Fedora, Mozilla, and trademarks [LWN.net]

By Jake Edge
April 28, 2010

Trademarks and free software can make a volatile mix. It is understandable that a project would want to ensure that code shipping under its name is "the real McCoy", but modifying the source and distributing the result is a hallmark of free software. Trademark policy can place limits on what changes—for bugs, features, or even policy compliance—downstream projects can make and still use the trademarked names. The tension between the two has led some, like Debian, to re-brand Mozilla projects, so that they can ship the changes they want; some Fedora developers would like to see that distribution follow suit.

A Thunderbird crashing bug, reported by Felix Schwarz to the fedora-devel mailing list, is the proximate cause for the current controversy. Numerous Fedora users were running into the bug, and it had been patched upstream for several weeks, but there had been no release of Thunderbird for Fedora to fix the problem. Schwarz reported that the patch fixed the crash for him and others, and asked that it be pushed out: "However it is still not fixed in Thunderbird F-12 CVS. Can you please push the fix to CVS and push builds to testing/stable?"

Martin Stransky, one of the Fedora Mozilla maintainers, noted that "we're patching mozilla packages only for really critical issues because of mozilla trademarks", which caused concern that the trademarks were causing Fedora to ship a buggy Thunderbird. While the patch was available in the upstream repository, it hadn't been merged into the branch for the next release. Stransky said that he had requested that the next Thunderbird release include the fix in Mozilla's bugzilla entry, but that wasn't sufficient for some.

It turns out that the bug had been reported back in early March, but wasn't given a very high priority by the Thunderbird developers for two reasons: they couldn't reproduce it and it was not showing up with any frequency in their crash statistics. Meanwhile, since early April, it was crashing fairly frequently for Fedora users, leading to multiple bugs being filed, all of which were eventually collapsed into one with numerous commenters and "me too" posts.

But the idea that the trademark policy might prevent Fedora from shipping a working Thunderbird led to calls for rebranding the mail client (and the Firefox web browser) with different icons and names, as Debian has done. In fact, adopting the "Iceweasel" and "Icedove" names, if Debian is amenable, was one of the suggestions. Ralf Corsepius put it this way:

Thanks for providing evidence of how trademarks are being applied to void the benefits of "open source".

The obvious logical consequences of what you say would be
* either to remove the packages you are referring to from Fedora because they are effectively unmaintainable.
* or to remove the trademarks and re-brand the packages.

Fedora engineering steering committee (FESCo) member Kevin Kofler seems to be spearheading the effort to get out from under Mozilla's trademark policy. He agreed with Corsepius in a series of posts, which laid out multiple reasons that Fedora should consider renaming, including the Mozilla project's habit of bundling its own versions of system libraries—something that goes against Fedora packaging policies. He points to libpng as one example:

Another big issue is libpng: xulrunner is bundling a forked libpng for APNG [Animated PNG] support (which isn't even available for anything else to build against, so e.g. Konqueror can't support APNG). (APNG is a nonstandard extension to PNG which Mozilla is arbitrarily pushing instead of the existing MNG format which the PNG developers are supporting. [...]) Debian is patching it to use the system libpng (which removes APNG support, so it's unlikely to pass trademark approval, ever), we aren't. This is a blatant violation of our own packaging guidelines.

Several other examples were noted by Kofler, including Thunderbird bundling Gecko, rather than using the system xulrunner, and xulrunner bundling libffi. He also expressed frustration with the integration of Mozilla projects into desktops, in particular KDE. Overall, Kofler and others are completely fed up. Bruno Wolff III doesn't see any advantages to using the trademarks:

I don't see how using Mozilla trademarks provides significant benefit to Fedora. It seems to mostly benefit Mozilla. I don't see why we should be breaking our rules to help them.

Other posters defended Mozilla, but the loudest voices were clearly those who were unhappy with the status quo. Chris Tyler pointed out several reasons that Fedora should continue using the Mozilla trademarks including the well-established Mozilla brands:

Mozilla's brands are very well-known: They have 350+ million users across multiple platforms (Windows, Mac, Linux), far more than we have in Fedora. The ability to use these apps in Fedora helps to assures new users that switching costs will be low.

Tyler also sees the trademark rules as reasonable to protect users from distributions that unintentionally introduce vulnerabilities when patching Firefox or Thunderbird. He is optimistic that things can be worked out:

Fedora has a great relationship with Mozilla. They're an amazing project filled with people that Get It, and we can work out issues with them in a cooperative way.

It is possible for Fedora to patch its version of Thunderbird or Firefox, but it must get approval from Mozilla for each patch. the team that maintains the Fedora packages for Mozilla projects only wants to go through that process for "really critical issues" as Stransky noted. Later in the thread, he outlines which issues qualify: zero-day vulnerabilities and crashes that affect everyone. Kofler is, unsurprisingly, not happy with that Mozilla policy either:

They're the ONLY Free Software upstream insane enough to require approval for EVERY SINGLE patch as a condition to use their trademark. Imagine if Linus Torvalds did that for the kernel Linux, or the GNOME and/or KDE developers for their desktop environments. It would be impossible to maintain a distro under such conditions! Why does Mozilla get that sort of preferential treatment?

Fedora also wants to control its trademarks, though, and has its own trademark guidelines which are substantially similar in spirit—at least—to those of Mozilla. Adam Williamson described it this way:

You can't modify Fedora under F/OSS principles and still call it Fedora, just like you can't modify Firefox under F/OSS principles and still call it Firefox. Both of us do this to protect the good name of the project. We'd be in an extremely glass house-y situation if we tried to 'call out' Mozilla over this. It'd be ridiculous.

Lead Mozilla package maintainer Christopher Aillon tried to clarify the situation somewhat. The "impact of the bug was misjudged", he said, which is frustrating users: "I think we have a responsibility to both Fedora and Mozilla to include a fix for it". He intends to get a fix into updates-testing for a few days to ensure there are no regressions for other users. He also defends the process that the packaging team uses:

The main purpose to get patches accepted by upstream before inclusion in Fedora is to make sure we are doing things the right way. For example, some patches may inadvertently break standards compliance, have ill side effects with JavaScript, may fix connecting to some mail servers at the expense of others, etc.

[...] We do have an agreement with Mozilla and as such, we are permitted to use the Firefox and Thunderbird trademarks. But even if we did not or it were decided those marks were not important to us, I strongly feel that we should continue do things the right way and get patches accepted upstream first.

Furthermore, Aillon stated that the trademark policy wasn't really an issue for this particular bug. In a comment on the ticket filed by Rahul Sundaram asking FESCo to look into the issue, Aillon said it was simply a misjudgment by the packaging team about the importance of the bug. FESCo discussed the matter at its April 27 meeting, but decided not to change anything with respect to the Mozilla packages.

There are a number of different issues swirling around this bug. It seems likely that if the packagers had noticed how many users were being affected, it would have been quickly patched and the larger issue might never have come up—at least temporarily. But, the problem that some—particularly strong free software advocates—have with Mozilla's trademark policy is not likely to go away.

There are some legitimate concerns regarding the ways in which the Fedora packaging guidelines are being routed around for Mozilla packages. There are also some odd, seemingly political, questions around the use of APNG, which will require Fedora to either patch APNG into libpng or ignore the "no bundled libraries" rule for Mozilla for the foreseeable future. It is, in short, something of a mess, but not enough of one to send Fedora down the Debian path.

There is hope that some of the other concerns that Kofler and others raised will improve in time. Tyler points to the recent addition of Fedora systems to the Mozilla test farm as a step in the right direction. Previously, CentOS systems, with older versions of the system libraries, were used. Testing on Fedora could well lead to better system integration as well as bundling fewer libraries in the Mozilla packages.

There doesn't seem to be any movement toward weakening the Mozilla trademark requirements, and that policy will always be anti-freedom to some. There are lots of other projects with much looser trademark guidelines; even some high-profile projects like Linux itself. Some may feel that Mozilla is overreaching, but Fedora is in no position to lecture Mozilla about trademark policies. Those who are bothered by those policies will either need to avert their eyes or find another distribution.

to post comments

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 16:30 UTC (Wed) by cry_regarder (subscriber, #50545) [Link] (4 responses)

Mozilla is within its rights to limit trademark usage as is Fedora. Similarly Fedora is within its rights to rebrand Mozilla (like Debian has) and Fedora downstreams are completely free to rebrand Fedora for their re-spins.

If the Fedora community chooses to take the Debian path then that is not "calling out" Mozilla. That is simply exercising free software rights to best accomplish the Fedora mission.

Cry

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 18:01 UTC (Wed) by JoeBuck (subscriber, #2330) [Link] (3 responses)

It would be preferable to work with Mozilla to clean up the worst of the problems. I think that the two top issues are:

Bundling of libraries. Try to work with Mozilla to use libraries provided by the distro to the extent possible, and provide separate packaging of some modified libraries (the APNG issue, for example). This might take some time to work out. If Mozilla refuses to go along with this, then a fork might be needed in the long term, but I think that the right way to go is to have a bundled version (so you can download a working Firefox even for an older distro) and also a build structure that lets distros provide an unbundled version, so the system has only one libpng, libz, xulrunner, etc.
Making the patch approval process less burdensome. The right to fork is a key part of software freedom, but every case where a fork becomes necessary is a kind of failure, it means that there were two groups of people who couldn't come to consensus about what the correct fix is. In general, I think that upstream should always be consulted before distros try to do non-emergency bug fixes, assuming that there is an active upstream, even when it isn't required.

The ability to continue to run well-known programs like Firefox makes it easier to convince people that they can switch to Linux-based systems. Going to iceweasel/icedove is a possible backup plan if nothing can be worked out, but I think it's too early for Fedora to resort to that.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 18:40 UTC (Wed) by jspaleta (subscriber, #50639) [Link] (2 responses)

For bundled libraried...

Chris tyler went into some detail about the path forward for supporting system libraries.

http://thread.gmane.org/gmane.linux.redhat.fedora.devel/1...

Having mozilla internally running unit test on fedora target systems is most certainly going to help.

The patch approval process definitely fell down here. There needs to be a clearer path for patches that are approved by upstream for inclusion in the next mozilla release to be approved for inclusion in downstream packaging in the time between upstream review/approval and upstream release.

-jef

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 7:33 UTC (Thu) by glandium (guest, #46059) [Link] (1 responses)

> Having mozilla internally running unit test on fedora target systems is most certainly going to help.

It won't, unless mozilla builds with all possible --with-system- and --enable-system- flags.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 11:47 UTC (Thu) by ewan (guest, #5533) [Link]

I think the point is that it will enable them to do that if they want to. AIUI the problem is that in many cases the CentOS system libraries are simply too old for Firefox, leaving Mozilla little option to bundle their own. With a Fedora base there's at least the potential to try using the system versions.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 17:23 UTC (Wed) by iabervon (subscriber, #722) [Link] (3 responses)

It seems to me that the trademark policy is not all that different from section 2a of the GPL; if you do make changes, you have to take responsibility for those changes, and it's a logical extension to the case where some of the recipients don't read all of the source files.

Regardless of the legal mechanism, it makes sense to ask who is supposed to be responsible for the pre-installed web browser executables on Linux systems. If Fedora is accepting bug reports and fixing them, they're taking the direct responsibility, and so these executables should be branded in such a way as to let users know this. If the user has downloaded the browser directly from Mozilla, or the distro is just redistributing Mozilla's download, it should be Mozilla-branded.

Of course, most packages don't need a distro branding, but that's because most packages don't have a big user-facing project. Kernel.org, for example, doesn't distribute and maintain built Linux kernel executables that end users are expected to download and start using because they saw a Linux ad in the newspaper.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 18:09 UTC (Wed) by JoeBuck (subscriber, #2330) [Link] (2 responses)

This GPLv2 requirement is frequently ignored: You may modify your copy or copies of the Program or any portion of it ... and copy and distribute such modifications ... provided that ... You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change.

I've seen claims that without such a clause, the GPL wouldn't be valid in countries with a concept of the moral rights of a creator (like France): this kind of thing is to protect the reputation of the original authors, so that if someone comes along and makes a botch of the program, it doesn't reflect on the original creator. However, IANAL so I don't know if this claim is correct.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 21:57 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (1 responses)

Conveniently just using a DVCS like git (and allowing public access to the git repository from which you build any binaries) seems to meet your obligation here unless you imagine a legal opponent will convince a court that the resulting metadata doesn't constitute a "prominent notice" carried by the file. All bets are off on language like that where lay people (in this case people who don't read source code for a living) have no intuition about what's intended.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 0:57 UTC (Thu) by JoeBuck (subscriber, #2330) [Link]

Git certainly has the information about what changed and when, as does any other revision control system, distributed or not. But as soon as you do a release from it, and make a tarball, if that tarball doesn't have changelogs and updated version numbers, you've broken the rules (besides, it's just bad manners and confusing).

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 22:03 UTC (Wed) by roc (subscriber, #30627) [Link] (22 responses)

One thing we (Mozilla) should do is make it clear that any upstream patch approved for a branch landing (but perhaps not yet landed on that branch, or shipped in a release off that branch) is fair game for trademark-using distributors to apply to their own copy of that branch without further approval. This would help grease the wheels.

The MNG vs APNG issue is a joke. MNG is a huge bloaty spec that has never been implemented by any mainstream browser. It's grotesquely over-engineered --- it includes stuff like scaling of image chunks and linear gradients in its "image format". APNG is a much lighter extension to PNG. We submitted patches for APNG to libpng but the libpng owners rejected them because they like MNG. APNG is implemented in Opera as well as Firefox.

I think we would love people to submit patches to upstream to enable use of system libffi and other system libraries. As far as I know that hasn't happened yet.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 22:44 UTC (Wed) by jspaleta (subscriber, #50639) [Link] (2 responses)

Yes i think reading over the bulk of the thread... knowing when an upstreamed patch meets the approval bar for trademark approved downstream use is the fundamental communication breakdown here.

Looking at the bug history...
This was reported on March 5th upstream
This was approved upstream on April 1st.

So if I'm reading what you saying correctly...from Mozilla's POV... this would have been acceptable under trademark policy to use the patch downstream as of the approval on April 1st?

-jef

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 22:59 UTC (Wed) by roc (subscriber, #30627) [Link] (1 responses)

Well, we might not want to say "any patch on trunk can be backported to any branch, no problem" because that can often have undesirable side effects.

But I think we should certainly say that any patch approved to land on a particular branch (but not yet landed there) can be applied by anyone packaging that branch. That alone wouldn't have helped here since there was an extra problem --- the branch approval request was ignored for four weeks :-(. That was simply a mistake, people should never sit on requests for long. But if you read the Mozilla bug --- https://bugzilla.mozilla.org/show_bug.cgi?id=550455 --- there's not much communication from Fedora there. Just one comment on the 23rd saying this matters to Fedora, and the patch was branch-approved five days later (which is definitely too long, but not outrageous IMHO). A more detailed hurry-up nag, especially if it happened earlier, would have really helped.

So what should have happened IMHO is as soon as the Fedora packagers realized they needed this patch, they should have commented in the bug "our users are hurting and we need branch approval for this patch NOW NOW NOW", that approval should have been granted immediately, and then via a previously clearly-communicated blanket approval to use such patches with our trademarks, Fedora would have packaged the fix.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 23:22 UTC (Wed) by jspaleta (subscriber, #50639) [Link]

"Well, we might not want to say "any patch on trunk can be backported to any branch, no problem" because that can often have undesirable side effects."

Granted, the question is how can what is allowed be more obvious and more explicit so its less likely someone will be waiting for approval mistakenly?

As for the unnecessary delay before it was made clear that this was a wide impactor. I'm sure an invitation for people to nag in upstream tickets more won't go...unrewarded. Something for the day-to-day Fedora Mozilla package maintainers and triagers and possible FESCO to be aware of.

-jef

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 4:04 UTC (Thu) by roc (subscriber, #30627) [Link] (10 responses)

I am informed that we have libffi changes that need to go upstream before we can use "system libffi". That is being worked on. Also, someone has submitted system-libffi patches and they are being reviewed and will be merged. So the libffi problem, at least, is being resolved.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 5:51 UTC (Thu) by pabs (subscriber, #43278) [Link] (8 responses)

Shouldn't you be getting these changes merged upstream first and once they have been merged and released, then start depending on the version that includes them?

Autoconf is for this exact needs...

Posted Apr 29, 2010 6:15 UTC (Thu) by khim (subscriber, #9252) [Link]

These two processes can go in any order. Correctly written autoconf will just refuse to use library without changes so till upstream will include fix the autoconf will refuse all upstream releases.

But of course it's good idea to wait till upstream includes the patch at least in VCS: this will mean there WILL be supported release in the future...

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 8:05 UTC (Thu) by roc (subscriber, #30627) [Link] (6 responses)

Getting changes merged upstream first would slow down development a lot.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 11:49 UTC (Thu) by ewan (guest, #5533) [Link] (1 responses)

Fedora has a strong policy of 'upstream first' and, to say the least, it doesn't generally get people complaining that the pace of change is too slow.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 23:48 UTC (Thu) by roc (subscriber, #30627) [Link]

Fedora is not building a large application on top of those upstream libraries that needs features added to those libraries.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 13:51 UTC (Thu) by ccurtis (guest, #49713) [Link] (3 responses)

Perhaps it would help if you posted a message like "our project is hurting and we need a release with this patch NOW NOW NOW" to their bug tracker.

Fedora, Mozilla, and trademarks

Posted May 2, 2010 3:31 UTC (Sun) by jwalden (guest, #41159) [Link] (2 responses)

Someone please correct me quickly if I'm wrong, but my understanding -- just from hearing people talking about it passing over lunch, or something like that, and this is the haze of a memory, so maybe I didn't hear correctly -- is that libffi isn't sufficiently strongly maintained to permit this.

I would also note (conveniently, one might say, but I believe accurately) that the browser market is a uniquely competitive one, as far as distribution-packaged software goes. What other distro package sees such widespread use, where time-to-market matters as much? (I'm speaking to the benefits of quick iteration through in-tree, lightly-patched third-party code now, not to the backport approval process.) I just skimmed my Applications menu in Fedora, and I can't see anything where new functionality and features have such high demand as for browsers.

Fedora, Mozilla, and trademarks

Posted May 3, 2010 20:12 UTC (Mon) by atgreen (guest, #33284) [Link] (1 responses)

I'm the upstream libffi author/maintainer.

Dan Witte, from Mozilla, has been terrific about feeding back libffi patches over the past few months. But from what I recall they mostly have to do with building libffi on Windows and OS/2. I don't remember anything that would prevent them from using the system libffi on Linux.

In any case, this sudden batch of contributions from Dan and Mozilla is appreciated.

Anthony Green

Fedora, Mozilla, and trademarks

Posted May 3, 2010 22:19 UTC (Mon) by jwalden (guest, #41159) [Link]

Ah, yeah, that's what I actually heard -- Windows support being the big thing. As I said, the memory was quite hazy. :-)

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 7:32 UTC (Thu) by glandium (guest, #46059) [Link]

> I am informed that we have libffi changes that need to go upstream before we can use "system libffi"

On the 1.9.2 branch, there are no upstream libffi changes required on linux to be able to use system libffi. Though one may wonder why js-ctypes is enabled on 1.9.2 at all, since it's only advertised for 1.9.3.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 4:59 UTC (Thu) by tzafrir (subscriber, #11501) [Link] (2 responses)

I just wonder what would happen if libpng would have had the same branding requirements as Mozilla. Surely upstream knows best.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 8:05 UTC (Thu) by roc (subscriber, #30627) [Link] (1 responses)

Then Mozilla wouldn't be able to use the libpng trademark. I think we'd deal.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 8:44 UTC (Thu) by tzafrir (subscriber, #11501) [Link]

As someone reading this through IceWeasel, I fully agree with you that this would be the wise thing to do.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 9:15 UTC (Thu) by niner (guest, #26151) [Link] (4 responses)

"MNG is a huge bloaty spec that has never been implemented by any mainstream browser."

If I am not mistaken it actually has. And ironically enough it was Mozilla that once upon a time had this support. It has been checked in 2000-06-12.

https://bugzilla.mozilla.org/show_bug.cgi?id=18574

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 23:51 UTC (Thu) by roc (subscriber, #30627) [Link] (2 responses)

OK OK. MNG is a huge bloaty spec that has never been shipped in any mainstream browser.

Fedora, Mozilla, and trademarks

Posted Apr 30, 2010 0:44 UTC (Fri) by foom (subscriber, #14868) [Link] (1 responses)

I couldn't care less about MNG, but...

According to https://bugzilla.mozilla.org/show_bug.cgi?id=18574#c653:
"It shipped in Mozilla 1.0 through 1.3, and Netscape 7.0, and had trivial usage on the web, and was
backed out before Mozilla 1.4 and Firebird 0.6.1."

Perhaps mozilla didn't really count as a mainstream browser back in 2001-2003? :)

Fedora, Mozilla, and trademarks

Posted Apr 30, 2010 4:47 UTC (Fri) by lambda (subscriber, #40735) [Link]

No, it didn't. If I remember correctly, Mozilla had maybe 1-2% market share back then, making it pretty far from mainstream. Those were pretty dark days, when there really was only one mainstream browser.

These days, there are 4 browsers I'd consider to be "mainstream" (IE, Firefox, Safari, and Chrome), and one, Opera, on the edge (in some countries, it has enough marketshare to be considered mainstream, and their mobile browser seems to be pretty popular).

Fedora, Mozilla, and trademarks

Posted Apr 30, 2010 0:34 UTC (Fri) by bronson (subscriber, #4806) [Link]

That was ten years ago! It seems pretty clear that MNG standard is stone dead.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 22:15 UTC (Wed) by wonder (guest, #64293) [Link] (1 responses)

why not shipping the packages unbranded? we are doing that in arch for a long time.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 14:25 UTC (Thu) by NAR (subscriber, #1313) [Link]

Then you can't put a nice shiny Firefox logo on your desktop which would reassure the user coming from e.g. Windows that there's at least one piece of software on her new netbook she's familiar with.

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 22:27 UTC (Wed) by roc (subscriber, #30627) [Link]

It looks like the problem with merging openSUSE's KDE integration work upstream is simply that they're not pushing it. There's only one open bug with a patch, and it's had plenty of attention from the Mozilla side...
https://bugzilla.mozilla.org/show_bug.cgi?id=526717

Fedora, Mozilla, and trademarks

Posted Apr 28, 2010 23:25 UTC (Wed) by TRS-80 (guest, #1804) [Link] (6 responses)

Ubuntu is going the other direction, with plans to "Use in-source libraries rather than system libs" and eliminate users of xulrunner by porting them to webkit. So much for XUL being a platform - Mozilla's Firefox focus has backfired.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 1:13 UTC (Thu) by joedrew (guest, #828) [Link] (3 responses)

The in-tree versions of our libraries are the best tested, both in terms of performance and correctness. Linking to arbitrary versions of Cairo, for example, can cause regressions in any number of areas, in ways that we can't predict because our hundreds of thousands of tests haven't been run against Mozilla + that particular library.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 2:14 UTC (Thu) by TRS-80 (guest, #1804) [Link] (1 responses)

Why can't distributions run those tests themselves? I know the Debian maintainer wants to, given his bugs about the testsuite not being freely licensed.

I'd forgotten that LWN had covered this topic so recently in Applications and bundled libraries. Jake did link it in this article, but only as a single word talking about a FESCo member's position so I skipped over it, probably thinking it was another mailing list post.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 2:52 UTC (Thu) by joedrew (guest, #828) [Link]

They can! And, as you note, Mike Hommey is moving towards doing so for Debian, at least for our unit tests (getting reliable perf data would be involved at the minimum you'd need the same machines we have running our perf tests. But it's doable!). And I'm sure that passing all our tests would give the folks involved in making the trademark decisions a lot of faith in the changes distributions have made!

But without that, and a commitment to keep running the tests (when you upgrade a library from under Mozilla, you have to make sure you keep things working!), justifying using an external library is a lot harder.

Fedora, Mozilla, and trademarks

Posted May 9, 2010 20:34 UTC (Sun) by damentz (subscriber, #41789) [Link]

This has happened with Iceweasel already; I've needed to pull from the Firefox ppa on Ubuntu because Iceweasel crashes more frequently for unexplainable reasons and is out of date more often than not.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 1:26 UTC (Thu) by The_Barbarian (guest, #48152) [Link]

Posted by TRS-80:
>Ubuntu is going the other direction, with plans to "Use in-source libraries rather than system libs"

Well, I can't say that such a decision surprises me, coming from Canonical. They can be almost as silly as Mozilla.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 9:14 UTC (Thu) by xnox (subscriber, #63320) [Link]

http://packages.ubuntu.com/search?keywords=chromium-browser

I wonder why that was uploaded, when system-libs & copyrights are not yet compltly sorted out. I bet it's aimed at replacing firefox.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 13:26 UTC (Thu) by NRArnot (subscriber, #3033) [Link] (4 responses)

Would it be legally defineable for Mozilla (or any other trademark owner) to permit unapproved patches, if and only if they do not change the user interface? In other words, the patched version and the official version appear and behave exactly the same to the user when they don't crash or hang?

Presumably it could also be possible to design logos with such patching in mind. A Firefox icon with a strip for a "patched by <distro name>" message/ warning / disclaimer. That's similar in concept to the "tainted" kernel you get when you load closed-source modules.

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 15:12 UTC (Thu) by cry_regarder (subscriber, #50545) [Link]

frame problem.

Cry

Fedora, Mozilla, and trademarks

Posted Apr 29, 2010 23:52 UTC (Thu) by roc (subscriber, #30627) [Link] (2 responses)

It's not just the UI that matters. Changes to Gecko can easily cause Web compatibility problems. "Site X works in Firefox 3.6 over there but not here" is one problem we're trying to avoid.

Fedora, Mozilla, and trademarks

Posted Apr 30, 2010 6:56 UTC (Fri) by glandium (guest, #46059) [Link] (1 responses)

As someone maintaining Gecko in a large distribution for years, I can tell you the vast majority (if not all) of the cases like "Site X works in Firefox 3.6 over there but not here" nowadays come from the fact that here is not Firefox but something named differently (guess why). Not from patching.

There was a time when we had problems, and they weren't due to patching, but to enabling an unsupported feature, namely the pango backend, at a time where both backends had their drawbacks.

Fedora, Mozilla, and trademarks

Posted Jul 1, 2010 9:12 UTC (Thu) by oak (guest, #2786) [Link]

Yea, there are quite a few www-sites that check browsers by the name they report and refuse to work if browser's name isn't one they explicitly check for.

(I've recently seen this in one company's internal SAP www-service and many years ago one large scandinavian bank had this kind of a beginner mistake on its public site main page, it gave a completely blank page e.g. in Konqueror...)