Twitter Updates

    follow me on Twitter

    Twitter Updates

      follow me on Twitter

      « Joga.com Companion Update | Main | Quick Update »

      TrackBack

      TrackBack URL for this entry:
      http://www.typepad.com/services/trackback/6a00d83451eebe69e200e5507f3c6f8834

      Listed below are links to weblogs that reference Mozilla, Trademarks and Debian:

      Comments

      Axel Hecht

      I'd like to add that Firefox is not the first program getting into arguments with linux distributions.

      The cdrecord package got so whacked by distros that the author declined to support it further, see his side of the world (or fragments of that) on http://cdrecord.berlios.de/old/private/cdrecord.html. Went through the press in Germany quite a bit, not sure how it's on the other side of the ocean.

      Anyway, with the growing spread of FOSS, linux distros are increasingly forced to get their act together, and either fork a project, or not. Instead of forking happily and dropping the support requests on the original authors.

      Adam

      The right to keep the name of a derivative program is almost as important as the right to make a derivation in the first place.

      The point of being able to making a derivation of a program is to make it fit your needs and the needs of other users better.

      "grep" is a useful tool by itself, as are most of the basic Unix tools enshrined in the POSIX standard. From the command line, these tools allow you accomplish a huge variety of tasks which I'm sure I don't need to go into here. The right to modify grep, to fix bugs, to add new features, and to share these improvements with your neighbour, is freedom 3 as defined by the Free Software Definition.

      However, individual programs do not exist in a vacuum. "grep" - as you are probably aware - is not only invoked directly from the command line; it is also a vital component of the shell script toolbox, one of things that makes Unix OSs so powerful.

      Now imagine if, when improving grep, you had to make your program under another name. Say, "mygrep". Yes, you could fix bugs, and share these with your neighbour, as GPL is intended to let you do. What would be the point? You've got tens, hundreds (maybe thousands?) of shell scripts on your system that use grep, but none of them can take advantage of your bug fixes as they all call "grep", and not "mygrep".


      "grep", and the rest of the POSIX toolbox, might be a contrived example. POSIX tools have to be given a particular name, that's defined by the standard, they are generic names already. OK, what about libxml? libkdecore? What's the point of having the ability to fix and share fixes to the core of kde if no-one else can use them with the rest of their KDE system, as the rest of their system is not going to work with libmykdecore? Those are libraries - that also might be too contrived. What about another web browser, like "links". "links -dump" is pretty useful - what if you had other programs relying on that, either in shell scripts, or even something like a filter for a GUI text editor that takes a URL, does "links -dump" and inserts the result into your text document?

      What's the point of being able to fix bugs and share those fixes if no-one can use your modifications without having to track down every place in the rest of their system where that program might have been used, and all the places in each new bit of software they get that might use the program, before those changes can take effect?


      Heck, what about Firefox? One of the great things about Firefox is the extension mechanism. I'm using over 10 extensions in FF now, and they all rule, and I wouldn't want to use FF without them.

      FF extensions are FF-version specific. 1.0 extensions needed updating for 1.5, and I'm pretty sure 1.5 extensions won't work with 1.0 or the upcoming 2.0. So, all extensions do version checks. If Debian are forced to change the name of FF to apply integration fixes, but extensions do, say, a version check to match the regex "^Firefox 1\.5\..*" and Iceweasel returns "Iceweasel 1.5.0.7-4", then the extension might refuse to work. So although the extension might be otherwise completely compatible with Iceweasel, the requirement to change the name of the program could make other programs that depend on that program stop working.

      What about X programs that look through windows lists and do things based on window titles? Suddenly that KWin "Disable focus stealing prevention for Firefox" option might not work so well.

      How useful are your bugfixes now? Are they worth all the other stuff you've just broken? Do you *know* everything you've broken yet?

      Program names are important because they're meaninful to more than just people. Without being able to keep them constant, the power of the right to make changes and share them with your neighbour is so greatly reduced, in some cases it becomes pointless.

      Wladimir Palant

      Adam, you don't understand. Mozilla Corporation gained lots of trust by providing a high-quality application. It took years. Now people come to expect this quality from Firefox. And there are processes in place to make sure that this level of quality can be provided by MoCo. But what about Debian? Can they guarantee the same quality level? Maybe. Or maybe not. Who can't tell? Fact is, if Debian developers or even downstream developers screw up - who will get the blame? Who will loose the trust that was so difficult to earn? Right, Mozilla Corporation. And I can fully understand the decision that Mozilla Corporation cannot be responsible for the changes performed by Debian. If Debian needs to do the changes it should rename the product and remove the trademarks - so that it is a Debian product, Debian is doing support and Debian is responsible for anything that fails.

      As to your examples of how the name is essential - they are all flawed. Firefox is not grep, it isn't a common library either. Any application that depends on the name deserves to be broken - since the same application will not work correctly with Bon Echo, Minefield or Flock, even though these aren't essentially different from Firefox. There is even a browser called lolifox if want to go exotic.

      That's also the reason why no extension will break if the name of the application changes. Most extensions have been tested in either Bon Echo or Minefield (nightly builds) and don't rely on names. Version checking is done by the extension manager that identifies the application by its GUID, no names.

      extensions

      Adam, your example of extensions is demonstrating the problem very well: what happens if some code changes by Debian are altering the way extensions work and the compatibility of extensions? People would try to install extensions from addons.mozilla.org (seeing this is the "official" extension website for Firefox) and they would rightfully be annoyed if it doesn't work with the Debian-version of Firefox - you could ask whether Firefox without working extensions would still be Firefox...
      Besides this, trademark law _requires_ Mozilla Corporation to take some measures in order to protect their tratdemarks. Yes, that's annoying, but hardly anything you can avoid without changing the law (good luck!) or not using trademarks at all. If you have no trademark to define your product, it's really hard to build a successfull brand, like Mozilla has done with Firefox.

      raiph

      "it's really hard to build a successfull brand, like Mozilla has done with Firefox."

      Careful. It is indeed really hard to build a successful brand. While you guys might argue internally that MoCo was the cause, and fan activity a mere effect, you would be well advised to keep that view to yourselves. In contrast, I think a good case could be made that it's the fans that did most of the brand building; that the key factor in that fan force was its open source spirit; and that MoCo is in debt to the open source community. Treat the brand as a MoCo asset, "spiritually" speaking, at your (and our) deep peril.

      Adam

      Yeah, I guess that building a good brand, releasing a high-quality product, protecting a trademark and making sure that downstream changes don't overburden you with irrelevant bug reports are incompatible with Free Software.

      Which is why KDE, Gnome, the Linux kernel, MySQL, Python, etc... all have the same problems. Oh, no, they don't.

      Hey, maybe it is just them. Maybe you guys are the only people with competent legal counsel, and all the legal experts belonging to all those other projects have got it wrong, and releasing 99+% of the exact same code with a few minor changes will turn "Firefox" into a generic term for "Web browser".

      Look, if you don't want people to change the name of the product, then you could put that in the License. You wouldn't be the first - the Latex Project Public License has such a clause. It's even a Free Software license according to the FSF.

      If you don't want to release Free Software, that's fine too. No-one's making you. But you've put Mozilla and Firefox under the GPL, the whole *point* of which is to give it's users certain Freedoms, such as the ability to make their own bugfixes and enhancements, and to make those bugfixes and enhancements available to anyone else *as a drop-in replacement for the original*.

      (Yes, Iceweasel *might* screw up extensions. My alterations to grep *might* screw it, and a bunch of scripts that rely on it, up. But that shouldn't stop me being *able* to make those alterations available)

      You decided to release Mozilla and Firefox under the GPL of your own free will. No-one forced you. But you guys are the ones adding extra restrictions here that almost no other Free Software project - trademarked or no - is. *You* are the odd ones out. Don't get surprised if people who know the GPL expect certain rights from software licensed under it, and have a hard time figuring out why they can't treat this particular GPLd product the same as all the others.

      Maybe you should stop releasing under the GPL/LGPL. That would clear up a lot of confusion.

      Anonymous

      > Which is why KDE, Gnome, the Linux kernel, MySQL, Python, etc... all have the same problems. Oh, no, they don't.

      Hmm, of all users http://www.w3schools.com/browsers/browsers_stats.asp 27.3% use Firefox compared to 3.5% which use Linux (I would assume that Gnome/KDE share 90% of these), so it is obvious that they must protect their Trademark more than others .

      It should be kept in mind that none of these entities allow usage of their trademark for anything harming their reputation, something the Debian browser incorrectly dupped Firefox does...

      Anonymous

      > That would clear up a lot of confusion.
      There is no confusion except for those people that can't read: http://lxr.mozilla.org/mozilla/source/configure#123

      Adam

      Odd, I thought that you had to protect *every* abuse of trademarks that you were aware of, not just selective ones if you didn't think you had enough market share to be important.

      And although 60+% of web /servers/ run Linux, and a not insignificant proportion of those will be running MySQL, your argument still doesn't address those organisations.

      Still, I'm sure you're right and "Linux" is on its way to becoming a generic term for "operating system kernel" and "MySQL" is similarly becoming synonymous with "DBMS".


      IMO, it's a pity that the Free version is going to be called Iceweasel. My vote was for Freefox.

      extensions

      @raiph:
      I'm not a guy at MoCo and I'm fully aware that support from fans has a huge part in the success of Firefox. It wasn't my intent to paint this as a MoCo only succeess:-)

      @Adam:
      You are still free to enjoy all the freedoms the GPL provides. All MoCo asks you (as I understand it) is to not use their trademarked name and artwork if you make substantial changes. This sounds completely reasonable to me. The fact that you can fork it and give it a new name shows that the fundamentals of free software are not affected.

      Chris Davis

      So... why is it in Ubuntu you don't see the firefox logo...just the globe part? I have the real one on Windows and on Mac... even SuSe has one I think...

      pd

      Debian is dead anyway, at least in desktop terms. It's now the equivalent of a highly sophisticated DOS where Debuntian is now the Windows 95 shell.

      I agree with MoFoCo completely. If I want to run Firefox, I should be able to download and run it the same on all platforms.

      Anonymous

      > Odd, I thought that you had to protect *every* abuse of trademarks that you were aware of, not just selective ones if you didn't think you had enough market share to be important.

      Yes, but the point is that there is no one wanting these trademarks, otherwise they would already have lost it.

      Kelson

      Adam, I see what you mean. It would be a shame if someone wrote an improved version of such a ubiquitous tool like "more" and decided to call it something else, like "less."

      Wait a minute...

      Adam

      Extensions> No, I'm not. I'm not free to make as limited set of changes as I wish, such as changing one bit of code that does auto-updates, and not changing the bit of code that says 'char const * product_name = "Firefox";'.

      The name of a product is code. It is entirely defined by code just as much as any other string literal or bit of functionality. If you don't believe me, try changing the name of Firefox without changing any of the code that says "Firefox". Alternatively, try changing every bit of code that says "Firefox" to something else, say "RandomBrowser", and claiming the resulting browser is not called "Firefox".

      Mozilla has granted me, via paragraph 2 of the GPL, the right to make whatever changes I choose to Firefox's code, and to redistribute those changes to whoever I choose, subject to a few very limited exceptions. None of those exceptions say anything remotely like "you must, at the request of the author, change some bits of text when you make other completely unrelated changes, if the author decides they don't like those changes". In fact, such a clause would be almost antithetical to the ideas behind the GPL.

      That Mozilla choose to use an unrelated set of laws (i.e. Trademark) to remove this right/add this restriction (that a non-author would be forbidden from doing under section 6) that is otherwise granted by the GPL is, well, ... self-contradictory? Hypocritical? Certainly weird.


      And I can't use a forked version as a drop-in replacement for FF, as many of the extensions which make FF rock so goddamn much may not work if they do a compatibility check that includes the product name as well as version. I figure that to make sure that they're either running on Mozilla 1.8.0+ or Firefox 1.5.0+, they pretty much have to do this. So, guess what, I'm also pretty sure they won't work with Iceweasel.

      In which case I lose AdBlock, the Web Developer toolbar, Cookie Button, Named anchors, Tab preview, Resizeable textarea, Link Widgets, etc..., and may as well just use Konqueror instead.


      Kelson> No, it would have been a shame if they didn't have the *option* of calling it "more".

      Anonymous

      Adam:

      You have some serious misunderstanding here. You can change any portion of the source code (as long as it is legal) and even redistribute it, mozilla for example has still many source references to Phoenix ( http://lxr.mozilla.org/mozilla/search?string=phoenix ) and Firebird.

      What you are not allowed to, is to claim that it, nor the resulting product, is Firefox (neither would you be allowed to call it Debian Browser or Internet Explorer, unless you are the Trademark holder). The same applies to Konqueror or any other application...

      > "you must, at the request of the author, change some bits of text when you make other completely unrelated changes, if the author decides they don't like those changes"
      No one has asked for such. By default compiling firefox produces a trademark-free unbranded browser (Debian itself doesn't have such a mechanism). Debian has chosen to modify the code in such a way that the resulting application makes trademark infringements and Mozilla asked them to stop doing it.

      > much may not work if they do a compatibility check that includes the product name as well as version.
      There isn't such a check as you've been told already, and even if it were you could simply modify the application to allow those anyway.

      Adam

      *double checks*

      Sorry - I misread the explanation of how extensions check for product versions first time through. Must have been in too much of a rush to reply to the other parts of that post. Thanks for the clarification.

      As for whether other producers allow derivative works of the same name - I'm pretty sure I'm right about that, but I'm doing some digging into it. Will post findings if I get any...

      Brennender Fuchs

      I don't think it would be much of a problem to tell Extensions they're loaded up on "Firefox x.y" while displaying the string "Iceweasel x.y-zz" on screen. IANAL but I don't think you have much of a chance to prevent that legally. The same thing has btw been mulled over by the ReactOS people (www.reactos.com) who are building a binary compatible WNT-W2k-XP compatible OS. They have to include Strings like "Microsoft" or "Windows" in their system for example because of the HKEY_LOCAL_MACHINE\Microsoft registry key a lot of Windows software expects to be there. As far as the name "Iceweazel" goes I would call it Brennender Fuchs der ueber den faulen Hund steigt 1.5, but sigh, they didn't go for that.

      atom probe

      Is the Community Edition Policy a reasonable compromise?

      http://www.mozilla.org/foundation/trademarks/community-edition-policy.html

      Joshua

      I would just like to point out a few things. Calling your program "Iceweasel" does not mean that you can't make the program be labeled as something else for compatibility purposes. What would be wrong with using a program named "firefox-bin" to run Iceweasel? If the program itself doesn't identify as firefox, I don't see what problem there would be with a situation like that.

      You could be more cooperative

      For a group such as Debian, couldn't you just give them a license to use the trademark in good faith? Their aims are simply to provide the best version of firefox that they can, without changing the essence of it. I'm sure the two groups can come to an understanding that if Debian packages something inappropriately they'd have to rectify the situation on pain of losing the courtesy license.

      I understand the need to keep the trademark safe, but I'm sure you can extend a little more trust to Debian than you'd extend to J.Random malware author.

      Matthew Garrett

      I'm unclear on why you can't just grant a blanket revocable trademark license to derivative works, with the condition that any changes that the Mozilla Foundation believe to be excessive will result in the revocation of the trademark license? My (vague) understanding of trademark law is that it's not a problem for lots of derivatives to be using the trademark providing those uses are all licensed. In general, it seems that even a fairly token effort at enforcement is sufficient to prevent genericide - which is presumably what you're actually worried about.

      MD

      Joshua, that would be a trademark violation -- that's what would be wrong with that.

      Glo

      It would seem to me that the Mozilla organization in releasing Firefox under the GPL intentionally has forfeited the trademark rights it claims on its web site. Further, that even a registered trademark is worthless in the face of releasing Firefox under the GPL.

      But wait! IANAL. Maybe the mozilla folks should ask SCO how things are going.

      Synthesis

      Why THE HELL don't Debian & Ubuntu just use the pretty Firefox logos?

      I have never understood why they change it to that Blue Globe.

      I agree with Mozilla on this, different logos and icons are confusing for end users!

      Verify your Comment

      Previewing your Comment

      This is only a preview. Your comment has not yet been posted.

      Working...
      Your comment could not be posted. Error type:
      Your comment has been posted. Post another comment

      The letters and numbers you entered did not match the image. Please try again.

      As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

      Having trouble reading this image? View an alternate.

      Working...

      Post a comment