leads to a
pre-magna-carta logic; since those 2 groups define what FOSS is, and
they say OFL is FOSS, then OFL is FOSS.

[MJ Ray]

Post by Dave Crossland

Post by Denis Jacquerye
With some free software you have the freedom to sell it, to

Generally FOSS means freedom to sell it for any purpose. That the OFL
restricts selling and that the FSF and OSI have approved it as a free
software/open source license is surprising, and leads to a
pre-magna-carta logic; since those 2 groups define what FOSS is, and
they say OFL is FOSS, then OFL is FOSS.

Based on my past dealings with them, I seriously doubt that the FSF
would approve a non-program as a free software licence. They use a
weaker standard for non-program software like fonts, much to my
disappointment.

The OSI process is rather cumbersome and I don't think it produces the
right answer every time. Even for programs, their list used to have
some differences with FSF.

I think Debian (one of the larger public review processes) accepts the
OFL because it believes the sale restriction is useless and can be
avoided by packaging the font for sale with a "hello world".

Hope that helps,

[Dave Crossland]

Post by Vernon Adams
Not sure i understand 100% what you are saying :)
specifically - "Free software can be charged for; otherwise it is not free as in
freedom"
is there a 'not' missing?

No :)

Free software can be sold. Freedom to redistribute for a fee is one of
the FSF core 4 freedoms.

However, free software is not distributed for a fee 99.99% of the
time, and that's because after selling a copy, the buyer can put it on
the net and redistribute it without a fee if they wish to, so the
original distributor usually just cuts to the chase and becomes the
canonical source for the work.

[Nathan Willis]

I'm _generally_ in favor of pursuing the trademark front as an alternative
to RFN (although obviously at this point it's a bit experimental). I
wonder, though, if recommending trademarks would be problematic because of
the differences in trademark law between jurisdictions.

I don't know global TM law; maybe it's so convoluted that ever country
would require its own FAQ. But maybe it's not....

Nate

[Eric Schrijver]

Post by Nathan Willis
I'm _generally_ in favor of pursuing the trademark front as an
alternative to RFN (although obviously at this point it's a bit
experimental). I wonder, though, if recommending trademarks would be
problematic because of the differences in trademark law between
jurisdictions.
I don't know global TM law; maybe it's so convoluted that ever country
would require its own FAQ. But maybe it's not...

Hey girls and guys,

just some practical points:

the RFN works great against a usual argument of type designers who are
afraid that open sourcing their work will lead to forks of ‘questionable
quality’—you can say that they have control over ‘their’ version of the
font because other fonts need a distinguishly different name.

the RFN doesn’t work great in terms of the Git/github model of
development, because basically in the cycle forking, modifying,
submitting a pull request, you are already violating the RFN.

At the same time, the fact that there are many versions of one artwork /
typeface / software with the same name can be difficult. I think for
instance of the many versions of GitX floating around. That’s where I
like the RFN as well.

In theory, I would like a soft RFN that forces you to change the name if
you don’t intend to contribute your changes back to the font you forked. [1]

Trademarks, on the other hand, to me seem pretty ludicrous, but I might
be misinformed. As far as I know, they cost money, and require active
litigation to keep them valid (i.e., contrary to copyright, if you don’t
activily sue for trademark infringements you risk to loose the trademark
protection).
Can someone outline how asking for a trademark would work for an
indepent Font designer and how it would benefit them?

Good night!

xx
E

[1] BTW, hard forks, forks that don’t contribute back to the thing
forked from, seem more desirable in fonts then in software. Like Gill is
a fork of Johnston, or Helvetica of Akzidensk—that’s cool, and something
made a lot more easy by an ecosystem of OFL fonts.

[Nathan Willis]

Post by Eric Schrijver

Post by Nathan Willis
I'm _generally_ in favor of pursuing the trademark front as an
alternative to RFN (although obviously at this point it's a bit
experimental). I wonder, though, if recommending trademarks would be
problematic because of the differences in trademark law between
jurisdictions.
I don't know global TM law; maybe it's so convoluted that ever country
would require its own FAQ. But maybe it's not...

Hey girls and guys,
the RFN works great against a usual argument of type designers who are
afraid that open sourcing their work will lead to forks of ‘questionable
quality’—you can say that they have control over ‘their’ version of the
font because other fonts need a distinguishly different name.
the RFN doesn’t work great in terms of the Git/github model of
development, because basically in the cycle forking, modifying, submitting
a pull request, you are already violating the RFN.
At the same time, the fact that there are many versions of one artwork /
typeface / software with the same name can be difficult. I think for
instance of the many versions of GitX floating around. That’s where I like
the RFN as well.
In theory, I would like a soft RFN that forces you to change the name if
you don’t intend to contribute your changes back to the font you forked. [1]
Trademarks, on the other hand, to me seem pretty ludicrous, but I might be
misinformed. As far as I know, they cost money, and require active
litigation to keep them valid (i.e., contrary to copyright, if you don’t
activily sue for trademark infringements you risk to loose the trademark
protection).
Can someone outline how asking for a trademark would work for an indepent
Font designer and how it would benefit them?

In the US (at least), trademarks do not cost money. Like copyright is
"automatic" when you create a work, trademark is "automatic" when you
advertise / publicize your brand. However, also like copyright, there is a
"registered trademark" option that DOES cost money, and lets you somehow
get more legal muscle. The specifics there I don't recall.

As for litigation, litigation is only necessary if someone violates your
trademark and doesn't stop when you ask them to. That's not different that
copyright either. Worst-case scenario in copyright is a lawsuit; trademark
is the same.

What I think *is* missing and needed is a clear "trademark usage
guideline." The author gets to set those rules; they would be analogous to
a copyright license -- they set out your rules for the trademark, as the
OFL sets out the license for the work itself. But so far there's not a set
of discrete boilerplate TM-usage-docs akin to (for example) the CC
licenses. Having some simple options for that might make deploying a
trademark on your font a less confusing option.

At least in the US. Still don't know about anywhere else.

Nate

Post by Eric Schrijver
Good night!
xx
E
[1] BTW, hard forks, forks that don’t contribute back to the thing forked
from, seem more desirable in fonts then in software. Like Gill is a fork of
Johnston, or Helvetica of Akzidensk—that’s cool, and something made a lot
more easy by an ecosystem of OFL fonts.

[Vernon Adams]

Post by Dave Crossland

Post by Vernon Adams
Not sure i understand 100% what you are saying :)
specifically - "Free software can be charged for; otherwise it is not free as in
freedom"
is there a 'not' missing?

No :)
Free software can be sold. Freedom to redistribute for a fee is one of
the FSF core 4 freedoms.

Yes i know. I couldn't get my head around the sentence structure (was a late night), plus, the OFL does disallow selling (under specific circumstances). Hence my confusion at what Barry was saying exactly. I see it now.

[Vernon Adams]

Oh i see what you are saying now :)
Yes of course.
But the OFL prevents that particular freedom.
I've never understood the rationale of that aspect of the OFL.
Are you suggesting that restriction be removed from the OFL?

-vern

Post by Barry Schwartz
Free software can be charged for; otherwise it is not free as in
freedom. Once you pay for it, if it is copylefted, _then_ you can
redistribute, etc.
So that’s not really a ‘situation’ unless they try to prevent the
second part.

[Vernon Adams]

Post by Dave Crossland
Hi,

Post by Vernon Adams
Or, is there something i am not understanding in this? :)

I think its unreasonable to expect every person publishing a blog who
makes their own subset to contact every copyright holder every time
they want to use a new OFL-RFN web font.

But isn't it exactly for that scenario that a designer will publish without an RFN?
If a designer wishes to make a font available under the OFL without RFN-caused restrictions on subsetting etc then he/she will publish without an RFN. Not difficult :)
Still not seeing why no RFN needs to be the default position. I'm very open to persuasive argument tho.

Post by Dave Crossland
I think the desirable effect of RFNs can be preserved with a trademark
license, while alleviating the undesirable effect of prohibiting
subsetting and other trivial changes while retaining the name users
are familiar with.

Any chance you can give an 'idiots guide' on how a trademark license would preserved some of the effects of the RFN? I'm unsure how and when this trademark license would work? What would it be aimed at preventing?

cheers

vernon

Post by Dave Crossland
I note that SIL itself also asserts both RFN and trademark rights on
its font names; so if you are keen to maximize control over your font
names, it may make sense to do what I'm suggestion in addition to
RFNs, rather than replacing them.
Cheers
Dave
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups "Google Font Directory Discussions" group.
For more options, visit https://groups.google.com/groups/opt_out.

[Dave Crossland]

Post by Vernon Adams
Any chance you can give an 'idiots guide' on how a trademark license
would preserved some of the effects of the RFN? I'm unsure how
and when this trademark license would work? What would it be
aimed at preventing?

If you have no RFN, someone can make subsets (yay) but also release
Oswald Serif without additional permission (hmm) - but if you have a
trademark, they can subset Oswald but can't release Oswald Serif.

Please consider
http://svn-qavm.apache.org/repos/svn-org/trunk/legal/trademark-policy.html

[Barry Schwartz]

Post by Dave Crossland

Post by Vernon Adams
Any chance you can give an 'idiots guide' on how a trademark license
would preserved some of the effects of the RFN? I'm unsure how
and when this trademark license would work? What would it be
aimed at preventing?

If you have no RFN, someone can make subsets (yay) but also release
Oswald Serif without additional permission (hmm) - but if you have a
trademark, they can subset Oswald but can't release Oswald Serif.
Please consider
http://svn-qavm.apache.org/repos/svn-org/trunk/legal/trademark-policy.html

I have long thought and continue to think that RFNs are way more
trouble than they are worth. It makes the OFL look complicated and
frightening, which is the opposite of what should be the goal. Plus,
if someone intends to give a font a different name, they don’t need to
be told to do it; and, if they do not intend to, they are not going to
corrupt society to the core. The worst that will happen is you’ll have
to be careful where you got the font.

The rest of the software community has managed to get along for
decades without having everyone give their version of ‘ls’ a different
name. It creates problems, big ones, but the alternative is worse.

Trademark is a related but different issue.

[Khaled Hosny]

Post by Vernon Adams
Still not seeing why no RFN needs to be the default position. I'm very
open to persuasive argument tho.

Because almost all free software licenses do not have such a provision
(I can only think of TeX license, which is pre historical, and the LPPL
which is apparently influenced by TeX license), so what is so special
about fonts in the first place? Incidentally I removed RFN from my fonts
few weeks ago because I felt it is just a silly provision and I have no
idea why I went for it in the first place :)

Regards,
Khaled

[Khaled Hosny]

Post by Khaled Hosny

Post by Vernon Adams
Still not seeing why no RFN needs to be the default position. I'm very
open to persuasive argument tho.

Because almost all free software licenses do not have such a provision
(I can only think of TeX license, which is pre historical, and the LPPL
which is apparently influenced by TeX license), so what is so special
about fonts in the first place? Incidentally I removed RFN from my fonts
few weeks ago because I felt it is just a silly provision and I have no
idea why I went for it in the first place :)

Actually people were making modified versions of the said font even with
the RFN, and telling those people they are violating the license seemed
so harsh so I didn’t bother.

Regards,
Khaled

[Vernon Adams]

Is it not possible to include with a font, alongside the OFL, a 'pre-emptive permission' text that gives the user the go-ahead to use the RFN named in OFL text when a font has been modified by subsetting, re-hinting, etc, (i would have to think of the full list) ?? Seems a sensible solution to me.

-vernon

I had assumed that including the RFN as part of the new name would be inappropriate. I suppose it could depend on whether the RFN was a trademark or not. The OFL is rather silent in this area.

It's up to the copyright holder. Some would *love* to have the RFNs used in derivative names by those they trust, and could even require that in their permission agreement. Others want the opposite. The OFL does not speak to this other than to say "if you modify this font, and the author has declared an RFN, talk to them and work out together what is best for this specific situation".
Personally I love the idea of web fonts services (or frankly any service or individual that modifies fonts to improve them) appending a recognisable suffix to the name. This will still require separate permission if RFNs are declared, but would go a long way to avoiding the confusion and frustration that Adam has so colourfully described. It would also show a value-added quality to the font. For example, say Paratype were to take our Gentium Plus font and hint the Roman and Cyrillic to look awesome on Android. With our permission they could call it 'Gentium Plus PT' and people would flock to them to get their improved derivative (hey - we'd even use it!). The world would also then think - hmmmm maybe we could hire them to do the same wonders for ours, or to begin to trust 'PT' derivatives over others.
Have a great weekend,
V
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups "Google Font Directory Discussions" group.
For more options, visit https://groups.google.com/groups/opt_out.

[Vernon Adams]

Post by Vernon Adams
Is it not possible to include with a font, alongside the OFL, a 'pre-emptive permission' text that gives the user the go-ahead to use the RFN named in OFL text when a font has been modified by subsetting, re-hinting, etc, (i would have to think of the full list) ?? Seems a sensible solution to me.

A copyright holder can certainly do something like that. The OFL doesn't prohibit or discourage it. Such are really very separate permission agreements.
The main caveat is that the copyright holder should be sure they are very clear about the specific types of modifications that are allowed (and not allowed), and how they will be judged. IOW - I think that to give others a blanket permission to change hinting is dangerous. It would be better to allow hinting *if the results is generally more readable than the original*. Otherwise you could have some service that cares for only one platform (that ignores hints anyway) completely blow away your careful hinting and make your font look terrible on Windows.

Yes. I am suggesting that a generic text could be used that gives permission for those specific tasks that are commonly carried out when a font is used within a 'webfont server' service, and in a service such as Font Squirrel. Perhaps the authors of the OFL could create such a text? A sort of 'engineering exception' that can be voluntarilly used with the OFL? I really don't think 'standards' need to come into it, and they are too complicated to define.

So according to Adobe, they do the following with the OFL'd fonts they use in their Edge service;

• NAME table changes, including stripping of some attribution metadata and obfuscation of Name related entries.
• Injecting a EULA link that lives at typekit.com, which displays the attribution metadata in an easily discoverable location.
• format conversions: TTF (if the original is a CFF), SVG, SWF, WOFF, EOT.
• subset creation: we currently offer our users two subsets: "default" (omits everything but Latin-1 plus some useful typographic marks) and "all" (includes everything the font has to offer).
• miscellaneous updates related to web delivery, e.g., adjustments to CFF table's dictionary and index data in order to accommodate changes to data location within the table resulting from subsetting; vertical metrics adjustments for cross-browser consistency.

I would say that these type of modifications are of a different nature to the type of modifications that i think the RFN is really aimed at. I see the RFN as a way of way to protect the 'stylistic aspect' of a font within the freedoms to modify and derivate.
I read the RFN part of the OFL, not as a non-permission, but more as a way of imposing a 'common sense naming policy' when naming a derivative (and stylistically 'different') font from an already existing font. These engineering modifications do not effect the stylstic aspect of a font's visual appearance.

-vernon

If you're going to give people wide and generous permissions with no standards and no recourse, then why declare RFNs at all?
V
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups "Google Font Directory Discussions" group.
For more options, visit https://groups.google.com/groups/opt_out.

[Dave Crossland]

Post by Vernon Adams
Perhaps the authors of the OFL could create such a text?

I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this, and I would not like to see such
additional permissions to the OFL floating around because I know that
software corporation's legal departments (ie, HP) consider "license
plus additional permission notice" to be a wholly discrete, custom
copyright license which immediately rules it out of consideration for
their use because they have policies against license proliferation -
and for good reason, because such a bespoke license is untested and
carries a lot more legal risk.

Software corporations _do_ have standard business processes for
handling trademark agreements though and that is why I am so keen on
them. Trademarks allow us to stop using RFNs but retain the kind of
name control that you want.

Post by Vernon Adams
These engineering modifications do not effect the stylstic aspect of a font's visual appearance.

While I share your position, as Adam and Victor have said, they don't
see these changes as minor and advocate changing the name even when
such changes are made.

I think imposing a 'common sense naming policy' when naming derivative
(and stylistically 'different') fonts from an already existing font
should be done with trademarks, not copyrights.

Since SIL and Canonical and Adobe assert trademarks and require
separate trademark agreements with web font service providers anyway,
I am more leaning towards making a simple web app that makes it easy
to get RFN permission to do these things.

[Vernon Adams]

Post by Dave Crossland
I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this, and I would not like to see such
additional permissions to the OFL floating around because I know that
software corporation's legal departments (ie, HP) consider "license
plus additional permission notice" to be a wholly discrete, custom
copyright license which immediately rules it out of consideration for
their use because they have policies against license proliferation -
and for good reason, because such a bespoke license is untested and
carries a lot more legal risk.

But Adobe seem more than able to handle this 'additional permission' right here and now. So i really can't imagine that if this 'additional permission' had already existed within (or alongside) the OFL, that Adobe would have said 'whoa guys, don't go near those OFL fonts!!'. Or would they?

-v

[Dave Crossland]

Post by Vernon Adams

Post by Dave Crossland
I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this, and I would not like to see such
additional permissions to the OFL floating around because I know that
software corporation's legal departments (ie, HP) consider "license
plus additional permission notice" to be a wholly discrete, custom
copyright license which immediately rules it out of consideration for
their use because they have policies against license proliferation -
and for good reason, because such a bespoke license is untested and
carries a lot more legal risk.

But Adobe seem more than able to handle this 'additional permission' right here and now. So i really can't imagine that if this 'additional permission' had already existed within (or alongside) the OFL, that Adobe would have said 'whoa guys, don't go near those OFL fonts!!'. Or would they?

I know that anyone dealing with free software in software corporations
must get their legal departments to review a 'well known license +
other licensing text' license as a new, discrete license, and that
doing that can be (a) hard to get legal's attention in the first place
and (b) they are unlikely to be happy about it because license
profileration is bad.

Maybe the legal department will wave it though. Maybe not.

[Pablo Impallari]

A recent real-life example, the Rosario font by Omnibus-Type, was
hand-hinted by Adobe.
http://blogs.adobe.com/edgewebfonts/2013/05/23/adobe-contributes-improvements-to-open-source-font-family-rosario/
http://googlewebfonts.blogspot.com.ar/2013/05/typekit-improves-rosario.html
That's great! (Pablo Cosgaya granted them permission to use the RFN's.)

But... if you look at https://typekit.com/fonts/rosario or at the pages
having the license https://typekit.com/eulas/000000000000000000014188 they
make NO MENTION of Omnibus-Type whatsoever.

Post by Vernon Adams

Post by Vernon Adams

Post by Dave Crossland
I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this, and I would not like to see such
additional permissions to the OFL floating around because I know that
software corporation's legal departments (ie, HP) consider "license
plus additional permission notice" to be a wholly discrete, custom
copyright license which immediately rules it out of consideration for
their use because they have policies against license proliferation -
and for good reason, because such a bespoke license is untested and
carries a lot more legal risk.

But Adobe seem more than able to handle this 'additional permission'

right here and now. So i really can't imagine that if this 'additional
permission' had already existed within (or alongside) the OFL, that Adobe
would have said 'whoa guys, don't go near those OFL fonts!!'. Or would they?
I know that anyone dealing with free software in software corporations
must get their legal departments to review a 'well known license +
other licensing text' license as a new, discrete license, and that
doing that can be (a) hard to get legal's attention in the first place
and (b) they are unlikely to be happy about it because license
profileration is bad.
Maybe the legal department will wave it though. Maybe not.
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups
"Google Font Directory Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an
For more options, visit https://groups.google.com/groups/opt_out.

[Vernon Adams]

Not sure this is about RFN's, (except that within the situation permission to use the RFN was granted).
I think this is more to do with the way that Adobe / Typekit have approached using the OFL'd fonts from Google Webfonts.
-v

A recent real-life example, the Rosario font by Omnibus-Type, was hand-hinted by Adobe.
http://blogs.adobe.com/edgewebfonts/2013/05/23/adobe-contributes-improvements-to-open-source-font-family-rosario/
http://googlewebfonts.blogspot.com.ar/2013/05/typekit-improves-rosario.html
That's great! (Pablo Cosgaya granted them permission to use the RFN's.)
But... if you look at https://typekit.com/fonts/rosario or at the pages having the license https://typekit.com/eulas/000000000000000000014188 they make NO MENTION of Omnibus-Type whatsoever.

Post by Vernon Adams

Post by Dave Crossland
I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this, and I would not like to see such
additional permissions to the OFL floating around because I know that
software corporation's legal departments (ie, HP) consider "license
plus additional permission notice" to be a wholly discrete, custom
copyright license which immediately rules it out of consideration for
their use because they have policies against license proliferation -
and for good reason, because such a bespoke license is untested and
carries a lot more legal risk.

But Adobe seem more than able to handle this 'additional permission' right here and now. So i really can't imagine that if this 'additional permission' had already existed within (or alongside) the OFL, that Adobe would have said 'whoa guys, don't go near those OFL fonts!!'. Or would they?

I know that anyone dealing with free software in software corporations
must get their legal departments to review a 'well known license +
other licensing text' license as a new, discrete license, and that
doing that can be (a) hard to get legal's attention in the first place
and (b) they are unlikely to be happy about it because license
profileration is bad.
Maybe the legal department will wave it though. Maybe not.
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups "Google Font Directory Discussions" group.
For more options, visit https://groups.google.com/groups/opt_out.
--
Un Abrazo
Pablo Impallari
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google Groups "Google Font Directory Discussions" group.
For more options, visit https://groups.google.com/groups/opt_out.

[Pablo Impallari]

It may be, but still the fact that they omit to credit the designers smells
bad to me.

When you browse commercial fonts on Typekit, they display the Foundries
names in a prominent way.
When you browse libre fonts on Edge, they have chosen not to show them at
first glance. Instead they are somewhat hidden, and you only get to see it
if you click on the small "license" link.

In the case of Rosario, since it links to the same page as TK, you never
get to see even a small mention of Omnibus-type.

Post by Vernon Adams
Not sure this is about RFN's, (except that within the situation permission
to use the RFN was granted).
I think this is more to do with the way that Adobe / Typekit have
approached using the OFL'd fonts from Google Webfonts.
-v

Post by Pablo Impallari
A recent real-life example, the Rosario font by Omnibus-Type, was

hand-hinted by Adobe.
http://blogs.adobe.com/edgewebfonts/2013/05/23/adobe-contributes-improvements-to-open-source-font-family-rosario/
http://googlewebfonts.blogspot.com.ar/2013/05/typekit-improves-rosario.html

Post by Pablo Impallari
That's great! (Pablo Cosgaya granted them permission to use the RFN's.)
But... if you look at https://typekit.com/fonts/rosario or at the pages

having the license https://typekit.com/eulas/000000000000000000014188they make NO MENTION of Omnibus-Type whatsoever.

Post by Pablo Impallari

Post by Vernon Adams

Post by Dave Crossland
I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this, and I would not like to see such
additional permissions to the OFL floating around because I know that
software corporation's legal departments (ie, HP) consider "license
plus additional permission notice" to be a wholly discrete, custom
copyright license which immediately rules it out of consideration for
their use because they have policies against license proliferation -
and for good reason, because such a bespoke license is untested and
carries a lot more legal risk.

But Adobe seem more than able to handle this 'additional permission'

right here and now. So i really can't imagine that if this 'additional
permission' had already existed within (or alongside) the OFL, that Adobe
would have said 'whoa guys, don't go near those OFL fonts!!'. Or would they?

Post by Pablo Impallari
I know that anyone dealing with free software in software corporations
must get their legal departments to review a 'well known license +
other licensing text' license as a new, discrete license, and that
doing that can be (a) hard to get legal's attention in the first place
and (b) they are unlikely to be happy about it because license
profileration is bad.
Maybe the legal department will wave it though. Maybe not.
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google

Groups "Google Font Directory Discussions" group.

Post by Pablo Impallari
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/groups/opt_out.
--
Un Abrazo
Pablo Impallari
--
--
Google Font Directory Discussions
http://groups.google.com/group/googlefontdirectory-discuss
---
You received this message because you are subscribed to the Google

Groups "Google Font Directory Discussions" group.

Post by Pablo Impallari
To unsubscribe from this group and stop receiving emails from it, send
For more options, visit https://groups.google.com/groups/opt_out.

[Victor Gaultney]

Post by Dave Crossland

Post by Vernon Adams
Perhaps the authors of the OFL could create such a text?

I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this,

Yes - for the reasons Dave mentions, and the basic conceptual difficulty of defining and evaluating what changes would be allowed.

V

[Vernon Adams]

I can understand this, except for one thing;

Surely it would not be 'diluting' the OFL to reshape it to bring more clarity to the licensing of this whole 'minor modification' space that webfont services are opening up?
Imo the OFL needs to be ever so slightly tweaked, but only to better protect the freedom of OFL'd fonts. That's not a dilution, that's a re-concentration.

On the other hand, expecting designers to rely on an external triggers such as 'trademarks' to plug this issue, does seem to dilute the license.

-vernon

Post by Victor Gaultney

Post by Dave Crossland

Post by Vernon Adams
Perhaps the authors of the OFL could create such a text?

I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this,

Yes - for the reasons Dave mentions, and the basic conceptual difficulty of defining and evaluating what changes would be allowed.

[Dave Crossland]

The ofl has no upgrade model, the authors have no resources to make an
update, and believe the license provides for this situation.

I will try to make getting addition permission privately a convenient
process for those who must seek such permission, for rfns and trademarks.

Post by Vernon Adams
I can understand this, except for one thing;
Surely it would not be 'diluting' the OFL to reshape it to bring more
clarity to the licensing of this whole 'minor modification' space that
webfont services are opening up?
Imo the OFL needs to be ever so slightly tweaked, but only to better
protect the freedom of OFL'd fonts. That's not a dilution, that's a
re-concentration.
On the other hand, expecting designers to rely on an external triggers
such as 'trademarks' to plug this issue, does seem to dilute the license.
-vernon

Post by Victor Gaultney

Post by Dave Crossland

Post by Vernon Adams
Perhaps the authors of the OFL could create such a text?

I think Victor has been quite clear that he's not at all interested in
diluting the OFL model like this,

Yes - for the reasons Dave mentions, and the basic conceptual difficulty

of defining and evaluating what changes would be allowed.

[Nathan Willis]

Post by Dave Crossland
The ofl has no upgrade model, the authors have no resources to make an
update, and believe the license provides for this situation.

Just curious: what would be an "upgrade model" and what "resources"? I do
certainly appreciate the non-profit nature of SIL and how busy everyone
is. But AFAIK, when other free software licenses have been upgraded, the
community has participated, and it sure seems like there are a lot of
interested parties who are willing to comment on these FAQ changes and
white paper. And it seems rare that a license would be upgraded on any
sort of schedule; it's usually undertaken only when a real-world issue
triggers it or a really long time has elapsed.

Ultimately, that's orthogonal to the third reason mentioned, of course.
OFL is their license; no question about that, and it is and should be their
call.

Nate

[Dave Crossland]

Post by Nathan Willis

Post by Dave Crossland
The ofl has no upgrade model, the authors have no resources to make an
update, and believe the license provides for this situation.

Just curious: what would be an "upgrade model"

The GNU project suggest people declare 'This software is under GPLvX
or any later version' but the 'or any later version' isn't in the
license text itself. It is in all the CC licenses.

Post by Nathan Willis
and what "resources"?

Staff time, lawyer time.

Post by Nathan Willis
I do certainly appreciate the non-profit nature of SIL and how busy everyone is.
But AFAIK, when other free software licenses have been upgraded, the
community has participated, and it sure seems like there are a lot of
interested parties who are willing to comment on these FAQ changes and white
paper. And it seems rare that a license would be upgraded on any sort of
schedule; it's usually undertaken only when a real-world issue triggers it
or a really long time has elapsed.
Ultimately, that's orthogonal to the third reason mentioned, of course. OFL
is their license; no question about that, and it is and should be their
call.

:)

[Khaled Hosny]

Post by Nathan Willis

Post by Dave Crossland
The ofl has no upgrade model, the authors have no resources to make an
update, and believe the license provides for this situation.

Just curious: what would be an "upgrade model"

Like GPL's "or any later version". If OFL, say, v1.2 would be written
today, every OFL-using font will need to be explicitly re-licensed under
it, further more we might not be able to mix OFLv1.1 licensed fonts with
OFLv1.2 licensed ones, which pretty much defeat the whole purpose of
having a unified libre font license (AFAIC anyway).

Regards,
Khaled

[Victor Gaultney]

The ofl has no upgrade model, the authors have no resources to make an update, and believe the license provides for this situation.

I agree that the OFL 1.1 serves us well for now. Because of the huge undertaking it would be to form a 1.2 and get it generally approved and accepted, and the pain it would cause to the whole FOSS font world, there would need to be a very compelling, long-term, philosophical reason to craft a new license (and model) and pay that price. Even for licenses where there is an 'upgrade clause' such upgrades are far from smooth. GPLv3 anyone?

Web fonts are important, but are (in the longer term) an ephemeral issue. It is one of many challenges, past and future, that the OFL will face. At some point in the future there may be a need for a license change, but I don't see any reason for that now. OFL 1.1 handles web fonts just fine. RFNs and trademark issues are not trivial, but are more of an administrative issue than something that would require key conceptual changes in the model.

In 5, 10 or 25 years, were there to be some major shift in legal precedent or understanding or in technology that would render the current OFL a poor model, then a revision might be in order. But it would need to be a reconstruction and complete rethinking - an OFL 2.0, not 1.2 - and would require just as much effort by the community as adopting the OFL in the first place.

V