Terminating Community Edition Builds of Firefox

23 views
Skip to first unread message

Catherine Brady

unread,
Sep 29, 2008, 5:14:58 PM9/29/08
to gover...@lists.mozilla.org
 

We are in the process of updating our Trademark Policy.  One of the proposed modifications is to eliminate the Community Edition program, which people are using as a loophole to create unauthorized modified Firefox builds using the Firefox name and selling them to the unsuspecting general public.

During the beginning days of Firefox, when we did not have the bandwidth to localize Firefox, the idea of a Community Edition of Firefox was created. (See http://www.mozilla.org/foundation/trademarks/community-edition-policy.html )  The program’s intention was to allow members around the world to create localized versions of Firefox without having to obtain a license or approval from Mozilla.  However, as we continued to grow, so did our localization (“L10n”) teams and program.  Now all L10n efforts are part of the official Firefox release.  Our community is not using the Community Edition builds to create localized versions of Firefox. 

As far as we’re aware, the only people taking advantage of this legacy program are people trying to circumvent the licensing requirements of our trademark policy.  These people create a modified version of Firefox, call it a “Community Edition”, and sell this modified Firefox to make money and collect users’ personal information (including name, email address, address and credit card information).  We cannot ensure the quality of these builds nor the negative user experience created by these opportunistic builds.

Reason to Terminate Community Edition Builds of Firefox.  The following are the reasons why we believe that Community Edition builds are irrelevant today:

  • We have an established L10n program and the community no longer use the Community Edition builds to create localized versions of Firefox;
  • Community Edition builds are being used by individuals trying to circumvent the licensing requirements for modified builds of Firefox;
  • Often these Community Edition builds are based on older versions of Firefox or contain unknown additional software;
  • Often the individuals who have created these Community Edition builds represent themselves, through illegal use of our logos, as Mozilla and collect money and personal information from the users;
  • We are working to shut down fraudulent advertisers and Web sites.  Ken Kovash and David Rolnitzky have estimated that approximately 6.7 million users visit these fraudulent sites annually.
  • We cannot ensure the quality of these builds. 

The purpose of this post is to let you know that we’re thinking of terminating the Community Edition builds, set forth our reasons why, and to receive your feedback.  If you are aware of (i) any reasons why we should continue to allow these builds or (ii) anyone who is creating these builds for other than a commercial or fraudulent purpose, let me know.

Cédric Corazza

unread,
Sep 29, 2008, 5:27:53 PM9/29/08
to
Catherine Brady a écrit :

>
> The purpose of this post is to let you know that we’re thinking of
> terminating the Community Edition builds, set forth our reasons why,
> and to receive your feedback. If you are aware of (i) any reasons why
> we should continue to allow these builds or (ii) anyone who is
> creating these builds for other than a commercial or fraudulent
> purpose, let me know.
>
Though I'm not a user of Debian distro, I hope they will still have the
right to release Iceweasel and other customized Mozilla products (but
maybe these are not a community releases and follow some other agreement
or policy). It would be a really bad move imo to forbid that.

My two cents.

Mike Shaver

unread,
Sep 29, 2008, 5:32:01 PM9/29/08
to Cédric Corazza, gover...@lists.mozilla.org
On Mon, Sep 29, 2008 at 5:27 PM, Cédric Corazza
<cedric....@gmail.com> wrote:
> Though I'm not a user of Debian distro, I hope they will still have the
> right to release Iceweasel and other customized Mozilla products (but
> maybe these are not a community releases and follow some other agreement
> or policy). It would be a really bad move imo to forbid that.

Those are not affected by this at all, indeed.

Mike

Message has been deleted

Cédric Corazza

unread,
Sep 29, 2008, 5:40:08 PM9/29/08
to
Mike Shaver a écrit :
Thanks for the clarification Mike, this will kill any troll about this matter before its birth :)

Cédric

Jonas Sicking

unread,
Sep 29, 2008, 7:12:39 PM9/29/08
to

Like Mike said, this does not affect Iceweasel at all. Firefox is and
will indeed continue to be open source which means that anyone is free
to take the source and build whatever they want (within the limits of
the licenses for the source). None of that is affected by this in the least.

What this affects is using the Firefox trademarks, such as logo and name.

/ Jonas

John S. Thomsen

unread,
Sep 29, 2008, 7:44:37 PM9/29/08
to
Catherine Brady wrote:
> If you are aware of
> anyone who is creating these builds for other than a
> commercial or fraudulent purpose, let me know.

The official danish edition of Sunbird(TM) 0.7 could not import/export
calendars because a l10n file was saved in a wrong format.

I believe there was a community edition available until Sunbird 0.8

Announcement [in Danish]
http://forum.mozilladanmark.dk/viewtopic.php?t=3773

I think it was a smooth solution to a big problem for a (probably) small
group of users.

/johnst

Mook

unread,
Sep 30, 2008, 5:19:31 AM9/30/08
to
Catherine Brady wrote:
>
>
> We are in the process of updating our Trademark Policy. One of the
> proposed modifications is to eliminate the Community Edition program,
> which people are using as a loophole to create unauthorized modified
> Firefox builds using the Firefox name and selling them to the
> unsuspecting general public.
>

It might be useful to post this information in
http://forums.mozillazine.org/viewforum.php?f=42 - that's where, last I
checked, most third party builders hang out. (It's been over a year
since I tracked it, though; things may have changed.) I see some topics
there lablelled "CE"; presumably that's Community Edition.

--
Mook

Sergey Yanovich

unread,
Sep 30, 2008, 8:19:38 AM9/30/08
to
Catherine Brady wrote:
> _Reason to Terminate Community Edition Builds of Firefox_. The following are
> the reasons why we believe that Community Edition builds are irrelevant today:
>
> * [unrelated to license]
> * Community Edition builds are being used by individuals trying to

> circumvent the licensing requirements for modified builds of Firefox;
> * Often these Community Edition builds are based on older versions of

> Firefox or contain unknown additional software;
> * Often the individuals who have created these Community Edition builds

> represent themselves, through illegal use of our logos, as Mozilla and
> collect money and personal information from the users;
> * We are working to shut down fraudulent advertisers and Web sites. Ken

> Kovash and David Rolnitzky have estimated that approximately 6.7 million
> users visit these fraudulent sites annually.
> * We cannot ensure the quality of these builds.

There is a simple and cheap solution to all of the above problems. Just
relicense Firefox under GPL (de-support LGPL/MPL). This is not an
easiest task, but perfectly doable.

--
Sergey Yanovich

Gijs Kruitbosch

unread,
Sep 30, 2008, 10:36:06 AM9/30/08
to

IANAL, but isn't the GPL a license, and isn't the point of (and/or issue with)
the Community Edition branding (ergo, trademarks), rather than the license of
the code itself? As far as I know, the fact that the code can be used under the
LGPL and MPL license as well as the GPL has no bearing on the reuse of the
"Firefox" brandname, which is what most of these problems originate from: some
things/"products" are called Firefox when they really aren't, and this causes an
inferior user experience for people finding these things.

The only point I can possibly see in your point is that it would force such
Community Editions to open their source code. Which doesn't help. And in fact,
doing what you suggest now cannot retroactively be applied to previous versions
of the code (eg. Fx 2, 3, or other releases) and hence would not influence anything.

Finally, the last relicensing (tri-licensing) of the code took several years to
complete. I don't see why what you're suggesting would be any faster, and so it
would not solve anything at all in the short run.

~ Gijs

Sergey Yanovich

unread,
Sep 30, 2008, 10:57:25 AM9/30/08
to
Gijs Kruitbosch wrote:
> IANAL, but isn't the GPL a license, and isn't the point of (and/or issue
> with) the Community Edition branding (ergo, trademarks), rather than the
> license of the code itself? As far as I know, the fact that the code can
> be used under the LGPL and MPL license as well as the GPL has no bearing
> on the reuse of the "Firefox" brandname, which is what most of these
> problems originate from: some things/"products" are called Firefox when
> they really aren't, and this causes an inferior user experience for
> people finding these things.

RedHat's approach demostrates how to settle the trademark issue in a
GPLed product.

> The only point I can possibly see in your point is that it would force
> such Community Editions to open their source code. Which doesn't help.

Why it doesn't help? The problems raised in the original post were facts
of Firefox being *sold* and being *malitious*. GPL prevents selling and
requires source code to come along binaries.

GPL-violation-watch is a very popular game.

> And in fact, doing what you suggest now cannot retroactively be applied
> to previous versions of the code (eg. Fx 2, 3, or other releases) and
> hence would not influence anything.

This is true. The solution is for the future releases only.

> Finally, the last relicensing (tri-licensing) of the code took several
> years to complete. I don't see why what you're suggesting would be any
> faster

The switch can be done at the next commit. Tri-licence explicitly allows
this.

--
Sergey Yanovich

Atsushi Shimono

unread,
Sep 30, 2008, 10:57:31 AM9/30/08
to cbr...@mozilla.com, gover...@lists.mozilla.org
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

hi, list.

Catherine Brady wrote:
> _Reason to Terminate Community Edition Builds of Firefox_. The
> following are the reasons why we believe that Community Edition builds
> are irrelevant today:

I think i should argue with some of the reasons.

> * We have an established L10n program and the community no longer


> use the Community Edition builds to create localized versions of
> Firefox;

> * Community Edition builds are being used by individuals trying to
> circumvent the licensing requirements for modified builds of Firefox;

Are the community edition builds only for localizations and individuals?
I think no.

Many CE builds targets 'optimizations' but not localization tests. like
http://pryan.org/mozilla/firefox/roytam1/ : optimization + customization
http://wiki.mozilla-x86-64.com/Download : x64
http://marilab.hp.infoseek.co.jp/buildfx/ : SSE2+PGU optimization
http://www1.plala.or.jp/tete009/software.html : optimization + option customization

> * Often these Community Edition builds are based on older versions
> of Firefox or contain unknown additional software;

Are they based on older version? I think we can get even 3.1alpha.
Of cource, some of the community edition bulids might be based on older
ones.

> * Often the individuals who have created these Community Edition
> builds represent themselves, through illegal use of our logos, as
> Mozilla and collect money and personal information from the users;
> * We are working to shut down fraudulent advertisers and Web sites.
> Ken Kovash and David Rolnitzky have estimated that approximately
> 6.7 million users visit these fraudulent sites annually.
> * We cannot ensure the quality of these builds.
>

> The purpose of this post is to let you know that we’re thinking of
> terminating the Community Edition builds, set forth our reasons why, and
> to receive your feedback. If you are aware of (i) any reasons why we

> should continue to allow these builds or (ii) anyone who is creating


> these builds for other than a commercial or fraudulent purpose, let me know.

I ofcource agree with the importance of the trademark and logo restriction,
but i cannot see the overweigh benefit to shut them all down than allowing
such optimization builds.


regards,
- --
Atsushi Shimono - shi...@mozilla.gr.jp
http://www.mozilla.gr.jp/~shimono/blog/
Mozilla-Gumi : Japanese Mozilla Users Group / System Administrator
MDC Japan Project Leader, Bugzilla Localization Working Group Member
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)

iEYEAREKAAYFAkjiPloACgkQBygbhFUCKWZCEgCeMQDGPTV/DKBHyTvNNy9+Thpr
OxkAnjlvvKDxDLBViJdV2jLH8J9GRdEH
=6jX+
-----END PGP SIGNATURE-----

Jean-Marc Desperrier

unread,
Sep 30, 2008, 11:20:16 AM9/30/08
to
Sergey Yanovich wrote:
> [...]

> Why it doesn't help? The problems raised in the original post were facts
> of Firefox being *sold* and being *malitious*. GPL prevents selling and
> requires source code to come along binaries.
>
> GPL-violation-watch is a very popular game.

Some people have no problem at all reselling GPL programs for profits
(whilst bringing *nothing* to the commmunity).

It's not really difficult to both comply with the letter of the GPL by
making the source available, but trick many people into not noticing
they could have had exactly the same thing for free.

Gijs Kruitbosch

unread,
Sep 30, 2008, 11:20:49 AM9/30/08
to
Sergey Yanovich wrote:
> Gijs Kruitbosch wrote:
>> IANAL, but isn't the GPL a license, and isn't the point of (and/or
>> issue with) the Community Edition branding (ergo, trademarks), rather
>> than the license of the code itself? As far as I know, the fact that
>> the code can be used under the LGPL and MPL license as well as the GPL
>> has no bearing on the reuse of the "Firefox" brandname, which is what
>> most of these problems originate from: some things/"products" are
>> called Firefox when they really aren't, and this causes an inferior
>> user experience for people finding these things.
>
> RedHat's approach demostrates how to settle the trademark issue in a
> GPLed product.

I'm sure that it might, I'm not familiar with it. But that's a trademark
approach, and is therefore separate from the license issue. Changing the license
doesn't affect the trademark policy. If you want Mozilla to adopt RedHat's
"approach" (whatever it is), then say that, not that they should switch to the
GPL. :-)

>> The only point I can possibly see in your point is that it would force
>> such Community Editions to open their source code. Which doesn't help.
>
> Why it doesn't help? The problems raised in the original post were facts
> of Firefox being *sold* and being *malitious*. GPL prevents selling and
> requires source code to come along binaries.
>
> GPL-violation-watch is a very popular game.

The GPL does not forbid selling. See http://www.gnu.org/philosophy/selling.html

The availability of source code for malicious programs won't help in this case.
It can help expose their intentions, but not for Joe User, and usually not in a
way that could not already be done without the publication of the source code.

>> And in fact, doing what you suggest now cannot retroactively be
>> applied to previous versions of the code (eg. Fx 2, 3, or other
>> releases) and hence would not influence anything.
>
> This is true. The solution is for the future releases only.
>
>> Finally, the last relicensing (tri-licensing) of the code took several
>> years to complete. I don't see why what you're suggesting would be any
>> faster
>
> The switch can be done at the next commit. Tri-licence explicitly allows
> this.

Maybe, but Mozilla is still supported largely by a community. Even if legally,
they could switch licenses at any time, I'm sure people would be upset if they
decided to switch to GPL out of nowhere without consulting that community. Not
to mention it stifling innovation - there are many non-MoFo/MoCo products based
on Mozilla code, and their source code is sometimes not available, nor licensed
GPL. Switching would force these users to use obsolete versions of the source
code, or switch platforms, which would in the end have negative consequences for
the Mozilla Project, I think.

~ Gijs

Michael Connor

unread,
Sep 30, 2008, 11:45:10 AM9/30/08
to Atsushi Shimono, cbr...@mozilla.com, gover...@lists.mozilla.org

No one is arguing that there are not legitimate uses which comply with
the spirit of the original policy. The argument is that there is a
volume of abuse that is targeting unwary users that we cannot deem as
acceptable. That said, there is nothing about those builds that
requires the branding for their users to successfully use them. Other
optimized builders just use the default branding, with no ill
effects. (http://www.beatnikpad.com/archives/2008/07/16/firefox is a
fine example.)

>> * Often the individuals who have created these Community Edition
>> builds represent themselves, through illegal use of our logos,
>> as
>> Mozilla and collect money and personal information from the
>> users;
>> * We are working to shut down fraudulent advertisers and Web
>> sites.
>> Ken Kovash and David Rolnitzky have estimated that approximately
>> 6.7 million users visit these fraudulent sites annually.
>> * We cannot ensure the quality of these builds.
>>
>> The purpose of this post is to let you know that we’re thinking of
>> terminating the Community Edition builds, set forth our reasons
>> why, and
>> to receive your feedback. If you are aware of (i) any reasons why we
>> should continue to allow these builds or (ii) anyone who is creating
>> these builds for other than a commercial or fraudulent purpose, let
>> me know.
>
> I ofcource agree with the importance of the trademark and logo
> restriction,
> but i cannot see the overweigh benefit to shut them all down than
> allowing
> such optimization builds.

If the policy is being abused, and there's no obvious way to for a
random user to know that the builds are not from Mozilla, there is a
pretty compelling case that the policy is creating a bad situation.
Six million users a year is simply too many people to put at risk for
unspecified benefits.

-- Mike

Jonas Sicking

unread,
Sep 30, 2008, 1:42:16 PM9/30/08
to Sergey Yanovich
Sergey Yanovich wrote:
>> The only point I can possibly see in your point is that it would force
>> such Community Editions to open their source code. Which doesn't help.
>
> Why it doesn't help? The problems raised in the original post were facts
> of Firefox being *sold* and being *malitious*. GPL prevents selling and
> requires source code to come along binaries.

GPL does not prevent selling. It does however require that the source
code can be retrieved by the buying parties. However that would only
help with part of the problem that the community edition builds causes.

/ Jonas

Michael Connor

unread,
Sep 30, 2008, 2:16:57 PM9/30/08
to Sergey Yanovich, gover...@lists.mozilla.org

On 30-Sep-08, at 10:57 AM, Sergey Yanovich wrote:

> Gijs Kruitbosch wrote:
>> IANAL, but isn't the GPL a license, and isn't the point of (and/or
>> issue
>> with) the Community Edition branding (ergo, trademarks), rather
>> than the
>> license of the code itself? As far as I know, the fact that the
>> code can
>> be used under the LGPL and MPL license as well as the GPL has no
>> bearing
>> on the reuse of the "Firefox" brandname, which is what most of these
>> problems originate from: some things/"products" are called Firefox
>> when
>> they really aren't, and this causes an inferior user experience for
>> people finding these things.
>
> RedHat's approach demostrates how to settle the trademark issue in a
> GPLed product.

GPL has nothing, at all, to do with trademark policy. It doesn't
prevent trademarks from being enforced, and it doesn't prevent people
from shipping spyware-laden versions, they just need to make the
source available somewhere. The people who are getting suckered into
buying malicious software aren't going to benefit one iota from a
source license change. The problem is that the current policy lets
people do this in a way that's abusing the trademark to exploit users.

Your argument simply doesn't make any sense.

-- Mike

Sergey Yanovich

unread,
Sep 30, 2008, 2:37:59 PM9/30/08
to Michael Connor, gover...@lists.mozilla.org
Michael Connor wrote:
>
> On 30-Sep-08, at 10:57 AM, Sergey Yanovich wrote:
>> RedHat's approach demostrates how to settle the trademark issue in a
>> GPLed product.
>
> GPL has nothing, at all, to do with trademark policy. It doesn't
> prevent trademarks from being enforced, and it doesn't prevent people
> from shipping spyware-laden versions

I never meant to say any of this.

> The people who are getting suckered into buying
> malicious software aren't going to benefit one iota from a source
> license change. The problem is that the current policy lets people do
> this in a way that's abusing the trademark to exploit users.

GPL rings the bell for many: "This should be free, and where's the
source?". Apart from that, policy change won't solve any problem by
itself either. It will need to be enforced, probably with lawsuits. GPL
violation suits are known to work, and Mozilla will most likely be able
to outsource them for free to specially trained folks.

> Your argument simply doesn't make any sense.

"Any" is a very strong claim. ;)

--
Sergey Yanovich

Mike Shaver

unread,
Sep 30, 2008, 2:42:36 PM9/30/08
to Sergey Yanovich, gover...@lists.mozilla.org
On Tue, Sep 30, 2008 at 10:57 AM, Sergey Yanovich <ynv...@gmail.com> wrote:
> RedHat's approach demostrates how to settle the trademark issue in a
> GPLed product.

Unfortunately, no; Red Hat's trademark license says, last I checked,
that you cannot use the trademark without explicit written permission
(and Red Hat doesn't answer questions about trademark use unless you
have a business arrangement with them).

http://www.redhat.com/about/companyprofile/trademark/

Such a policy would not permit any of those builds to use the Firefox
name, just as CentOS couldn't use the RHEL name even if they wanted
to.

Mike

Sergey Yanovich

unread,
Sep 30, 2008, 2:47:37 PM9/30/08
to Mike Shaver, gover...@lists.mozilla.org
Mike Shaver wrote:
> On Tue, Sep 30, 2008 at 10:57 AM, Sergey Yanovich <ynv...@gmail.com> wrote:
>> RedHat's approach demostrates how to settle the trademark issue in a
>> GPLed product.
>
> Unfortunately, no; Red Hat's trademark license says, last I checked,
> that you cannot use the trademark without explicit written permission
> (and Red Hat doesn't answer questions about trademark use unless you
> have a business arrangement with them).

> Such a policy would not permit any of those builds to use the Firefox


> name, just as CentOS couldn't use the RHEL name even if they wanted
> to.

That's exactly my point! And exactly what Mozilla needs to control the
Firefox trademark, right?

--
Sergey Yanovich

Justin Dolske

unread,
Sep 30, 2008, 2:56:54 PM9/30/08
to
Atsushi Shimono wrote:

> Many CE builds targets 'optimizations' but not localization tests. like
> http://pryan.org/mozilla/firefox/roytam1/ : optimization + customization
> http://wiki.mozilla-x86-64.com/Download : x64
> http://marilab.hp.infoseek.co.jp/buildfx/ : SSE2+PGU optimization
> http://www1.plala.or.jp/tete009/software.html : optimization + option customization

Why do these need to be Community Editions? If people are finding things
to optimize, they should be filing bugs and pushing those patches upstream.

Justin

Mike Shaver

unread,
Sep 30, 2008, 3:34:17 PM9/30/08
to Sergey Yanovich, gover...@lists.mozilla.org
On Tue, Sep 30, 2008 at 2:47 PM, Sergey Yanovich <ynv...@gmail.com> wrote:
> That's exactly my point! And exactly what Mozilla needs to control the
> Firefox trademark, right?

Did you read Catherine's mail and the Community Edition program
description? The problem is that the program -- which does not exist
in the Red Hat model -- provides a loophole that is being widely
abused, and that we wish to close. It does require a policy change,
because the Community Edition policy is what makes these abuses harder
to act on.

Mike

Sergey Yanovich

unread,
Sep 30, 2008, 3:44:41 PM9/30/08
to Mike Shaver, gover...@lists.mozilla.org

Sure. It looks like I failed to articulate the main point. My proposal
isn't meant to replace, but to extend the proposed changes.

--
Sergey Yanovich

Michael Lefevre

unread,
Sep 30, 2008, 4:01:59 PM9/30/08
to Justin Dolske

They need to be Community Editions because Mozilla doesn't want to make
more builds, I think...

AIUI, these are not optimisations in terms of improving code, they are
tweaking build/compiler options which may also have significant negative
effects (e.g. greatly increasing file size, only running on some types
of system, tweaking config options to things that make sense only for
geeks, including experimental stuff which is checked in but not yet
enabled...)

I'm pretty sure there is a bug about providing 64-bit Windows builds,
but doing it is not a matter of necessary code changes, it's a matter of
Mozilla not wanting to have all the overheads associated with another
set of builds, when the resulting builds are only of interest to a small
group of people. Although I guess 64-bit Windows, and any other ports
out there doing CE builds could maybe move over to having the same
status as the ports in the contrib/ folder?

--
Michael

Michael Connor

unread,
Sep 30, 2008, 4:10:55 PM9/30/08
to Sergey Yanovich, gover...@lists.mozilla.org

That doesn't make any sense either. The source license and the
trademark usage policy are entirely orthogonal to each other.

If you want to propose dropping the tri-license and going GPL-only,
that's entirely unrelated and should be its own thread.

-- Mike

Benjamin Smedberg

unread,
Sep 30, 2008, 4:11:49 PM9/30/08
to
Sergey Yanovich wrote:

> Sure. It looks like I failed to articulate the main point. My proposal
> isn't meant to replace, but to extend the proposed changes.

For what purpose? What does this proposal, about changing our trademark
policy, have to do with a licensing change for GPL-only?

If you're going to advocate switching to GPL-only, please do so in a
different thread, and outline specifically what goal you'd be trying to
accomplish. I know that if Mozilla moved to GPL-only, I'd jump ship and look
for other pastures pretty damn quickly. If anything, I'd like to drop the
tri-license and be MPL-only.

--BDS

Michael Connor

unread,
Sep 30, 2008, 4:24:10 PM9/30/08
to Sergey Yanovich, gover...@lists.mozilla.org

On 30-Sep-08, at 2:37 PM, Sergey Yanovich wrote:

> Michael Connor wrote:


>>
>> On 30-Sep-08, at 10:57 AM, Sergey Yanovich wrote:
>>> RedHat's approach demostrates how to settle the trademark issue in a
>>> GPLed product.
>>

>> GPL has nothing, at all, to do with trademark policy. It doesn't
>> prevent trademarks from being enforced, and it doesn't prevent people
>> from shipping spyware-laden versions
>
> I never meant to say any of this.

If you're suggesting it as a solution to an issue of people abusing
our trademark policy to ship malicious builds, I think you are.

>> The people who are getting suckered into buying
>> malicious software aren't going to benefit one iota from a source
>> license change. The problem is that the current policy lets people
>> do
>> this in a way that's abusing the trademark to exploit users.
>
> GPL rings the bell for many: "This should be free, and where's the
> source?". Apart from that, policy change won't solve any problem by
> itself either. It will need to be enforced, probably with lawsuits.
> GPL
> violation suits are known to work, and Mozilla will most likely be
> able

> to outsource them for free to specially trained folks.

First, GNU encourages selling code: http://www.gnu.org/philosophy/selling.html

Second, complying with the GPL doesn't mean users are safe, or that
the builds can be trusted. If people are using letter-of-the-law to
distribute sketchy modifications to Firefox legally, why wouldn't they
just provide the source somewhere? They could easily make minimal
changes that pull down malicious bits and install them separately, if
they're protecting their malware...

Third, I still think you're misunderstanding the entire point of why
this policy is an issue. If we eliminate this loophole, people can't
use our trademarks without permission, and thus they're already
subject to trademark infringement suits, which are even more widely
known to work.

>> Your argument simply doesn't make any sense.
>
> "Any" is a very strong claim. ;)

It is, but it is one borne out by facts.

-- Mike

Eddy Nigg

unread,
Sep 30, 2008, 6:20:53 PM9/30/08
to
On 09/30/2008 11:01 PM, Michael Lefevre:

> AIUI, these are not optimisations in terms of improving code, they are
> tweaking build/compiler options which may also have significant negative
> effects (e.g. greatly increasing file size, only running on some types
> of system, tweaking config options to things that make sense only for
> geeks, including experimental stuff which is checked in but not yet
> enabled...)

Nobody will prevent such optimized builds. Not now and not in the
future...when no changes are made to the code, those can still be called
Firefox (if I understand the current trademark policy correctly).


--
Regards

Signer: Eddy Nigg, StartCom Ltd.
Jabber: star...@startcom.org
Blog: https://blog.startcom.org

Michael Connor

unread,
Sep 30, 2008, 6:52:29 PM9/30/08
to Eddy Nigg, gover...@lists.mozilla.org

On 30-Sep-08, at 6:20 PM, Eddy Nigg wrote:

> On 09/30/2008 11:01 PM, Michael Lefevre:
>> AIUI, these are not optimisations in terms of improving code, they
>> are
>> tweaking build/compiler options which may also have significant
>> negative
>> effects (e.g. greatly increasing file size, only running on some
>> types
>> of system, tweaking config options to things that make sense only for
>> geeks, including experimental stuff which is checked in but not yet
>> enabled...)
>
> Nobody will prevent such optimized builds. Not now and not in the
> future...when no changes are made to the code, those can still be
> called
> Firefox (if I understand the current trademark policy correctly).

I don't think you do. You can redistribute unaltered _binaries_ but
no mention is made in the policy that I can see or recall...

-- Mike

Eddy Nigg

unread,
Sep 30, 2008, 7:11:33 PM9/30/08
to
On 10/01/2008 01:52 AM, Michael Connor:

Then I think the issue of repackaging *without source modification* must
be addressed in an adequate way - specially for Linux vendors.

Michael Connor

unread,
Sep 30, 2008, 7:22:47 PM9/30/08
to Eddy Nigg, gover...@lists.mozilla.org

On 30-Sep-08, at 7:11 PM, Eddy Nigg wrote:

> On 10/01/2008 01:52 AM, Michael Connor:
>>
>> On 30-Sep-08, at 6:20 PM, Eddy Nigg wrote:
>>> Nobody will prevent such optimized builds. Not now and not in the
>>> future...when no changes are made to the code, those can still be
>>> called
>>> Firefox (if I understand the current trademark policy correctly).
>>
>> I don't think you do. You can redistribute unaltered _binaries_ but
>> no
>> mention is made in the policy that I can see or recall...
>>
>
> Then I think the issue of repackaging *without source modification*
> must
> be addressed in an adequate way - specially for Linux vendors.

That's entirely out of scope for this thread, of course, but most
vendors ship modified builds to some extent, so its largely theoretical.

-- Mike

Eddy Nigg

unread,
Sep 30, 2008, 7:33:54 PM9/30/08
to
On 10/01/2008 02:22 AM, Michael Connor:

>
> That's entirely out of scope for this thread, of course, but most
> vendors ship modified builds to some extent, so its largely theoretical.
>

Under which section are repackaging and distribution of Firefox by, lets
say, Red Hat / Fedora etc. handled? I'm aware that there aren't any or
very minor modifications to profile settings without source modifications.

It was my understanding that those were the accepted rules - but maybe
that was exactly covered under the community edition builds...

Michael Connor

unread,
Sep 30, 2008, 9:42:33 PM9/30/08
to Eddy Nigg, gover...@lists.mozilla.org

On 30-Sep-08, at 7:33 PM, Eddy Nigg wrote:

> On 10/01/2008 02:22 AM, Michael Connor:
>>
>> That's entirely out of scope for this thread, of course, but most
>> vendors ship modified builds to some extent, so its largely
>> theoretical.
>>
>
> Under which section are repackaging and distribution of Firefox by,
> lets
> say, Red Hat / Fedora etc. handled? I'm aware that there aren't any or
> very minor modifications to profile settings without source
> modifications.

Well, if you read the policy, there's no mention at all of this case,
so it falls under the "contact us if you want to do anything not
covered by this policy" bit. Which is why we have always worked with
vendors to sign off and approve their use of the marks for their own
builds.

Fedora is the closest to stock, actually, caillon does a lot of work
to upstream things. Ubuntu's current diff is https://launchpadlibrarian.net/17950309/firefox-3.0_3.0.3%2Bbuild1%2Bnobinonly-0ubuntu1.diff.gz
, et.

> It was my understanding that those were the accepted rules - but maybe
> that was exactly covered under the community edition builds...

Very minor modifications to specific bits were permitted under the
Community Edition policy, but I don't know of any distros that were
availing themselves of that policy.

The policy hasn't changed for quite some time, there may have been
some less formalized approval of zero-changes builds, but it was never
part of the public policy.

-- Mike

Sergey Yanovich

unread,
Oct 1, 2008, 5:34:12 AM10/1/08
to Michael Connor, gover...@lists.mozilla.org
Michael Connor wrote:
> First, GNU encourages selling code:
> http://www.gnu.org/philosophy/selling.html

No. It encourages charging fee for distributing copies. And I don't know
a person, who understands what GPL is and is still ready to pay for copy
distribution. Paid support is completely different business.

> Third, I still think you're misunderstanding the entire point of why
> this policy is an issue. If we eliminate this loophole, people can't
> use our trademarks without permission, and thus they're already subject
> to trademark infringement suits, which are even more widely known to work.

I still think you're misunderstanding the entire point of GPL compliance
enforcement being cheaper for Mozilla than trademark policy enforcement.
You cannot file a DMCA takedown notice on trademark reasons. And I doubt
EFF will be willing to handle Mozilla trademark infringement suits, much
unlike GPL violation suits.

That said, switching to GPL-only may have unrelated side effects, like
bsmedberg mentioned, which effective prevents anyone inside Mozilla from
thinking about being GPL-only. In this case, it is generally more polite
to name the side effects and their crucial importance, rather than to
claim other people's argument being nonsense.

--
Sergey Yanovich

Eddy Nigg

unread,
Oct 1, 2008, 8:09:26 AM10/1/08
to
On 10/01/2008 04:42 AM, Michael Connor:

>
> Fedora is the closest to stock, actually, caillon does a lot of work to
> upstream things.

There are many Red Hat contributions, but it still doesn't answer the
question in relation to the trademark policy of Mozilla (Firefox) then.
RHEL, Fedora and folks are all using the Firefox trademark, but
repackage and adjust to a certain extended.

> Very minor modifications to specific bits were permitted under the
> Community Edition policy, but I don't know of any distros that were
> availing themselves of that policy.
>
> The policy hasn't changed for quite some time, there may have been some
> less formalized approval of zero-changes builds, but it was never part
> of the public policy.
>

OK, then what does that mean for Linux distributions? I represent also
one of them, hence I'm interested to know what the impact of the change
would be for Linux vendors (if at all). Are we in line with the
trademark policy of Mozilla (seems like we aren't, but...)

Benjamin Smedberg

unread,
Oct 1, 2008, 8:37:40 AM10/1/08
to Michael Lefevre, Justin Dolske
Michael Lefevre wrote:
> Justin Dolske wrote:
>> Atsushi Shimono wrote:
>>
>>> Many CE builds targets 'optimizations' but not localization tests.
>>> like
>>> http://pryan.org/mozilla/firefox/roytam1/ : optimization + customization
>>> http://wiki.mozilla-x86-64.com/Download : x64
>>> http://marilab.hp.infoseek.co.jp/buildfx/ : SSE2+PGU optimization
>>> http://www1.plala.or.jp/tete009/software.html : optimization + option
>>> customization
>> Why do these need to be Community Editions? If people are finding things
>> to optimize, they should be filing bugs and pushing those patches upstream.
>
> They need to be Community Editions because Mozilla doesn't want to make
> more builds, I think...

Indeed, Mozilla does not want to make many builds optimized for specific
architectures.

The question seems to be though, whether we should allow a general trademark
license to anyone who wants to make such a build, or whether we should
require them to contact Mozilla and get a trademark agreement.

I would hope that, given the right circumstances, Mozilla would continue to
allow these community editions to exist with an explicit trademark
agreement. That continues to protect our brand, without allowing random
companies on the internet to abuse our brand for nefarious purposes.

--BDS

Mike Shaver

unread,
Oct 1, 2008, 9:16:46 AM10/1/08
to Eddy Nigg, gover...@lists.mozilla.org
Linux distributors do not use the Community Edition policy for their
builds, so this change will have no effect on them.

Mike

> _______________________________________________
> governance mailing list
> gover...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/governance
>

--
Sent from Gmail for mobile | mobile.google.com

Alex K.

unread,
Oct 1, 2008, 10:12:30 AM10/1/08
to
Sergey Yanovich wrote:
> Michael Connor wrote:
>> First, GNU encourages selling code:
>> http://www.gnu.org/philosophy/selling.html
>
> No. It encourages charging fee for distributing copies. And I don't know
> a person, who understands what GPL is and is still ready to pay for copy
> distribution. Paid support is completely different business.
>
>> Third, I still think you're misunderstanding the entire point of why
>> this policy is an issue. If we eliminate this loophole, people can't
>> use our trademarks without permission, and thus they're already subject
>> to trademark infringement suits, which are even more widely known to work.
>
> I still think you're misunderstanding the entire point of GPL compliance
> enforcement being cheaper for Mozilla than trademark policy enforcement.
> You cannot file a DMCA takedown notice on trademark reasons. And I doubt
> EFF will be willing to handle Mozilla trademark infringement suits, much
> unlike GPL violation suits.
>
IANAL, but, as I understand it, the issue is most definitely trademark
enforcement.

Even if they fully complied with the GPL, they would still be infringing
on the trademarks. Whether its cheaper, or not, filing complaints about
*copyright* violations does not protect the *trademarks*. They are two,
distinct areas of law.

Specifically, unlike copyrights, trademarks can be deemed 'abandoned',
if the owner "fails to protest the unauthorized use of the mark by
others or lets others use the mark without adequate supervision.":

http://legalcatch.wordpress.com/2006/10/24/abandonment-trademark-legal-definition-business-law/

So, my understanding of that, is that Mozilla *must* actively protest
*every* unauthorized use of their trademarks that comes to their
attention, *and* monitor the authorized uses, or they can lose the
subject trademark.

The way I see it, Mozilla is trying to protect itself here. It appears,
to me, that the goal is to remove any ambiguity, as to whether
permission is needed, to use the trademarks.

As it is currently, a defendant could, theoretically, argue that their
use was authorized, under the CE policy. Of course, Mozilla could then
argue that the use was 'disparaging', but that would all boil down to a
courtroom battle, for a judge to ultimately determine. That can get
very expensive, very quickly.

IMO, by removing the ambiguity, it makes for a rock solid case against
any unauthorized use, and much *less* of a chance that an infringer
would try to challenge it in court.

Personally, I don't see any reason why *legitimate community builds*
couldn't be granted permission to use the trademarks. The only
difference, would be that they would have to explicitly ask for it.
Mozilla could still grant permission, on a case-by-case basis.

Thats how anyone else in the business world gets permission to use
someone else's trademark, they ask, & negotiate the terms, conditions, etc.

Again, though, as I said in my opening, IANAL. This is just my own,
layman's understanding of the (US) law and the legal issues involved.
As always, YMMV.

--
Alex K.

Sergey Yanovich

unread,
Oct 1, 2008, 11:37:23 AM10/1/08
to Alex K.
Alex K. wrote:
> IMO, by removing the ambiguity, it makes for a rock solid case against
> any unauthorized use, and much *less* of a chance that an infringer
> would try to challenge it in court.

My point was Mozilla can avoid *going to court* at all, if it chooses to
use GPL as a solution to Firefox being sold for money or being tweaked
into malware.

Since GPL-only Firefox doesn't seem to fly, the case is closed.

--
Sergey Yanovich

Boris Zbarsky

unread,
Oct 1, 2008, 12:00:10 PM10/1/08
to
Sergey Yanovich wrote:
> My point was Mozilla can avoid *going to court* at all, if it chooses to
> use GPL as a solution to Firefox being sold for money or being tweaked
> into malware.

The point is that the GPL doesn't solve those problems.

-Boris

Michael Connor

unread,
Oct 1, 2008, 12:52:35 PM10/1/08
to Sergey Yanovich, gover...@lists.mozilla.org

On 1-Oct-08, at 5:34 AM, Sergey Yanovich wrote:

> Michael Connor wrote:
>> First, GNU encourages selling code:
>> http://www.gnu.org/philosophy/selling.html
>

> No. It encourages charging fee for distributing copies. And I don't
> know
> a person, who understands what GPL is and is still ready to pay for
> copy
> distribution. Paid support is completely different business.

Minor distinction, but you're still paying for GPLed software, so its
not like GPL === free in a universal sense. And as I have stated, the
people who understand what the GPL is are not really the people I'm
worried about with these builds. (If you know what the GPL means, you
probably know that you don't want to get Firefox off some random
website.

>> Third, I still think you're misunderstanding the entire point of why
>> this policy is an issue. If we eliminate this loophole, people can't
>> use our trademarks without permission, and thus they're already
>> subject
>> to trademark infringement suits, which are even more widely known
>> to work.
>

> I still think you're misunderstanding the entire point of GPL
> compliance
> enforcement being cheaper for Mozilla than trademark policy
> enforcement.
> You cannot file a DMCA takedown notice on trademark reasons. And I
> doubt
> EFF will be willing to handle Mozilla trademark infringement suits,
> much
> unlike GPL violation suits.

This has nothing to do with cost. You assuming that these builds
would somehow violate the GPL. If people are attempting to use legal
loopholes to distribute software using our marks, I doubt they'd then
turn around and violate the GPL and expose themselves to obvious
targeting. Why would they do so? If they were willing to violate the
GPL, they'd almost certainly just use the official branding and go
full out until they got a takedown notice.

Why do you think these people simply wouldn't provide the source code?

> That said, switching to GPL-only may have unrelated side effects, like
> bsmedberg mentioned, which effective prevents anyone inside Mozilla
> from
> thinking about being GPL-only. In this case, it is generally more
> polite
> to name the side effects and their crucial importance, rather than to
> claim other people's argument being nonsense.

Your argument is that GPL would help with the problems we're trying to
solve here. It won't make any difference in this case, and to
continue to claim there would be value in such is nonsense. The side
effects are extra reasons to not do it, but they're not why its
nonsense.

-- Mike

Sergey Yanovich

unread,
Oct 1, 2008, 1:37:31 PM10/1/08
to Michael Connor, gover...@lists.mozilla.org
Michael Connor wrote:
> Why do you think these people simply wouldn't provide the source code?

If for no other reason than to avoid providing evidence for criminal
cases against them. Malware coders and fraudsters are bad people, but
they are by no means stupid.

--
Sergey Yanovich

Catherine Brady

unread,
Apr 5, 2009, 2:42:56 PM4/5/09
to gover...@lists.mozilla.org
Several months ago, we submitted a proposal to eliminate the Community Edition program because its needs were being addressed elsewhere and this program had outlived the reason for which it was created in the first place. When something no longer works, you don't keep it going. Prior to closing out this discussion, I’d like to address the comments received, which may be summarized in a four categories, and then describe what I think the next steps are.

1. What does this mean for Linux distributions? Linux distributions do not use the Community Edition program for their builds so this change does not affect them.

2. May I still modify Mozilla products? Mozilla is, and will continue to be, open source. The flexibility and ability to customize our products to fit your own needs is core to Mozilla and we have no desire to change that. If you do not want to use any of the Mozilla trademarks, you may make any changes you wish without receiving our permission. If you would like to create a modified version of a Mozilla product and use our trademarks, please contact us at part...@mozilla.com or biz...@mozilla.com to discuss your proposed modifications.

3. Many of the existing Community Editions serve a purpose or fulfill a special need. We agree that there are legitimate Community Editions that comply with the spirit of the original program. This is why we’re continuing to allow legacy builds (see https://bugzilla.mozilla.org/show_bug.cgi?id=486785) that comply with the original requirements.

4. Community Editions should be relicensed under the GPL. This proposal is independent from, and makes no changes to, the licenses under which we make the source code available to the public. If there is material interest in licensing question, we should separate that from the current discussion on Community Edition program and address it in another thread.

I'd like to move forward on the proposal and close this out. To that end, I have filed Bug 486785. If further clarification is required, please let me know.

----- Original Message -----
From: "Catherine Brady" <cbr...@mozilla.com>
To: gover...@lists.mozilla.org
Sent: Monday, September 29, 2008 2:14:58 PM GMT -08:00 US/Canada Pacific
Subject: Terminating Community Edition Builds of Firefox

We are in the process of updating our Trademark Policy. One of the proposed modifications is to eliminate the Community Edition program, which people are using as a loophole to create unauthorized modified Firefox builds using the Firefox name and selling them to the unsuspecting general public.


During the beginning days of Firefox, when we did not have the bandwidth to localize Firefox, the idea of a Community Edition of Firefox was created. (See http://www.mozilla.org/foundation/trademarks/community-edition-policy.html ) The program’s intention was to allow members around the world to create localized versions of Firefox without having to obtain a license or approval from Mozilla. However, as we continued to grow, so did our localization (“L10n”) teams and program. Now all L10n efforts are part of the official Firefox release. Our community is not using the Community Edition builds to create localized versions of Firefox.


As far as we’re aware, the only people taking advantage of this legacy program are people trying to circumvent the licensing requirements of our trademark policy. These people create a modified version of Firefox, call it a “Community Edition”, and sell this modified Firefox to make money and collect users’ personal information (including name, email address, address and credit card information). We cannot ensure the quality of these builds nor the negative user experience created by these opportunistic builds.


Reason to Terminate Community Edition Builds of Firefox . The following are the reasons why we believe that Community Edition builds are irrelevant today:

• We have an established L10n program and the community no longer use the Community Edition builds to create localized versions of Firefox;
• Community Edition builds are being used by individuals trying to circumvent the licensing requirements for modified builds of Firefox;
• Often these Community Edition builds are based on older versions of Firefox or contain unknown additional software;
• Often the individuals who have created these Community Edition builds represent themselves, through illegal use of our logos, as Mozilla and collect money and personal information from the users;
• We are working to shut down fraudulent advertisers and Web sites. Ken Kovash and David Rolnitzky have estimated that approximately 6.7 million users visit these fraudulent sites annually.
• We cannot ensure the quality of these builds.


The purpose of this post is to let you know that we’re thinking of terminating the Community Edition builds, set forth our reasons why, and to receive your feedback. If you are aware of (i) any reasons why we should continue to allow these builds or (ii) anyone who is creating these builds for other than a commercial or fraudulent purpose, let me know.

Axel Hecht

unread,
Apr 5, 2009, 4:26:21 PM4/5/09
to
On 05.04.2009 20:42 Uhr, Catherine Brady wrote:

<...>


> 4. Community Editions should be relicensed under the GPL. This
> proposal is independent from, and makes no changes to, the licenses
> under which we make the source code available to the public. If
> there is material interest in licensing question, we should separate
> that from the current discussion on Community Edition program and
> address it in another thread.

Is there a rationale why community editions should be GPL? Seems like
it's not the closest license to the Mozilla ecosystem.

Axel

mco...@mozilla.com

unread,
Apr 5, 2009, 5:21:03 PM4/5/09
to Axel Hecht, gover...@lists.mozilla.org

On 5-Apr-09, at 4:30 PM, Axel Hecht <ax...@pike.org> wrote:

> Is there a rationale why community editions should be GPL? Seems
> like it's not the closest license to the Mozilla ecosystem.
>
> Axel

I would go further and say that this doesn't actually make sense at
all, and would prevent inclusion of modifications made into upstream
builds. I can't imagine any circumstance where this would be an
acceptable outcome.

- Mike

Benjamin Smedberg

unread,
Apr 5, 2009, 7:57:39 PM4/5/09
to
On 4/5/09 2:42 PM, Catherine Brady wrote:

> 3. Many of the existing Community Editions serve a purpose or fulfill a
> special need. We agree that there are legitimate Community Editions that
> comply with the spirit of the original program. This is why we’re
> continuing to allow legacy builds (see
> https://bugzilla.mozilla.org/show_bug.cgi?id=486785) that comply with the
> original requirements.

I think we should take one step further and continue to, with a written
agreement, continue to allow *new* community builds. The real change we're
making, then, is that you can't do a community build without an agreement in
place.

--BDS

Eddy Nigg

unread,
Apr 5, 2009, 8:17:19 PM4/5/09
to
On 04/05/2009 09:42 PM, Catherine Brady:

> 1. What does this mean for Linux distributions? Linux distributions do not use the Community Edition program for their builds so this change does not affect them.
>

Can you provide some more information about what Linux distros do use?
Usually it's a branded strait re-build, but...??

Catherine Brady

unread,
Apr 6, 2009, 12:50:31 AM4/6/09
to mco...@mozilla.com, Axel Hecht, gover...@lists.mozilla.org
Let me clarify.  As mentioned in the beginning of my post, I was (i) summarizing the comments received, which fell into 4 categories and then (ii) responding to the comments.  The summary of the comment was "The Community Editions are not licensed under GPL" The response is that licensing is a separate discussion. There is not intention of changing the way we currently license our products.

- Mike

Catherine Brady

unread,
Apr 6, 2009, 12:56:05 AM4/6/09
to Benjamin Smedberg, gover...@lists.mozilla.org
Correct Benjamin. If someone wants to make a new community build or modified version of Firefox, or other Mozilla product, they should contact us to discuss their proposed changes. Again, there is no restriction to changes anyone may make to the products if they do not want to use our trademarks. If they want to make their changes and call it XYZ browser, no permission is required from Mozilla.
--ccb

----- Original Message -----
From: "Benjamin Smedberg" <benj...@smedbergs.us>
To: gover...@lists.mozilla.org
Sent: Sunday, April 5, 2009 4:57:39 PM GMT -08:00 US/Canada Pacific
Subject: Re: Terminating Community Edition Builds of Firefox

--BDS

_______________________________________________

Mark Surman

unread,
Apr 7, 2009, 10:51:29 AM4/7/09
to
On Apr 5, 7:57 pm, Benjamin Smedberg <benja...@smedbergs.us> wrote:

> I think we should take one step further and continue to, with a written
> agreement, continue to allow *new* community builds. The real change we're
> making, then, is that you can't do a community build without an agreement in
> place.

This is the right principle, and my understanding is that this is
effectivley what 'partnership builds' are.

Is there any benefit in saying 'partnership builds are community
builds, but w/ an agreement' on the new redirect page?

Maybe just confuses things. But it's an important point for people to
understand.

ms

Reply all
Reply to author
Forward
0 new messages