1:39 p.m.
Hi,
I'm concerned about bz 579023 [1] which is a Thunderbird crasher bug.
This bug was fixed upstream [2] for about 3-4 weeks. I ran a thunderbird
koji build version [3] with an adapted version of that patch since then
without any problems. Other users confirmed that this patch fixes their
problems as well.
The Fedora bug has a number of duplicates with quite some number of cc'd
users so I guess a lot of people experiencing these crashes.
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?
fs
[1] https://bugzilla.redhat.com/show_bug.cgi?id=579023
[2] https://bugzilla.mozilla.org/show_bug.cgi?id=550455
[3] http://koji.fedoraproject.org/koji/taskinfo?taskID=2092397
1:43 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/22/2010 02:39 PM, Felix Schwarz wrote:
I'm concerned about bz 579023 [1] which is a Thunderbird crasher
bug.
This bug was fixed upstream [2] for about 3-4 weeks. I ran a thunderbird
koji build version [3] with an adapted version of that patch since then
without any problems. Other users confirmed that this patch fixes their
problems as well.
The Fedora bug has a number of duplicates with quite some number of cc'd
users so I guess a lot of people experiencing these crashes.
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?
+1! This bug is brutal. TB crashes several times a day for me.
- Mike
3:27 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Thu, Apr 22, 2010 at 1:39 PM, Felix Schwarz
<felix.schwarz(a)oss.schwarz.eu> wrote:
I'm concerned about bz 579023 [1] which is a Thunderbird crasher bug.
This bug was fixed upstream [2] for about 3-4 weeks. I ran a thunderbird
koji build version [3] with an adapted version of that patch since then
without any problems. Other users confirmed that this patch fixes their
problems as well.
The Fedora bug has a number of duplicates with quite some number of cc'd
users so I guess a lot of people experiencing these crashes.
And probably a lot of people aren't.
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?
Why isn't Mozilla releasing a new version that contains the fix?
And even if the Fedora Thunderbird maintainer decides to push a
patched package to F-12, there's no way it should be pushed directly
to stable.
--
Jeff Ollie
7:30 p.m.
Jeffrey Ollie said the following on 04/22/2010 01:27 PM Pacific Time:
On Thu, Apr 22, 2010 at 1:39 PM, Felix Schwarz
<felix.schwarz(a)oss.schwarz.eu> wrote:
>
> I'm concerned about bz 579023 [1] which is a Thunderbird crasher bug.
> This bug was fixed upstream [2] for about 3-4 weeks. I ran a thunderbird
> koji build version [3] with an adapted version of that patch since then
> without any problems. Other users confirmed that this patch fixes their
> problems as well.
>
> The Fedora bug has a number of duplicates with quite some number of cc'd
> users so I guess a lot of people experiencing these crashes.
And probably a lot of people aren't.
That's rather dismissive considering the number of bug reports:
http://bit.ly/9uUdGx
It got so bad for me that I went back to the GA version for Fedora 12.
> 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?
Why isn't Mozilla releasing a new version that contains the fix?
And even if the Fedora Thunderbird maintainer decides to push a
patched package to F-12, there's no way it should be pushed directly
to stable.
An unofficial patched version would be better than the current situation.
John
7:45 p.m.
On 04/22/2010 08:30 PM, John Poelstra wrote:
An unofficial patched version would be better than the current
situation.
John
FWIW - I use the nightly builds from mozilla.org .. no crashes at all
with 3.1 nightly - be aware if you use enigmail, that versions after the
4/05 build do not work (so far as of 4/22).
You might want to try an upstream build of 3.0.x see if it fares better
for you.
Also, 64 bit firefox/thunderbird may not work properly - I have found
running the 32 bit version to be less problematic.
YMMV
gene/
8:18 p.m.
Jeffrey Ollie wrote:
Why isn't Mozilla releasing a new version that contains the fix?
I don't know, but in the absence of a new release, the maintainer is
supposed to backport the fix. 3-4 weeks or more is not a nice response time.
And even if the Fedora Thunderbird maintainer decides to push a
patched package to F-12, there's no way it should be pushed directly
to stable.
Why not, if it contains an important fix?
Kevin Kofler
10:13 p.m.
On Fri, 2010-04-23 at 03:18 +0200, Kevin Kofler wrote:
> And even if the Fedora Thunderbird maintainer decides to push a
> patched package to F-12, there's no way it should be pushed directly
> to stable.
Why not, if it contains an important fix?
Because if we don't test the updated package we don't know:
a) does it actually fix the bug?
b) does it break anything else?
Just because an update 'contains an important fix' doesn't mean it's
fine to release it without testing.
However, I certainly agree this should be fixed ASAP, and saying 'well
it doesn't crash for EVERYONE' and 'upstream should fix it instead' are
pretty weak excuses.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
12:20 a.m.
On Apr 22, 2010, at 9:13 PM, Adam Williamson wrote:
On Fri, 2010-04-23 at 03:18 +0200, Kevin Kofler wrote:
>> And even if the Fedora Thunderbird maintainer decides to push a
>> patched package to F-12, there's no way it should be pushed directly
>> to stable.
>
> Why not, if it contains an important fix?
Because if we don't test the updated package we don't know:
a) does it actually fix the bug?
b) does it break anything else?
Just because an update 'contains an important fix' doesn't mean it's
fine to release it without testing.
I'll test it. I don't need it in stable right away...
2:03 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Hi,
we're patching mozilla packages only for really critical issues because
of mozilla trademarks. We can't put any patch we want to the mozilla
package and ship it as 'Firefox' or 'Thunderbird'.
I've asked for inclusion at upstream bug,
https://bugzilla.mozilla.org/show_bug.cgi?id=550455, if you want to
speed up the process you can reply there too.
ma.
On 04/22/2010 08:39 PM, Felix Schwarz wrote:
Hi,
I'm concerned about bz 579023 [1] which is a Thunderbird crasher bug.
This bug was fixed upstream [2] for about 3-4 weeks. I ran a thunderbird
koji build version [3] with an adapted version of that patch since then
without any problems. Other users confirmed that this patch fixes their
problems as well.
The Fedora bug has a number of duplicates with quite some number of cc'd
users so I guess a lot of people experiencing these crashes.
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?
fs
[1] https://bugzilla.redhat.com/show_bug.cgi?id=579023
[2] https://bugzilla.mozilla.org/show_bug.cgi?id=550455
[3] http://koji.fedoraproject.org/koji/taskinfo?taskID=2092397
2:18 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 09:03 AM, Martin Stransky wrote:
Hi,
we're patching mozilla packages only for really critical issues because
of mozilla trademarks. We can't put any patch we want to the mozilla
package and ship it as 'Firefox' or 'Thunderbird'.
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.
/me ducks and hides for cover.
Ralf
2:24 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 09:18 AM, Ralf Corsepius wrote:
On 04/23/2010 09:03 AM, Martin Stransky wrote:
> Hi,
>
> we're patching mozilla packages only for really critical issues because
> of mozilla trademarks. We can't put any patch we want to the mozilla
> package and ship it as 'Firefox' or 'Thunderbird'.
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.
/me ducks and hides for cover.
No, you get it wrong. It's about cooperation, we work with upstream to
release one valid product. See the upstream bug, the fix may be included
in next security update.
3:46 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Fri, Apr 23, 2010 at 9:24 AM, Martin Stransky <stransky(a)redhat.com> wrote:
On 04/23/2010 09:18 AM, Ralf Corsepius wrote:
> On 04/23/2010 09:03 AM, Martin Stransky wrote:
>> Hi,
>>
>> we're patching mozilla packages only for really critical issues because
>> of mozilla trademarks. We can't put any patch we want to the mozilla
>> package and ship it as 'Firefox' or 'Thunderbird'.
>
> 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.
>
> /me ducks and hides for cover.
No, you get it wrong. It's about cooperation, we work with upstream to
release one valid product. See the upstream bug, the fix may be included
in next security update.
...*may be included* in next security update.
Well, Ralf is right. That situation is just sick. To have a patch that
fixes a crashing application but it can't be applied, because of
Trademark/Branding problems. And even worse, that the app has to crash
for *everyone* to get it faster.
But i guess it's better i shut up, since i don't use Mozilla products.
--
LG Thomas
Dubium sapientiae initium
2:28 a.m.
Thomas Janssen wrote:
...*may be included* in next security update.
Well, Ralf is right. That situation is just sick. To have a patch that
fixes a crashing application but it can't be applied, because of
Trademark/Branding problems. And even worse, that the app has to crash
for *everyone* to get it faster.
Indeed, this is completely unacceptable.
Kevin Kofler
4:34 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 09:24 AM, Martin Stransky wrote:
On 04/23/2010 09:18 AM, Ralf Corsepius wrote:
> On 04/23/2010 09:03 AM, Martin Stransky wrote:
>> Hi,
>>
>> we're patching mozilla packages only for really critical issues because
>> of mozilla trademarks. We can't put any patch we want to the mozilla
>> package and ship it as 'Firefox' or 'Thunderbird'.
>
> 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.
>
> /me ducks and hides for cover.
No, you get it wrong.
Well, c.f. freedom 3 on http://www.gnu.org/philosophy/free-sw.html
You told us, you can't modify the sources and ship modified binaries
=> thunderbird and firefox are non-free, because of the trademarks
Mozilla apply.
=> These packages should not be part of Fedora.
It's about cooperation, we work with upstream to
release one valid product. See the upstream bug, the fix may be included
in next security update.
You are confusing their product with yours: Your product is broken and
you are unable to maintain it because of legal reasons.
Ralf
2:28 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Ralf Corsepius wrote:
Well, c.f. freedom 3 on http://www.gnu.org/philosophy/free-sw.html
You told us, you can't modify the sources and ship modified binaries
=> thunderbird and firefox are non-free, because of the trademarks
Mozilla apply.
=> These packages should not be part of Fedora.
+1
You are confusing their product with yours: Your product is broken
and
you are unable to maintain it because of legal reasons.
+1 again.
Kevin Kofler
12:07 a.m.
New subject: Mozilla trademarks (Re: Thunderbird bz 579023 still not fixed even though there is an upstream fix available)
On Fri, 2010-04-23 at 11:34 +0200, Ralf Corsepius wrote:
Well, c.f. freedom 3 on http://www.gnu.org/philosophy/free-sw.html
You told us, you can't modify the sources and ship modified binaries
=> thunderbird and firefox are non-free, because of the trademarks
Mozilla apply.
You're right, the trademarks are detrimental to Fedora's stated goal of
freedom, which includes the freedom to modify. I have entered a tracker
bug requesting that trademark-unencumbered versions of all software at
least be offered as an alternative, and a specific bug for the case of
Firefox:
https://bugzilla.redhat.com/show_bug.cgi?id=582778
https://bugzilla.redhat.com/show_bug.cgi?id=582784
The tracker bug could benefit from a memorable alias. What are the
guidelines for choosing aliases? How about "F-TrademarkFree"?
--
Matt
2:27 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Martin Stransky wrote:
No, you get it wrong. It's about cooperation, we work with
upstream to
release one valid product. See the upstream bug, the fix may be included
in next security update.
That's too late. It should have been applied weeks ago! That crash has been
known for 7 weeks, a quick&dirty fix which could have been applied at least
temporarily has been proposed only days later.
Kevin Kofler
2:47 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Ralf Corsepius wrote:
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.
/me ducks and hides for cover.
No need to hide. I couldn't agree more, and in fact I've been arguing for
this all this time.
I really don't see why we're putting up with this kind of stupidity when we
could just rebrand the packages like e.g. Debian is doing.
Firefox, Thunderbird and xulrunner are also exempt from provenpager commits
because of those trademark reasons. They're the ONLY packages for which an
exemption has been granted. I really don't see why those packages deserve
this kind of special treatment. They should be treated the same as all the
other software in Fedora! If upstream insists on enforcing their trademarks
in a fascist way, then the only option is to do what Debian did and rename
the software.
Those packages are also sometimes not compliant with Fedora policies such as
usage of system libraries because any patches to use a system library need
trademark approval. This is also just unacceptable. See e.g. the Hunspell
fiasco:
* In February 2008, xulrunner was set to build against the system hunspell.
* In May 2008, it was discovered that xulrunner is using an old private
hunspell header even when using the system hunspell, which causes crashes
due to an ABI mismatch. But since xulrunner was untouchable, hunspell was
hacked instead to make it work.
* In July 2008, it was noticed that the above hack made hunspell
incompatible with upstream and all other distros, so it was reverted. The
fix for xulrunner would have been trivial, but as xulrunner was still
untouchable, it was reverted to use the private copy instead, a blatant
violation of the Fedora Packaging Guidelines.
* Only in June 2009, xulrunner finally got fixed to use the system hunspell
again.
Another big issue is libpng: xulrunner is bundling a forked libpng for APNG
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. They removed MNG support for very
unconvincing reasons and then decided to reinvent the wheel to help fragment
the web.) 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. I'm really fed up of
Firefox and Thunderbird continually getting special treatment.
Kevin Kofler
3 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
I wrote:
Those packages are also sometimes not compliant with Fedora policies
such
as usage of system libraries because any patches to use a system library
need trademark approval.
Another one: Thunderbird STILL bundles its own Gecko instead of using the
system xulrunner, another blatant violation of our packaging guidelines.
Nothing is done to fix this issue because upstream does not care and we
can't change what they ship.
Kevin Kofler
2:29 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/25/2010 10:00 AM, Kevin Kofler wrote:
I wrote:
> Those packages are also sometimes not compliant with Fedora policies such
> as usage of system libraries because any patches to use a system library
> need trademark approval.
Another one: Thunderbird STILL bundles its own Gecko instead of using the
system xulrunner, another blatant violation of our packaging guidelines.
Nothing is done to fix this issue because upstream does not care and we
can't change what they ship.
Hi,
Thunderbird is unable to use system xulrunner yet. We are working with
upstream to fix it. Believe me, it's not a trivial task:
https://bugzilla.mozilla.org/show_bug.cgi?id=377319 (see also depends bugs).
--
jh
3:04 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
I wrote:
Those packages are also sometimes not compliant with Fedora policies
such
as usage of system libraries because any patches to use a system library
need trademark approval. This is also just unacceptable.
PPS: And another one: xulrunner uses a bundled libffi. Another blatant
violation of our packaging guidelines. This is fixed in Debian, but we
aren't applying their patch.
Kevin Kofler
10:08 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, Apr 25, 2010 at 09:47:26 +0200,
Kevin Kofler <kevin.kofler(a)chello.at> wrote:
Those packages are also sometimes not compliant with Fedora policies such as
usage of system libraries because any patches to use a system library need
trademark approval. This is also just unacceptable. See e.g. the Hunspell
fiasco:
Isn't this a FESCO issue? Maybe it is time to reopen this issue?
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.
10:35 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, Apr 25, 2010 at 5:08 PM, Bruno Wolff III <bruno(a)wolff.to> wrote:
On Sun, Apr 25, 2010 at 09:47:26 +0200,
Kevin Kofler <kevin.kofler(a)chello.at> wrote:
>
> Those packages are also sometimes not compliant with Fedora policies such as
> usage of system libraries because any patches to use a system library need
> trademark approval. This is also just unacceptable. See e.g. the Hunspell
> fiasco:
Isn't this a FESCO issue? Maybe it is time to reopen this issue?
I don't see how using Mozilla trademarks provides significant benefit
to Fedora.
By shipping software using names known to users coming from other OSes?
Btw. the fedora trademark guidelines aren't less restrictive either.
10:51 a.m.
On Sun, Apr 25, 2010 at 17:35:13 +0200,
drago01 <drago01(a)gmail.com> wrote:
By shipping software using names known to users coming from other OSes?
While in general it would be confusing if everything was renamed, I think
the default web browser name is less of an issue since it is installed
by default. People don't need to know the name to get it installed or
run it. Besides iceweasel is also familiar to people.
Btw. the fedora trademark guidelines aren't less restrictive
either.
The Fedora trademark doesn't cause problems within Fedora. We also even
have things in place to make rebranding easier for downstream.
The issue is that the Mozilla trademark rules are preventing us from
packaging software using those trademarks in accordance with our rules.
I think it would be better for the trademarks to go, rather than granting
exceptions to the rules.
We could even try coordinating names with Debian to reduce confusion.
11:03 a.m.
Bruno Wolff III <bruno(a)wolff.to> writes:
The issue is that the Mozilla trademark rules are preventing us from
packaging software using those trademarks in accordance with our rules.
I think it would be better for the trademarks to go, rather than granting
exceptions to the rules.
Wouldn't it be sensible to approach the Mozilla folk about getting them
to relax their requirements so that sane packaging is possible? ISTM
that this must be a problem for other distros too.
regards, tom lane
11:08 a.m.
On Sun, Apr 25, 2010 at 12:03:28 -0400,
Tom Lane <tgl(a)redhat.com> wrote:
Bruno Wolff III <bruno(a)wolff.to> writes:
> The issue is that the Mozilla trademark rules are preventing us from
> packaging software using those trademarks in accordance with our rules.
> I think it would be better for the trademarks to go, rather than granting
> exceptions to the rules.
Wouldn't it be sensible to approach the Mozilla folk about getting them
to relax their requirements so that sane packaging is possible? ISTM
that this must be a problem for other distros too.
I though we did that several years ago. I wasn't a packager at that time,
so I didn't follow what was happening closely.
11:19 a.m.
Bruno Wolff III <bruno(a)wolff.to> writes:
On Sun, Apr 25, 2010 at 12:03:28 -0400,
Tom Lane <tgl(a)redhat.com> wrote:
> Wouldn't it be sensible to approach the Mozilla folk about getting them
> to relax their requirements so that sane packaging is possible? ISTM
> that this must be a problem for other distros too.
I though we did that several years ago. I wasn't a packager at
that time,
so I didn't follow what was happening closely.
Well, if we say to them "either you fix this licensing problem or your
trademarks will disappear from Fedora", it might get their attention.
Especially if we can get Debian and some other distros to tell them the
same.
regards, tom lane
11:31 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, Apr 25, 2010 at 6:19 PM, Tom Lane <tgl(a)redhat.com> wrote:
Bruno Wolff III <bruno(a)wolff.to> writes:
> On Sun, Apr 25, 2010 at 12:03:28 -0400,
> Tom Lane <tgl(a)redhat.com> wrote:
>> Wouldn't it be sensible to approach the Mozilla folk about getting them
>> to relax their requirements so that sane packaging is possible? ISTM
>> that this must be a problem for other distros too.
> I though we did that several years ago. I wasn't a packager at that time,
> so I didn't follow what was happening closely.
Well, if we say to them "either you fix this licensing problem or your
trademarks will disappear from Fedora", it might get their attention.
Especially if we can get Debian and some other distros to tell them the
same.
Well, since Debian already has Iceweasel, Mozilla obviously don't care.
IMO, we should go the Debian way.
--
LG Thomas
Dubium sapientiae initium
11:33 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, Apr 25, 2010 at 6:31 PM, Thomas Janssen
<thomasj(a)fedoraproject.org> wrote:
On Sun, Apr 25, 2010 at 6:19 PM, Tom Lane <tgl(a)redhat.com>
wrote:
> Bruno Wolff III <bruno(a)wolff.to> writes:
>> On Sun, Apr 25, 2010 at 12:03:28 -0400,
>> Tom Lane <tgl(a)redhat.com> wrote:
>>> Wouldn't it be sensible to approach the Mozilla folk about getting them
>>> to relax their requirements so that sane packaging is possible? ISTM
>>> that this must be a problem for other distros too.
>
>> I though we did that several years ago. I wasn't a packager at that time,
>> so I didn't follow what was happening closely.
>
> Well, if we say to them "either you fix this licensing problem or your
> trademarks will disappear from Fedora", it might get their attention.
> Especially if we can get Debian and some other distros to tell them the
> same.
Well, since Debian already has Iceweasel, Mozilla obviously don't care.
IMO, we should go the Debian way.
Whoops, sorry for the PM Bruno and Kevin, i did just click on reply to
all. Forgot to check for a cc.
--
LG Thomas
Dubium sapientiae initium
11:36 a.m.
On Sun, Apr 25, 2010 at 18:33:27 +0200,
Thomas Janssen <thomasj(a)fedoraproject.org> wrote:
Whoops, sorry for the PM Bruno and Kevin, i did just click on reply to
all. Forgot to check for a cc.
If I didn't want PM copies, I'd set mail-followup-to to not get them.
I sometimes find it useful to get the extra copy.
Also the default for the Fedora lists seems to be to try to detect duplicates
and not send a copy through the list to them. (I have turned that off, so
I get both copies.)
And to keep things a bit on topic I added a question about Mozilla trademarks
to the election questions. I will be interested in seeing what FESCO and
Board candidates have to say on this issue.
4:55 p.m.
Tom Lane wrote:
Wouldn't it be sensible to approach the Mozilla folk about
getting them
to relax their requirements so that sane packaging is possible? ISTM
that this must be a problem for other distros too.
We have tried, Debian has tried, other distros have tried, Mozilla just said
"no". The problem is that, unlike most other upstreams, Mozilla does not see
even a coalition of all GNU/Linux distros as an almighty power you have to
give its way, but only as a secondary platform next to Window$ from whose
users they get millions of downloads. So they feel they're in the position
of power and get to dictate to us what we have to do.
They also care very little about the needs of distros and it took years for
some of the system libs to get used rather than bundled, for things like
system icons getting adopted etc. They still suck in the system integration
domain in many ways, e.g. openSUSE's KDE integration patches have yet to be
merged, and of course our maintainers refuse to merge openSUSE's patches due
to the usual trademark concerns (which openSUSE doesn't seem to be concerned
about, they just ship those patches in branded packages, so either Mozilla
approved them, which means we can ship them too, or they just didn't care,
so why should we?).
Kevin Kofler
5:02 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, Apr 25, 2010 at 11:55 PM, Kevin Kofler <kevin.kofler(a)chello.at> wrote:
They also care very little about the needs of distros and it took
years for
some of the system libs to get used rather than bundled, for things like
system icons getting adopted etc. They still suck in the system integration
domain in many ways, e.g. openSUSE's KDE integration patches have yet to be
merged, and of course our maintainers refuse to merge openSUSE's patches due
to the usual trademark concerns (which openSUSE doesn't seem to be concerned
about, they just ship those patches in branded packages, so either Mozilla
approved them, which means we can ship them too, or they just didn't care,
so why should we?).
What really strikes me here is that we're not even talking about
adding random downstream patches, but an upstream one.
--
Gianluca Sforna
http://morefedora.blogspot.com
http://www.linkedin.com/in/gianlucasforna
8:55 p.m.
On Sun 25 April 2010 2:55:58 pm Kevin Kofler wrote:
They still suck in the system integration
domain in many ways, e.g. openSUSE's KDE integration patches have yet to
be merged, and of course our maintainers refuse to merge openSUSE's
patches due to the usual trademark concerns
In their defense, those patches are fairly messy.
--
Ryan Rix
== http://hackersramblings.wordpress.com | http://rix.si/ ==
11:56 p.m.
I wrote:
Yes, definitely. We should ask Debian about using the ice* names
they're
using, and also share patches with them.
An alternative would be using GNU IceCat:
http://www.gnu.org/software/gnuzilla/
but they don't have a rebranded Thunderbird.
Kevin Kofler
12:37 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
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.
I think you are grossly misjudging the relative visibility and
importance of the Firefox and Fedora brands... nobody knows what Fedora
is, while most computer users will have at least heard about Firefox.
12:45 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, Apr 25, 2010 at 13:37:11 -0400,
Matthias Clasen <mclasen(a)redhat.com> wrote:
On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
> 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.
I think you are grossly misjudging the relative visibility and
importance of the Firefox and Fedora brands... nobody knows what Fedora
is, while most computer users will have at least heard about Firefox.
Yeah, but "most computer users" isn't relevant. The question is about what
is relevant to Fedora users. Changing the name of Firefox will have little
affect on them since it is installed as the default web browser. Being able
to fix bugs in a timely manner on the other hand, is going to have a
significant affect on them.
1:48 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, 2010-04-25 at 12:45 -0500, Bruno Wolff III wrote:
On Sun, Apr 25, 2010 at 13:37:11 -0400,
Matthias Clasen <mclasen(a)redhat.com> wrote:
> On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
>
> > 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.
>
>
> I think you are grossly misjudging the relative visibility and
> importance of the Firefox and Fedora brands... nobody knows what Fedora
> is, while most computer users will have at least heard about Firefox.
Yeah, but "most computer users" isn't relevant. The question is about what
is relevant to Fedora users. Changing the name of Firefox will have little
affect on them since it is installed as the default web browser. Being able
to fix bugs in a timely manner on the other hand, is going to have a
significant affect on them.
Wait, let's not get silly here.
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.
Consider:
* Mozilla is currently implementing unit tests *on Fedora* in addition
to their long-standing tests on CentOS. This benefits both communities.
See Armen's blog posts at
http://armenzg.blogspot.com/2010/04/unit-tests-for-fedora-utont-project.html and
http://armenzg.blogspot.com/2010/04/one-more-fedora-unit-test-suite-visib...
* 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.
* The trademark rules are there for a reason. Browser and e-mail clients
are some of the most common attack points on desktop systems, and
Mozilla needs to ensure that they don't get a black eye for some
vulnerability introduced by a distro. And distros definitely introduce
vulnerabilities: think about the Debian ssh-keygen patch fiasco as an
example. We wouldn't do something so rash, of course -- or would we? The
suggestion earlier in this thread that we patch TB and push directly to
stable does not instill confidence. (We have the freedom to turn off the
branding anytime and use the code however we want, but why give up the
marketing value? and why give up the testing?)
Let's not be brash. If we want to ship TB with one small patch, it's a
simple matter of asking.
-Chris
2:03 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 12:18 AM, Chris Tyler wrote:
Let's not be brash. If we want to ship TB with one small patch, it's a
simple matter of asking.
If it was so simple, why haven't we done it already? What about patches
to use system libraries?
Rahul
3:11 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, 2010-04-26 at 00:33 +0530, Rahul Sundaram wrote:
On 04/26/2010 12:18 AM, Chris Tyler wrote:
>
> Let's not be brash. If we want to ship TB with one small patch, it's a
> simple matter of asking.
If it was so simple, why haven't we done it already?
We did, with Firefox and Pango.
What about patches to use system libraries?
I'm sure they'd love to receive 'em!
Part of the problem here is that Mozilla's reference image for Linux is
CentOS-based -- which means old library versions. Now that they're
adding Fedora to their test farm, they're going to be getting better
exposure to current versions. Sending patches to use current libraries
upstream would be great.
I'm sure Armen would love some help with nailing the remaining Fedora
oranges -- see the tracking bug at
https://bugzilla.mozilla.org/show_bug.cgi?id=554934
Also, back to the original TB bug -- the reason this wasn't a top
priority upstream was that it wasn't showing up in the Moz crashstats
(they only had one report). We need to finish the work of tying our
crash reports into their system, something they've been asking for since
well before ABRT.
-Chris
3:20 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 01:41 AM, Chris Tyler wrote:
On Mon, 2010-04-26 at 00:33 +0530, Rahul Sundaram wrote:
> On 04/26/2010 12:18 AM, Chris Tyler wrote:
>
>> Let's not be brash. If we want to ship TB with one small patch, it's a
>> simple matter of asking.
>>
> If it was so simple, why haven't we done it already?
>
We did, with Firefox and Pango.
Well, I was referring to Thunderbird and the fix for the crash that is
part of this thread obviously.
> What about patches to use system libraries?
>
I'm sure they'd love to receive 'em!
Aren't they aware of the existing patches already?
Also, back to the original TB bug -- the reason this wasn't a
top
priority upstream was that it wasn't showing up in the Moz crashstats
(they only had one report). We need to finish the work of tying our
crash reports into their system, something they've been asking for since
well before ABRT.
Till then, how do we get them to prioritize downstream crashes?
Alternatively are they willing to let Fedora patch it without requiring
a rebranding? They could be be more flexible.
Rahul
5:39 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Chris Tyler wrote:
On Mon, 2010-04-26 at 00:33 +0530, Rahul Sundaram wrote:
> What about patches to use system libraries?
I'm sure they'd love to receive 'em!
http://patch-tracker.debian.org/patch/series/view/xulrunner/1.9.2.3-2/deb...
http://patch-tracker.debian.org/patch/series/view/xulrunner/1.9.2.3-2/deb...
Kevin Kofler
5:21 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Chris Tyler wrote:
Wait, let's not get silly here.
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.
I'm fed up of the "they're great, so let's do all they want"
rhetoric. No
matter how nice the other guys are, if what they do conflicts with our
ideals and they aren't flexible enough to change this, we CANNOT work with
them.
I strongly disagree about them "Getting It". 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?
Even some copyright holders who previously enforced this kind of
restrictions in their copyright licenses, e.g. DJB and the UW (PINE), have
seen that patch approval requirements are a major roadblock and made their
software Free Software. So why does Mozilla have to use trademarks to
restrict the freeness of their software?
I also think Mozilla doesn't "Get It" in several other ways. System
integration on GNU/Linux is generally an afterthought and moves very slowly,
e.g. it has taken years to get Firefox to use system icons, and I have no
idea when or even if openSUSE's KDE integration will be merged. Firefox is
set up to send all the URLs you consult to Google by default, which also
means you have to manually disable the relevant settings (which don't even
say anywhere in the dialog that they send all your browsing data to Google!)
if you don't agree with the ToS for Google's anti-phishing service. They
forked libpng to add support for the nonstandard "APNG" extension they added
despite the explicit opposition of the PNG group, bundle that forked version
and don't support building against a system libpng without their patch; they
removed support for the official MNG format which serves the same purpose
for bogus reasons, probably ensuring that no animated PNG format will ever
reach critical mass (which is quite sad considering the obsolescence of GIF,
and which might even contribute to spreading proprietary crap such as
Flash).
Consider:
* Mozilla is currently implementing unit tests *on Fedora* in addition
to their long-standing tests on CentOS. This benefits both communities.
So what? That has nothing to do with the issues at hand.
* 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.
Whether it's called "Firefox" or "Iceweasel" or "Water
Buffalo" or whatever
doesn't change how it works at all.
* The trademark rules are there for a reason. Browser and e-mail
clients
are some of the most common attack points on desktop systems, and
Mozilla needs to ensure that they don't get a black eye for some
vulnerability introduced by a distro.
LOL, now that's a funny excuse! They should start fixing their own
vulnerabilities, they have at least an order of magnitude more security
issues than Konqueror. For each one they fix, a new one pops up somewhere
else.
And in addition, this is not a problem for the kernel Linux, for GNOME
(including Epiphany and Evolution), for KDE (including Konqueror and KMail)
and many other software, so why would this be a problem for Mozilla?
And distros definitely introduce vulnerabilities: think about the
Debian
ssh-keygen patch fiasco as an example.
That was an isolated case. Extrapolating from that that distros introduce
vulnerabilities in all the software they patch is quite silly.
We wouldn't do something so rash, of course -- or would we? The
suggestion
earlier in this thread that we patch TB and push directly to stable does
not instill confidence.
Well, it depends on how complicated the patch is. If it's rather trivial and
if the problem is important (which it is), I think pushing it out as fast as
possible is the right thing to do. If the patch is more complex, it needs
some testing.
Let's not be brash. If we want to ship TB with one small patch,
it's a
simple matter of asking.
Then let's ask, NOW, so we can get that crasher finally fixed!
But that still doesn't fix the underlying issue in the long run.
Kevin Kofler
5:37 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/25/2010 06:21 PM, Kevin Kofler wrote:
Can someone explain why the fedora version has a bug which upstream
version does not ? Or am I missing something ?
6:17 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Mail Lists wrote:
Can someone explain why the fedora version has a bug which upstream
version does not ? Or am I missing something ?
The upstream version has that bug too, they just don't care about it enough
to release a fixed version in a timely manner.
Kevin Kofler
6:56 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/25/2010 07:17 PM, Kevin Kofler wrote:
The upstream version has that bug too, they just don't care about
it enough
to release a fixed version in a timely manner.
OH - FYI, I am running upstream and I don't have that problem ... can
disconnect the network all i want .. no crash.
3:18 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, Apr 26, 2010 at 1:56 AM, Mail Lists <lists(a)sapience.com> wrote:
On 04/25/2010 07:17 PM, Kevin Kofler wrote:
> The upstream version has that bug too, they just don't care about it enough
> to release a fixed version in a timely manner.
OH - FYI, I am running upstream and I don't have that problem ... can
disconnect the network all i want .. no crash.
The fedora build works fine for me too.
4:55 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, 2010-04-25 at 14:48 -0400, Chris Tyler wrote:
* The trademark rules are there for a reason. Browser and e-mail
clients
are some of the most common attack points on desktop systems, and
Mozilla needs to ensure that they don't get a black eye for some
vulnerability introduced by a distro. And distros definitely introduce
vulnerabilities: think about the Debian ssh-keygen patch fiasco as an
example. We wouldn't do something so rash, of course -- or would we? The
suggestion earlier in this thread that we patch TB and push directly to
stable does not instill confidence. (We have the freedom to turn off the
branding anytime and use the code however we want, but why give up the
marketing value? and why give up the testing?)
I think a rather large part of the problem here is that all the above
'special exception' pleading applies far more to Firefox than it does to
Thunderbird. Firefox is a special exception; it's a phenomenon, the
single most successful F/OSS app, an app with its own very definite
distinct identity which many people know. It clearly does have a
reputation to protect and I can entirely agree with it being very
careful about that.
Thunderbird...uhhh, not so much. It's nowhere near as popular as
Firefox. Most people don't know what it is. It's not really a 'special
exception'; it's just another application like the ten gazillion others
we ship which don't have onerous trademark restrictions attached.
I think it'd be appropriate for Mozilla to take a rather more liberal
line with Thunderbird than it does with Firefox, if that's legally
plausible.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
4:35 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Adam Williamson wrote:
I think a rather large part of the problem here is that all the
above
'special exception' pleading applies far more to Firefox than it does to
Thunderbird. Firefox is a special exception; it's a phenomenon, the
single most successful F/OSS app, an app with its own very definite
distinct identity which many people know. It clearly does have a
reputation to protect and I can entirely agree with it being very
careful about that.
Thunderbird...uhhh, not so much. It's nowhere near as popular as
Firefox. Most people don't know what it is. It's not really a 'special
exception'; it's just another application like the ten gazillion others
we ship which don't have onerous trademark restrictions attached.
I think it'd be appropriate for Mozilla to take a rather more liberal
line with Thunderbird than it does with Firefox, if that's legally
plausible.
Agreed, but I think even Firefox doesn't deserve this kind of preferential
treatment.
In fact, I don't see Firefox as being the "absolute requirement" it's
painted to be at all, we could even consider just not shipping it at all and
picking a different default browser for the GNOME spin, e.g. Epiphany which
is the official GNOME browser. (The KDE spin already does not ship Firefox,
we default to Konqueror.) It's just a web browser.
Kevin Kofler
4:55 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Tue, 2010-04-27 at 23:35 +0200, Kevin Kofler wrote:
In fact, I don't see Firefox as being the "absolute
requirement" it's
painted to be at all, we could even consider just not shipping it at all and
picking a different default browser for the GNOME spin, e.g. Epiphany which
is the official GNOME browser.
Epiphany is a non-starter. In the default configuration, it doesn't
validate SSL certificates at all (bug 569577). An unbranded Mozilla
browser would be a much better choice.
--
Matt
4:59 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Tue, Apr 27, 2010 at 17:55:39 -0400,
Matt McCutchen <matt(a)mattmccutchen.net> wrote:
Epiphany is a non-starter. In the default configuration, it doesn't
validate SSL certificates at all (bug 569577). An unbranded Mozilla
browser would be a much better choice.
The way Firefox does it, is more to help companies sell certificates than to
actually help security.
5:15 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Tue, 2010-04-27 at 16:59 -0500, Bruno Wolff III wrote:
On Tue, Apr 27, 2010 at 17:55:39 -0400,
Matt McCutchen <matt(a)mattmccutchen.net> wrote:
>
> Epiphany is a non-starter. In the default configuration, it doesn't
> validate SSL certificates at all (bug 569577). An unbranded Mozilla
> browser would be a much better choice.
The way Firefox does it, is more to help companies sell certificates than to
actually help security.
I don't want to repeat that debate here. Please see:
http://www.gerv.net/security/self-signed-certs/
--
Matt
5:37 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Bruno Wolff III wrote:
The way Firefox does it, is more to help companies sell certificates
than
to actually help security.
+1
All it does is it leads people to use completely unencrypted HTTP instead,
to avoid the "big scary warnings". How does that provide any added security?
I like the way Konqueror handles this: it does complain about self-signed or
otherwise invalid certs, but it allows you to accept them either temporarily
(for the duration of the session) or permanently in 2 clicks (one to accept
and one to choose whether to accept it for the session or forever).
Kevin Kofler
6:41 a.m.
On Tue, Apr 27, 2010 at 04:59:55PM -0500, Bruno Wolff III wrote:
On Tue, Apr 27, 2010 at 17:55:39 -0400,
Matt McCutchen <matt(a)mattmccutchen.net> wrote:
>
> Epiphany is a non-starter. In the default configuration, it doesn't
> validate SSL certificates at all (bug 569577). An unbranded Mozilla
> browser would be a much better choice.
The way Firefox does it, is more to help companies sell certificates than to
actually help security.
agreed.
I did recently look into the list of CAs trusted by Firefox, it looks bad. There
are CAs from countries all over the world.
I would say that 99% of users do not need a CA from some mid-eastern or far-eastern
countries. But each and every of these can give a forged certificate for anything that
will be gladly accepted by Firefox.
To me the security model of Firefox appears too permissive. I have seen online banks
which do include page elements, even javascript from 3 parties severs, different domains
and certificates. Yet there is one URL shown and the user is lead to believe everthing
is certified by the same authority.
Richard
5:35 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Matt McCutchen wrote:
Epiphany is a non-starter. In the default configuration, it
doesn't
validate SSL certificates at all (bug 569577). An unbranded Mozilla
browser would be a much better choice.
That's a wrong bug ID. RH/Fedora bug 569577 is a Nautilus crash. GNOME bug
569577 is a Banshee bug.
Kevin Kofler
5:37 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Wed, 2010-04-28 at 00:35 +0200, Kevin Kofler wrote:
Matt McCutchen wrote:
> Epiphany is a non-starter. In the default configuration, it doesn't
> validate SSL certificates at all (bug 569577). An unbranded Mozilla
> browser would be a much better choice.
That's a wrong bug ID. RH/Fedora bug 569577 is a Nautilus crash. GNOME bug
569577 is a Banshee bug.
Whoops. It's Fedora bug 569677.
--
Matt
3:05 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Sun, 2010-04-25 at 12:45 -0500, Bruno Wolff III wrote:
On Sun, Apr 25, 2010 at 13:37:11 -0400,
Matthias Clasen <mclasen(a)redhat.com> wrote:
> On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
>
> > 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.
>
>
> I think you are grossly misjudging the relative visibility and
> importance of the Firefox and Fedora brands... nobody knows what Fedora
> is, while most computer users will have at least heard about Firefox.
Yeah, but "most computer users" isn't relevant. The question is about what
is relevant to Fedora users. Changing the name of Firefox will have little
affect on them since it is installed as the default web browser. Being able
to fix bugs in a timely manner on the other hand, is going to have a
significant affect on them.
Not a nice idea, but, at least as a temporary
workaround, could Fedora
ship both a Firefox and an Iceweasel; Firefox complying with the
trademark rules, and Iceweasel working as users would want it.
Could the Fedora shipped Firefox even have a home page that says "Have
you tried Iceweasel ..."? And bugs reported against Firefox could be
closed with "Fixed in Iceweasel".
3:16 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, Apr 26, 2010 at 10:05 AM, Quentin Armitage
<Quentin(a)armitage.org.uk> wrote:
On Sun, 2010-04-25 at 12:45 -0500, Bruno Wolff III wrote:
> On Sun, Apr 25, 2010 at 13:37:11 -0400,
> Matthias Clasen <mclasen(a)redhat.com> wrote:
> > On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
> >
> > > 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.
> >
> >
> > I think you are grossly misjudging the relative visibility and
> > importance of the Firefox and Fedora brands... nobody knows what Fedora
> > is, while most computer users will have at least heard about Firefox.
>
> Yeah, but "most computer users" isn't relevant. The question is about
what
> is relevant to Fedora users. Changing the name of Firefox will have little
> affect on them since it is installed as the default web browser. Being able
> to fix bugs in a timely manner on the other hand, is going to have a
> significant affect on them.
Not a nice idea, but, at least as a temporary workaround, could Fedora
ship both a Firefox and an Iceweasel; Firefox complying with the
trademark rules, and Iceweasel working as users would want it.
Could the Fedora shipped Firefox even have a home page that says "Have
you tried Iceweasel ..."? And bugs reported against Firefox could be
closed with "Fixed in Iceweasel".
That is nonsense ... it just creates confusion and maintenance overhead.
3:25 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/25/2010 01:37 PM, Matthias Clasen wrote:
I think you are grossly misjudging the relative visibility and
importance of the Firefox and Fedora brands... nobody knows what Fedora
is, while most computer users will have at least heard about Firefox.
Agreed a fortiori - in fact everyone I know chooses the browser and
mail client by name and definitely not by function.
If it was fedora branded then I'd guess a goodly chunk would just go
and install the upstream anyway coz they would not know nor trust the
browser called 'Fedora-Browser' or whatever.
Those wishing to use firefox will use firefox - those wishing to use
google-chrome will use it (in spite of chromium sounding similar by the
by, it was far less functional).
To be honest - i had never heard of iceweasel (yeh I get the name now)
until this thread .. but I sure have heard of firefox/thunderbird.
gene/
4:58 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Mail Lists wrote:
If it was fedora branded then I'd guess a goodly chunk would
just go
and install the upstream anyway coz they would not know nor trust the
browser called 'Fedora-Browser' or whatever.
Those wishing to use firefox will use firefox - those wishing to use
google-chrome will use it (in spite of chromium sounding similar by the
by, it was far less functional).
To be honest - i had never heard of iceweasel (yeh I get the name now)
until this thread .. but I sure have heard of firefox/thunderbird.
If we use the same name as Debian and get the press to publish articles with
headlines like "Iceweasel is the Linux Firefox" (we don't have to worry
about trademark issues with such a statement if THEY do it for us :-p), that
might help.
Kevin Kofler
3:36 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/25/2010 07:37 PM, Matthias Clasen wrote:
On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
> 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.
I think you are grossly misjudging the relative visibility and
importance of the Firefox and Fedora brands... nobody knows what Fedora
is, while most computer users will have at least heard about Firefox.
Does this justify Fedora throwing the principles of "Free SW" over board?
IMO, *no* - it's time to spread the world about Mozilla's trademark
policy violating the prinicples of Free SW and Fedora's Mozilla being
hostage of it.
In other words, if Fedora hasn't already entirely abandoned the original
motivations and foundations it once was based on, it's time for Fedora
to very serously reconsider their attitude on Mozilla packages.
Ralf
5:09 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, 2010-04-26 at 10:36 +0200, Ralf Corsepius wrote:
IMO, *no* - it's time to spread the world about Mozilla's
trademark
policy violating the prinicples of Free SW and Fedora's Mozilla being
hostage of it.
You mean, much like the Fedora and Red Hat trademark policies, which say
almost the exact same things?
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.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
1:49 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 12:09 AM, Adam Williamson wrote:
On Mon, 2010-04-26 at 10:36 +0200, Ralf Corsepius wrote:
> IMO, *no* - it's time to spread the world about Mozilla's trademark
> policy violating the prinicples of Free SW and Fedora's Mozilla being
> hostage of it.
You mean, much like the Fedora and Red Hat trademark policies, which say
almost the exact same things?
Correct. The Fedora distro itself is non-free because
of similar
trademark games.
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.
Correct.
Similar as Fedora/RH's restrictive trademark policies prevents people
from "enhancing/bugfixing" Fedora, Mozilla's trademarks policies
prevents Fedora from "enhancing/bugfixing" Firefox/Thunderbird.
The resort to both Fedora/RH's and Mozilla's trademark policies is to
"rebrand or abandon".
Wrt. to Mozilla package in Fedora, the problem is Fedora containing
packages which clearly violate the prinicples Fedora once was founded on
("Freedom") and which infect Fedora with effects of "Non-Free SW"
(e.g.
unfixable bugs) == Mozilla's trademark games voids the effects of "OSS".
Ralf
4:38 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Adam Williamson wrote:
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.
That has not been an issue for Debian.
It perfectly makes sense that you'd want to protect your own name, but still
not use somebody else's similarly protected name because it prevents you
from shipping what you want to ship.
Kevin Kofler
6:06 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Tue, 2010-04-27 at 23:38 +0200, Kevin Kofler wrote:
Adam Williamson wrote:
> 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.
That has not been an issue for Debian.
It perfectly makes sense that you'd want to protect your own name, but still
not use somebody else's similarly protected name because it prevents you
from shipping what you want to ship.
I wasn't suggesting it'd prevent us from shipping an unbranded Firefox /
Thunderbird if we wanted to. I was just saying that it would stop us
from pitching some kind of public hissy fit about how unreasonable
Mozilla was being, as Ralf seemed to be advocating.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
3:36 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 4/25/2010 8:37 PM, Matthias Clasen wrote:
On Sun, 2010-04-25 at 10:08 -0500, Bruno Wolff III wrote:
> 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.
>
I think you are grossly misjudging the relative visibility and
importance of the Firefox and Fedora brands... nobody knows what Fedora
is, while most computer users will have at least heard about Firefox.
I disagree with your criteria. The visibility in of itself is not
relevant; its consequences are. The important criteria, in my eyes, are:
1) Does the Mozilla name attract users to Fedora or keep them with
Fedora once they have tasted it?
2) Would Fedora lose users if it dropped the trademarked name?
3) Does Fedora lose users by having unpatched versions of Firefox etc.?
4:48 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Bruno Wolff III wrote:
Isn't this a FESCO issue? Maybe it is time to reopen this issue?
Knowing my fellow FESCo members, I don't think I'll get a majority to agree
with me. :-(
Kevin Kofler
3:41 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/25/2010 11:48 PM, Kevin Kofler wrote:
Bruno Wolff III wrote:
> Isn't this a FESCO issue? Maybe it is time to reopen this issue?
Knowing my fellow FESCo members, I don't think I'll get a majority to agree
with me. :-(
Well, may-be FESCO should decide upon on whether the FSF's
"freedom 3" [1] is a inclusion/exclusion criterion for packages in Fedora.
Ralf
[1] http://www.gnu.org/philosophy/free-sw.html
3:52 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 02:11 PM, Ralf Corsepius wrote:
Well, may-be FESCO should decide upon on whether the FSF's
"freedom 3" [1] is a inclusion/exclusion criterion for packages in Fedora.
Ralf
[1] http://www.gnu.org/philosophy/free-sw.html
It is (except for firmware) but before you wave it around, you will do
well to ask FSF whether the Mozilla trademark guidelines violate it. My
understanding is that, they don't consider it a violation and yes, I
have actually asked them.
Rahul
4:09 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, Apr 26, 2010 at 9:52 AM, Rahul Sundaram <metherid(a)gmail.com> wrote:
On 04/26/2010 02:11 PM, Ralf Corsepius wrote:
> Well, may-be FESCO should decide upon on whether the FSF's
> "freedom 3" [1] is a inclusion/exclusion criterion for packages in Fedora.
>
> Ralf
>
> [1] http://www.gnu.org/philosophy/free-sw.html
>
It is (except for firmware) but before you wave it around, you will do
well to ask FSF whether the Mozilla trademark guidelines violate it. My
understanding is that, they don't consider it a violation and yes, I
have actually asked them.
Rahul
May I ask whether anyone has the same crash occurring with Thunderbird
3.1b2? I have been running the nightlies for that version for months
without any problem - is the crash only occurring with version 3.0.x?
Rather than planning on breaking away from Mozilla perhaps when TB
releases 3.1 the issue will become a non-issue?
Just a thought!
--
mike c
4:17 a.m.
On 04/26/2010 02:39 PM, mike cloaked wrote:
May I ask whether anyone has the same crash occurring with Thunderbird
3.1b2? I have been running the nightlies for that version for months
without any problem - is the crash only occurring with version 3.0.x?
Rather than planning on breaking away from Mozilla perhaps when TB
releases 3.1 the issue will become a non-issue?
Just a thought!
I see the crash issue as a symptom of a large issue of whether we are
comfortable with the trademark boundaries of Mozilla. Besides the beta
of 3.1 might have other regressions and not very suitable as an update
to Fedora 12. If Mozilla does a release quickly for 3.0.x, then it
would solve the current issue but I think it still leaves downstream in
a rather unpleasant situation in general where we are entirely dependant
on getting permission from Mozilla for even minor patches. Mozilla
certainly are well within their legal rights to dictate how their brand
should be used by downstream distributions but whether it is the right
balance is questionable.
Rahul
8:35 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 10:52 AM, Rahul Sundaram wrote:
On 04/26/2010 02:11 PM, Ralf Corsepius wrote:
> Well, may-be FESCO should decide upon on whether the FSF's
> "freedom 3" [1] is a inclusion/exclusion criterion for packages in Fedora.
>
> Ralf
>
> [1] http://www.gnu.org/philosophy/free-sw.html
>
It is (except for firmware)
... and on Fedora's own trademark encumbered packages.
but before you wave it around, you will do
well to ask FSF whether the Mozilla trademark guidelines violate it.
I know - I had
a discussion with a FSF-representative on this.
But ... many people disagree with this.
As well as the facts speak for themselves:
* The Fedora Mozilla packages can't be bug-fixed/patched.
Cause: The package is non-free.
* The Fedora Mozilla package can't be made compliant to the FPG.
Cause the packages are non-free.
Openly speaking, I feel RH is not interested addressing this issues,
likely for political reasons or for marketing reason - A too high price
to pay, if you ask me.
My
understanding is that, they don't consider it a violation and yes, I
have actually asked them.
Well, Mozilla license definitely is an open-source license.
It's just that the Mozilla packages in Fedora are non-free.
8:54 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 07:05 PM, Ralf Corsepius wrote:
On 04/26/2010 10:52 AM, Rahul Sundaram wrote:
> On 04/26/2010 02:11 PM, Ralf Corsepius wrote:
>> Well, may-be FESCO should decide upon on whether the FSF's
>> "freedom 3" [1] is a inclusion/exclusion criterion for packages in
>> Fedora.
>>
>> Ralf
>>
>> [1] http://www.gnu.org/philosophy/free-sw.html
>>
>
> It is (except for firmware)
... and on Fedora's own trademark encumbered packages.
No. Only firmware. Fedora's trademarks are not a problem at all.
http://www.gnu.org/philosophy/common-distros.html.
Well, Mozilla license definitely is an open-source license.
It's just that the Mozilla packages in Fedora are non-free.
How? Your claim here is certainly not supported by the FSF. although
they have some issues with the recommendation of non-free plugins and if
you consider it a legal question, it is outside the scope of FESCo.
FESCo can still discuss it from a policy perspective but then pointing
to FSF guidelines confuses a policy discussion with a legal one.
Rahul
8:56 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 09:35 AM, Ralf Corsepius wrote:
* The Fedora Mozilla packages can't be bug-fixed/patched.
Cause: The package is non-free.
* The Fedora Mozilla package can't be made compliant to the FPG.
Cause the packages are non-free.
Neither of these are true.
The Fedora Mozilla packages can be bug-fixed/patched. If Mozilla doesn't
accept the patches upstream first, we would no longer have permission to
use their trademarks, and would need to remove them when we did so.
As to why we have not simply patched at will, and discarded the
trademarks, well, I think that is ultimately up to FESCo and the
Maintainer(s) to decide how we wish to operate in that manner.
Personally, I feel that there is name-recognition value in the Mozilla
trademarks, and we should make every effort to try to discuss a
compromise with them to allow us a bit more flexibility while retaining
the trademark use. Perhaps this is a discussion that could be opened
with Luis Villa?
~spot
9:14 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
2010/4/26 Tom "spot" Callaway <tcallawa(a)redhat.com>:
The Fedora Mozilla packages can be bug-fixed/patched. If Mozilla
doesn't
accept the patches upstream first, we would no longer have permission to
use their trademarks, and would need to remove them when we did so.
You just said something like "yes we can, but Mozilla will deny" -
this is exactly what Ralf told us earlier.
As to why we have not simply patched at will, and discarded the
trademarks, well, I think that is ultimately up to FESCo and the
Maintainer(s) to decide how we wish to operate in that manner.
It's not up to maintainer to decide whether to provide non-free
package in Fedora. And I don't see why we need to ask FESCo for
resolution of this (clearly visible for almost everyone) violation of
our guidelines.
Personally, I feel that there is name-recognition value in the
Mozilla
trademarks, and we should make every effort to try to discuss a
compromise with them to allow us a bit more flexibility while retaining
the trademark use. Perhaps this is a discussion that could be opened
with Luis Villa?
Rebranding can be a difficult task, but this task also can be easily
measured in man-hours, man-days or man-months, and this would be a
ultimate solution, while chatting with lawers can consume much more
time w/o success (nothing personal here).
--
With best regards, Peter Lemenkov.
9:23 a.m.
On 04/26/2010 07:44 PM, Peter Lemenkov wrote:
It's not up to maintainer to decide whether to provide non-free
package in Fedora. And I don't see why we need to ask FESCo for
resolution of this (clearly visible for almost everyone) violation of
our guidelines.
Mozilla has some restrictive trademark guidelines but it nevertheless
Free software w.rt to copyright licensing (triple licensing with GPL,
LGPL and MPL) and is fully compliant with Fedora licensing guidelines.
It does violate Fedora packaging guidelines but FESCo is authorized to
give exceptions and decide on policy matters and I have filed it at
https://fedorahosted.org/fesco/ticket/369
Rahul
4:47 p.m.
Peter Lemenkov wrote:
Rebranding can be a difficult task, but this task also can be easily
measured in man-hours, man-days or man-months, and this would be a
ultimate solution, while chatting with lawers can consume much more
time w/o success (nothing personal here).
And the rebranding work has already been done by Debian. We should ask them
whether they're OK with us using the "iceweasel" name even if we apply
different patches than they do. If they're OK with that, we could just use
their patch as is, if not, we'd pick another name, but their patch would
still point us to the exact places to change.
Kevin Kofler
9:17 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 07:26 PM, Tom "spot" Callaway wrote:
As to why we have not simply patched at will, and discarded the
trademarks, well, I think that is ultimately up to FESCo and the
Maintainer(s) to decide how we wish to operate in that manner.
Alright. So I have filed this issue with FESCo since we are unlikely to
find consensus on our own
https://fedorahosted.org/fesco/ticket/369
Rahul
9:48 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 03:56 PM, Tom "spot" Callaway wrote:
On 04/26/2010 09:35 AM, Ralf Corsepius wrote:
> * The Fedora Mozilla packages can't be bug-fixed/patched.
> Cause: The package is non-free.
>
> * The Fedora Mozilla package can't be made compliant to the FPG.
> Cause the packages are non-free.
>
Neither of these are true.
The Fedora Mozilla packages can be bug-fixed/patched.
This was what Martin Stransky
had claimed.
A I understood his claim: "He can't apply patches/bug fixes to Fedora's
Thunderbird, because this would void Mozilla's trademark".
IMO, the techical impacts of this are clearly visible in Fedora's
packages, such as the clear FPG violations Kevin has elaborated.
Personally, I feel that there is name-recognition value in the
Mozilla
trademarks,
As I already said, I feel this what RH is interested in for political
or
marketing reasons.
As a propopent of "Free SW", my interest is to fight those who are
applying trademarks to undermine the principles of "free SW".
Ralf
5:30 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, 2010-04-26 at 16:48 +0200, Ralf Corsepius wrote:
As a propopent of "Free SW", my interest is to fight those
who are
applying trademarks to undermine the principles of "free SW".
This is not what Mozilla is doing. They are applying trademarks to
protect the names Mozilla, Firefox and Thunderbird. Consider what could
happen if they didn't; any Tom, Dick or Harry could 'patch' their
advert, badly-conceived 'feature' or, hell, piece of malware into
Firefox and call it Firefox, and Mozilla couldn't say boo to them for
it. Yes, this is true of many pieces of software that Fedora also ships,
but there's no real motive to do it to most of those; they don't have
Mozilla's profile.
If Mozilla's intent was in fact to undermine the principles of free SW,
they would make it as hard as possible to debrand Firefox; they do
rather the opposite.
(I've had the distinct displeasure of working with a bit of software
which actually did seem to go out of its way to make it as difficult as
possible to comply with the license gotchas - Truecrypt - and it's a
very different experience from working with Mozilla).
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
10:33 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 12:30 AM, Adam Williamson wrote:
On Mon, 2010-04-26 at 16:48 +0200, Ralf Corsepius wrote:
> As a propopent of "Free SW", my interest is to fight those who are
> applying trademarks to undermine the principles of "free SW".
This is not what Mozilla is doing. They are applying trademarks to
protect the names Mozilla, Firefox and Thunderbird. Consider what could
happen if they didn't; any Tom, Dick or Harry could 'patch' their
advert, badly-conceived 'feature' or, hell, piece of malware into
Firefox and call it Firefox, and Mozilla couldn't say boo to them for
it. Yes, this is true of many pieces of software that Fedora also ships,
Correct.
but there's no real motive to do it to most of those; they
don't have
Mozilla's profile.
Untrue, c.f. Gnome, KDE, QT, GCC, GNU and many more.
If Mozilla's intent was in fact to undermine the principles of
free SW,
they would make it as hard as possible to debrand Firefox; they do
rather the opposite.
I disagree. Their trademark policy is already doing it.
2:30 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 12:33 PM, Martin Stransky wrote:
Hi,
we're patching mozilla packages only for really critical issues because
of mozilla trademarks. We can't put any patch we want to the mozilla
package and ship it as 'Firefox' or 'Thunderbird'.
I've asked for inclusion at upstream bug,
https://bugzilla.mozilla.org/show_bug.cgi?id=550455, if you want to
speed up the process you can reply there too.
What is the exact definition of "really critical issues" here. A
frequent crash seems a critical issue to me.
Rahul
2:42 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 09:30 AM, Rahul Sundaram wrote:
On 04/23/2010 12:33 PM, Martin Stransky wrote:
> Hi,
>
> we're patching mozilla packages only for really critical issues because
> of mozilla trademarks. We can't put any patch we want to the mozilla
> package and ship it as 'Firefox' or 'Thunderbird'.
>
> I've asked for inclusion at upstream bug,
> https://bugzilla.mozilla.org/show_bug.cgi?id=550455, if you want to
> speed up the process you can reply there too.
>
What is the exact definition of "really critical issues" here. A
frequent crash seems a critical issue to me.
- 0day vulnerabilities
- critical crashes (like app fails to start for *everyone*, app crashes
for *everyone* in five minutes after start)
- fedora customization (build fixes)
ma.
2:50 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 01:12 PM, Martin Stransky wrote:
On 04/23/2010 09:30 AM, Rahul Sundaram wrote:
> What is the exact definition of "really critical issues" here. A
> frequent crash seems a critical issue to me.
- 0day vulnerabilities
- critical crashes (like app fails to start for *everyone*, app
crashes for *everyone* in five minutes after start)
- fedora customization (build fixes)
It is really annoying that Mozilla is restricting fixes for fairly
serious issues via their trademark guidelines. Their effort to protect
their brand has to balanced against downstream needs. It is
unacceptable for them to tie down our hands if they are not going to do
a release soon. It is very much possible that a bug might affect a
particular distribution only (even without patches) because of the rest
of the stack it is build against and Mozilla has to be responsive to all
that.
Rahul
4:11 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Friday 23 of April 2010 09:03:37 Martin Stransky wrote:
Hi,
we're patching mozilla packages only for really critical issues because
of mozilla trademarks. We can't put any patch we want to the mozilla
package and ship it as 'Firefox' or 'Thunderbird'.
just curious: is it possible to ship snapshots or trademarks does not allow
this too?
I've asked for inclusion at upstream bug,
https://bugzilla.mozilla.org/show_bug.cgi?id=550455, if you want to
speed up the process you can reply there too.
ma.
On 04/22/2010 08:39 PM, Felix Schwarz wrote:
> Hi,
>
> I'm concerned about bz 579023 [1] which is a Thunderbird crasher bug.
> This bug was fixed upstream [2] for about 3-4 weeks. I ran a thunderbird
> koji build version [3] with an adapted version of that patch since then
> without any problems. Other users confirmed that this patch fixes their
> problems as well.
>
> The Fedora bug has a number of duplicates with quite some number of cc'd
> users so I guess a lot of people experiencing these crashes.
>
> 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?
>
> fs
>
> [1] https://bugzilla.redhat.com/show_bug.cgi?id=579023
> [2] https://bugzilla.mozilla.org/show_bug.cgi?id=550455
> [3] http://koji.fedoraproject.org/koji/taskinfo?taskID=2092397
4:14 a.m.
On 04/23/2010 11:11 AM, Michal Hlavinka wrote:
On Friday 23 of April 2010 09:03:37 Martin Stransky wrote:
> Hi,
>
> we're patching mozilla packages only for really critical issues because
> of mozilla trademarks. We can't put any patch we want to the mozilla
> package and ship it as 'Firefox' or 'Thunderbird'.
just curious: is it possible to ship snapshots or trademarks does not allow
this too?
We can ship anything as an unbranded package. We do so in rawhide for
pre-released packages like firefox/thunderbird betas.
5:45 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/23/2010 12:03 AM, Martin Stransky wrote:
Hi,
we're patching mozilla packages only for really critical issues because
of mozilla trademarks. We can't put any patch we want to the mozilla
package and ship it as 'Firefox' or 'Thunderbird'.
To clarify a little further...
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. Since the browser and mail client are an
integral part of the daily user experience, we don't want to patch
things which would have an unintended effect.
Also, in the past, certain distributors have altered or broken standards
compliance in their clients with patches, and in continuing to do so,
they no longer ship with Mozilla trademarks. They have effectively
created a different browser and mail client that behaves differently on
some web sites or mail servers. Correctly following open standards is
extremely important for the internet, and the last thing I want is to
effectively create a fork in this way.
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.
I've asked for inclusion at upstream bug,
https://bugzilla.mozilla.org/show_bug.cgi?id=550455, if you want to
speed up the process you can reply there too.
Looking at both the Red Hat and upstream bugs,
The patch actually has already been accepted upstream already on
2010-04-01, just not in the same branch. There have been no complaints
of regression in the bug however from trunk testers.
As Firefox and Thunderbird are high usage and high visibility products
which tend to get many bugs and patches, we limit consideration to
certain classes of bugs with high impact. Based on the number of
different users commenting and the number of duplicates of the bug, I
think the impact of the bug was misjudged. Clearly this is leaving a
bad taste in users mouths, and I think we have a responsibility to both
Fedora and Mozilla to include a fix for it.
However, I'd like to put this in updates-testing for several days to
make sure it does not regress users of IMAP servers that do not suffer
from this issue. Given the number of affected users, I think it would
get high karma relatively quickly, so this is a case where I think karma
automatism should be disabled. Let's see if anyone responds negatively
in say 3 or 4 days before pushing this manually.
5:52 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Mon, Apr 26, 2010 at 15:45:53 -0700,
Christopher Aillon <caillon(a)redhat.com> wrote:
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.
And they should do things the right way as well. If they are bundling
libraries, they should stop doing that.
7:47 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/26/2010 03:52 PM, Bruno Wolff III wrote:
And they should do things the right way as well. If they are
bundling
libraries, they should stop doing that.
Mozilla has to bundle to ship on Windows, Mac, even their builds for
Linux where they don't control what versions of libraries are present on
the system, if they are installed at all (hooray choice!). That has
absolutely no bearing at all on Fedora however because we can and do
build with system libraries, and have done so for ages, precisely
because they are good at providing build support for that.
4:58 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Christopher Aillon wrote:
Mozilla has to bundle to ship on Windows, Mac, even their builds for
Linux where they don't control what versions of libraries are present on
the system, if they are installed at all (hooray choice!). That has
absolutely no bearing at all on Fedora however because we can and do
build with system libraries, and have done so for ages, precisely
because they are good at providing build support for that.
We can build with some system libraries. But there are still bundled ones
being used:
* libpng is bundled due to APNG support,
* libffi is bundled because there's no option to use the system version,
Debian has a patch to fix that, we don't use it.
There may also be more bundled libraries, I've just mentioned the ones where
I've noticed Debian has fixes.
(In addition, Thunderbird bundles xulrunner, but there's no fix available
for that issue at this time.)
Kevin Kofler
7:43 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 05:58 PM, Kevin Kofler wrote:
The OP had an issue w. thunderbird - which many find to be a pretty
decent mail client.
This thread has morphed ...
As for Firefox, I'd actually prefer to put fedora effort behind
chromium - google-chrome is an order of magnitude better than firefox -
if chromium can be made comparable we don't need to fuss about firefox -
lets fuss about chromium/google-chrome instead :-)
gene/
1:39 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Wed, Apr 28, 2010 at 1:43 AM, Mail Lists <lists(a)sapience.com> wrote:
On 04/27/2010 05:58 PM, Kevin Kofler wrote:
The OP had an issue w. thunderbird - which many find to be a pretty
decent mail client.
This thread has morphed ...
As for Firefox, I'd actually prefer to put fedora effort behind
chromium - google-chrome is an order of magnitude better than firefox -
if chromium can be made comparable we don't need to fuss about firefox -
lets fuss about chromium/google-chrome instead :-)
One more point which may not be directly on thread but which IS
important - many people use their browser for online banking and a
good number of banks will not allow login from any browser not on
their "approved" list. At present if you are running Firefox then
most banks will allow login - if you switch to another less well
recognised browser then users may find their bank will then not permit
online access (possibly even if you switch user agent string!). For
example in general chrome is working pretty well (yes it is not FOSS)
- but my bank will not permit login using chrome in linux (though it
will from Windows! This is because it won't allow a beta as it is not
released code!). A "released" chromium may get allowed for such use
but this is uncertain at present.
I think you need to be careful with such issues - and also ensure that
security is looked after vigorously for similar reasons. Switching
away from Firefox to Iceweasel or Coldcoyote or whatever may not play
nice with important sites such as banks!
--
mike c
7:22 a.m.
On Wed, Apr 28, 2010 at 07:39:37AM +0100, mike cloaked wrote:
On Wed, Apr 28, 2010 at 1:43 AM, Mail Lists
<lists(a)sapience.com> wrote:
> On 04/27/2010 05:58 PM, Kevin Kofler wrote:
>
>
> The OP had an issue w. thunderbird - which many find to be a pretty
> decent mail client.
>
> This thread has morphed ...
>
> As for Firefox, I'd actually prefer to put fedora effort behind
> chromium - google-chrome is an order of magnitude better than firefox -
> if chromium can be made comparable we don't need to fuss about firefox -
> lets fuss about chromium/google-chrome instead :-)
>
One more point which may not be directly on thread but which IS
important - many people use their browser for online banking and a
good number of banks will not allow login from any browser not on
their "approved" list.
thats why many popular browsers still use spoofed user-agent ids by
default.
If the banks do adhere to standards there is no use for checking user-agent.
Richard
1:24 a.m.
mike cloaked wrote:
One more point which may not be directly on thread but which IS
important - many people use their browser for online banking and a
good number of banks will not allow login from any browser not on
their "approved" list. At present if you are running Firefox then
most banks will allow login - if you switch to another less well
recognised browser then users may find their bank will then not permit
online access (possibly even if you switch user agent string!). For
example in general chrome is working pretty well (yes it is not FOSS)
- but my bank will not permit login using chrome in linux (though it
will from Windows! This is because it won't allow a beta as it is not
released code!). A "released" chromium may get allowed for such use
but this is uncertain at present.
I think you need to be careful with such issues - and also ensure that
security is looked after vigorously for similar reasons. Switching
away from Firefox to Iceweasel or Coldcoyote or whatever may not play
nice with important sites such as banks!
If we have it send Firefox's UA (or add an option to do it), which has been
proven in court not to be trademark infringement, and also have it return
"Firefox" for JS queries for the browser name, which is not trademark
infringement for the same reason (it's required for interoperability), they
have no way to tell the difference. (After all the code is the same except
for the name and for distro integration patches they have no way to detect.)
Kevin Kofler
12:58 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 02:58 PM, Kevin Kofler wrote:
* libffi is bundled because there's no option to use the system
version,
This option doesn't exist because it's impossible to use right now.
Just adding a --with-libffi doesn't actually make it useful since the
minimum required version of libffi hasn't been released yet, and libffi
releases don't come out that frequently. Note that a large portion of
the patchset between the latest libffi release and latest HEAD has come
from Mozilla, who have pushed all their changes upstream and have had
them merged successfully, and there are still more changes in the pipeline.
2:29 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Christopher Aillon (caillon(a)redhat.com) said:
This option doesn't exist because it's impossible to use
right now.
Just adding a --with-libffi doesn't actually make it useful since the
minimum required version of libffi hasn't been released yet, and libffi
releases don't come out that frequently. Note that a large portion of
the patchset between the latest libffi release and latest HEAD has come
from Mozilla, who have pushed all their changes upstream and have had
them merged successfully, and there are still more changes in the pipeline.
At least in F-13, we ship libffi-3.0.9, and the xulrunner-included version
claims to be 3.0.8. Of course, older releases may be a different story.
Bill
5 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/29/2010 12:29 PM, Bill Nottingham wrote:
Christopher Aillon (caillon(a)redhat.com) said:
> This option doesn't exist because it's impossible to use right now.
> Just adding a --with-libffi doesn't actually make it useful since the
> minimum required version of libffi hasn't been released yet, and libffi
> releases don't come out that frequently. Note that a large portion of
> the patchset between the latest libffi release and latest HEAD has come
> from Mozilla, who have pushed all their changes upstream and have had
> them merged successfully, and there are still more changes in the pipeline.
At least in F-13, we ship libffi-3.0.9, and the xulrunner-included version
claims to be 3.0.8. Of course, older releases may be a different story.
Sorry, Mozilla HEAD requires the latest HEAD of libffi (3.0.10pre),
which is what I was thinking of. Still, the 3.0.8 that Mozilla ships is
not vanilla and would need patching, though I don't know whether 3.0.9
has those patches or not. Since the work was still relatively recent,
it wasn't a huge concern of mine especially given the other work we've
got going on. I'm happy that Debian is adding this support and getting
it upstream, just as I'm sure they're happy we're doing the Thunderbird
XULRunner work. When it's upstreamed, we'll definitely use it, and I'm
sure when our work gets upstream, they'll use that too. I don't see why
we're the bad guy in this situation.
1:09 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Christopher Aillon wrote:
This option doesn't exist because it's impossible to use
right now.
Just adding a --with-libffi doesn't actually make it useful since the
minimum required version of libffi hasn't been released yet, and libffi
releases don't come out that frequently. Note that a large portion of
the patchset between the latest libffi release and latest HEAD has come
from Mozilla, who have pushed all their changes upstream and have had
them merged successfully, and there are still more changes in the
pipeline.
Debian builds xulrunner-1.9.2.3 against the system libffi-3.0.9 and it just
works. They even claim a minimum version of only >= 3.0.5 for the
dependency.
Kevin Kofler
1:59 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
I wrote:
Debian builds xulrunner-1.9.2.3 against the system libffi-3.0.9 and
it
just works. They even claim a minimum version of only >= 3.0.5 for the
dependency.
Actually they even claim only >= 3.0.4.
See for yourself:
http://packages.debian.org/experimental/xulrunner-1.9.2
Kevin Kofler
1:24 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 02:58 PM, Kevin Kofler wrote:
(In addition, Thunderbird bundles xulrunner, but there's no fix
available
for that issue at this time.)
I'm not sure why I'm bothering responding if you're not going to even
read responses, such as:
http://lists.fedoraproject.org/pipermail/devel/2010-April/135250.html
But, I'll re-iterate what Jan told you earlier in the thread that we've
been working on it with upstream and have been for a while, and it's a
HUGE undertaking. We've already made significant progress and have
gotten quite a bit of the changes needed merged upstream already. See
moz bug 377319 for the details.
2:06 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Thu, Apr 29, 2010 at 10:24 AM, Christopher Aillon <caillon(a)redhat.com> wrote:
But, I'll re-iterate what Jan told you earlier in the thread that
we've
been working on it with upstream and have been for a while, and it's a
HUGE undertaking. We've already made significant progress and have
gotten quite a bit of the changes needed merged upstream already. See
moz bug 377319 for the details.
And this is exactly why I personally find its okay to have the
exceptional status of thunderbird packaging. There's observable
progress towards fixing the problems that requires the exceptional
status.
I also think its perfectly fine for FESCO to review exceptional
situations on a timely basis. Granting exceptions to our packaging
policy shouldn't be done lightly..but it can also be beneficial when
there is recognition that the problem is significant and is going to
take some collaboration between Fedora developers and upstream to
solve. If there is a demonstrable commitment to untangling the
underlying problem..and there is observable progress..then renewing
exceptional status on review is reasonable to me personally and I
would hope reasonable for FESCO members.
-jef
10:50 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Thu, 2010-04-29 at 11:24 -0700, Christopher Aillon wrote:
On 04/27/2010 02:58 PM, Kevin Kofler wrote:
> (In addition, Thunderbird bundles xulrunner, but there's no fix available
> for that issue at this time.)
I'm not sure why I'm bothering responding if you're not going to even
read responses, such as:
http://lists.fedoraproject.org/pipermail/devel/2010-April/135250.html
That was uncalled for. The way I read Kevin's message, he
was /acknowledging/ that the xulrunner unbundling requires significant
development work and therefore left it out of his list of current
complaints about the packaging.
--
Matt
12:57 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Christopher Aillon wrote:
On 04/27/2010 02:58 PM, Kevin Kofler wrote:
> (In addition, Thunderbird bundles xulrunner, but there's no fix available
> for that issue at this time.)
I'm not sure why I'm bothering responding if you're not going to even
read responses, such as:
http://lists.fedoraproject.org/pipermail/devel/2010-April/135250.html
I know about this. No fix available at this time means that we can't expect
you to apply a fix which doesn't exist. That doesn't mean the OTHER problems
don't need fixing.
But, I'll re-iterate what Jan told you earlier in the thread that
we've
been working on it with upstream and have been for a while, and it's a
HUGE undertaking. We've already made significant progress and have
gotten quite a bit of the changes needed merged upstream already. See
moz bug 377319 for the details.
The biggest problem there is that you're trying to port the whole
Thunderbird to the "stable" APIs. Firefox can get away with using the
unstable API just fine, why can't this be done for Thunderbird? It just
means it needs to be updated in lockstep, even if the matching upstream
branch is not declared stable.
But of course the underlying true issue is that Mozilla is refusing to
guarantee backwards compatibility for the interfaces pretty much all
existing apps used and in several cases still use, instead trying to force
everyone to port to their new public API. I can't judge whether the original
interfaces were so poorly designed they really can't be kept compatible
(which reflects poorly on Mozilla) or whether they're just adding a useless
wrapper because they refuse to stay compatible (which reflects even more
poorly on them). But of course this issue is something only a full-blown
fork could even attempt to solve. And of course Firefox gets to use the
private API forever. Wasn't Mozilla among the ones complaining about IE
using private APIs? Why are they doing the same in Firefox?
Kevin Kofler
4:15 p.m.
On Fri, 2010-04-30 at 07:57 +0200, Kevin Kofler wrote:
But of course the underlying true issue is that Mozilla is refusing
to
guarantee backwards compatibility for the interfaces pretty much all
existing apps used and in several cases still use, instead trying to force
everyone to port to their new public API. I can't judge whether the original
interfaces were so poorly designed they really can't be kept compatible
(which reflects poorly on Mozilla) or whether they're just adding a useless
wrapper because they refuse to stay compatible (which reflects even more
poorly on them).
With the caveat that I'm not really familiar with the situation, I think
that assessment is unfair. If Mozilla feels the private interfaces are
suboptimal and wants to adjust the design before committing to long-term
stability, I think that's a reasonable decision. Let he who gets
everything right the first time throw the first stone.
--
Matt
2:04 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 12:45 AM, Christopher Aillon wrote:
On 04/23/2010 12:03 AM, Martin Stransky wrote:
We do have an agreement with Mozilla and as such, we are permitted
to
use the Firefox and Thunderbird trademarks.
You certainly mean: RH has an agreement
with Mozilla.
That's of no importance in a supposed to be "free" distro such as Fedora.
Ralf
4:55 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Christopher Aillon wrote:
Also, in the past, certain distributors have altered or broken
standards
compliance in their clients with patches, and in continuing to do so,
they no longer ship with Mozilla trademarks. They have effectively
created a different browser and mail client that behaves differently on
some web sites or mail servers. Correctly following open standards is
extremely important for the internet, and the last thing I want is to
effectively create a fork in this way.
You mean "compliance" with Mozilla's own "standards" such as APNG
which
require a bundled hacked version of a system library to support?
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.
I think that, sure, we should try to get patches upstreamed, but I don't see
why we'd need to wait for their approval before applying them, other than
due to the aforementioned trademark bureaucracy.
Firefox and Thunderbird are the ONLY high-profile packages in Fedora working
that way, and there must be very few packages in Fedora being maintained in
this style.
Kevin Kofler
5:46 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Tue, 2010-04-27 at 23:55 +0200, Kevin Kofler wrote:
You mean "compliance" with Mozilla's own
"standards" such as APNG which
require a bundled hacked version of a system library to support?
Kevin, you keep bringing up APNG, so let me address that one. I know the
story because the Mozilla implementation was originally written by a
former student, Andrew Smith. The decision to back PNG and fork libpng
was not made lightly -- it's a significant maintenance burden for
Mozilla.
APNG was created to fill a void -- there was a need for a modern
animated format with two qualities: it needed to be lightweight and
backward-compatible (degrade gracefully). After nearly a year of
discussion and consultation, the PNG group decided not to back it;
Mozilla gave it further thought and decided they needed it. It has since
been implemented by other browsers (such as Opera) and used in other
applications (such as WorldDMB, see ESTI TS 101 499 V2.2.1).
The alternative, MNG, is a heavy spec that offers no graceful
degradation for older browsers. It's been around for years, but is
almost never used. If it was a lighter spec, surely we'd be using it for
animated cursors and throbbers by now.
-Chris
6:16 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Chris Tyler wrote:
APNG was created to fill a void -- there was a need for a modern
animated format with two qualities: it needed to be lightweight and
backward-compatible (degrade gracefully). After nearly a year of
discussion and consultation, the PNG group decided not to back it;
Because they already have a perfectly fine format, MNG, which Mozilla
decided to drop support for. And in fact the main reason graceful
degradation is needed at all is browsers not supporting the format. What do
you suggest next? That all images need to "gracefully degrade" in a browser
which only supports GIF, such as the original Mosaic? The fact that support
for animations is being hacked into a format designed for still images is
one of the complaints the PNG group had about APNG.
Mozilla gave it further thought and decided they needed it. It has
since
been implemented by other browsers (such as Opera) and used in other
applications (such as WorldDMB, see ESTI TS 101 499 V2.2.1).
That's 2 of the major browsers. Come back when you get M$, Google, Apple
etc. to support it. Even MNG had wider support before Mozilla dropped it,
e.g. Konqueror in KDE 3 supported it just fine. (Sadly, in KDE 4, support
for animations in Konqueror regressed and MNG was one of the victims of it.
Other problems include: "stop animations" initially didn't work, I fixed
that; animated GIFs don't animate when resized, at least last I checked. But
that's not a Mozilla issue.)
The alternative, MNG, is a heavy spec that offers no graceful
degradation for older browsers. It's been around for years, but is
almost never used. If it was a lighter spec, surely we'd be using it for
animated cursors and throbbers by now.
The main reason it's almost never used is lack of browser support. Mozilla
dropping their existing support for it definitely did not help!
Kevin Kofler
12:58 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On 04/27/2010 02:55 PM, Kevin Kofler wrote:
I think that, sure, we should try to get patches upstreamed, but I
don't see
why we'd need to wait for their approval before applying them, other than
due to the aforementioned trademark bureaucracy.
You really don't see the value in having the engineers that own the code
give technical review?
Firefox and Thunderbird are the ONLY high-profile packages in Fedora
working
that way, and there must be very few packages in Fedora being maintained in
this style.
Getting sign-off is standard practice for the kernel too. Maybe we
should drop that package?
Anyway, it's unfortunate that this really isn't done more often. I
really think that as a project, we'd be doing a lot better if we
mandated upstream review before applying patches to any package if you
aren't an upstream maintainer of the code. As it is now, it's somewhat
scary to think how many packagers would take a bugfix patch and apply it
without being able to figure out if there's a potential hidden exploit
in it...
2:19 p.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
On Thu, Apr 29, 2010 at 9:58 AM, Christopher Aillon <caillon(a)redhat.com> wrote:
Anyway, it's unfortunate that this really isn't done more
often. I
really think that as a project, we'd be doing a lot better if we
mandated upstream review before applying patches to any package if you
aren't an upstream maintainer of the code. As it is now, it's somewhat
scary to think how many packagers would take a bugfix patch and apply it
without being able to figure out if there's a potential hidden exploit
in it...
The question is... is there a communication breakdown which let this
particular patch linger in the review process for too long ? And if
so, what can 'we' do to address that breakdown?
It definitely seems there's recognition from Mozilla that something in
the communication broke down from this sidebar discussion at LWN:
http://lwn.net/Articles/385171/
The question I have is.. do 'we' understand our role in driving
important issues up into upstream's review que to make sure it gets
looked at in a timely way?
It seems to me the review process worked like it was suppose to
here...but it just didn't get triggered in a timely manner...partly
because we didn't jump up and down about it being important.
-jef
10:46 p.m.
On Thu, 2010-04-29 at 10:58 -0700, Christopher Aillon wrote:
I really think that as a project, we'd be doing a lot better if
we
mandated upstream review before applying patches to any package if you
aren't an upstream maintainer of the code. As it is now, it's somewhat
scary to think how many packagers would take a bugfix patch and apply it
without being able to figure out if there's a potential hidden exploit
in it...
Review, perhaps, but not approval. Fedora and upstream are independent
organizations each pursing their own goals. Trademarks aside, Fedora
shouldn't be bound by upstream decisions any more than upstream is bound
by our packaging guidelines or obliged to accept patches to comply with
them. For comparison, disapproval from upstream libpng sure didn't stop
Mozilla from patching libpng with APNG support.
And the relevant qualification for a reviewer is knowledge of the code,
not affiliation with upstream.
--
Matt
1:04 a.m.
New subject: Thunderbird bz 579023 still not fixed even though there is an upstream fix available
Christopher Aillon wrote:
You really don't see the value in having the engineers that own
the code
give technical review?
I don't think this should be a requirement for each and every patch to ANY
Fedora package.
It is generally not necessary and delays fixing bugs a lot.
Anyway, it's unfortunate that this really isn't done more
often. I
really think that as a project, we'd be doing a lot better if we
mandated upstream review before applying patches to any package if you
aren't an upstream maintainer of the code. As it is now, it's somewhat
scary to think how many packagers would take a bugfix patch and apply it
without being able to figure out if there's a potential hidden exploit
in it...
And you think the average upstream is any better at this? Seriously?
Kevin Kofler
5103
days inactive
5111
days old
112 comments
32 participants
participants (32)
-
Adam Williamson -
Bill Nottingham -
Bruno Wolff III -
Chris Tyler -
Christopher Aillon -
Dr. Michael J. Chudobiak -
drago01 -
Felix Schwarz -
Genes MailLists -
Gianluca Sforna -
Jan Horak -
Jef Spaleta -
Jeffrey Ollie -
John Poelstra -
Kevin Kofler -
Martin Stransky -
Matt McCutchen -
Matthew Garrett -
Matthias Clasen -
Michal Hlavinka -
mike cloaked -
Nathanael D. Noblet -
Peter Lemenkov -
Quentin Armitage -
Rahul Sundaram -
Ralf Corsepius -
Richard Z -
Ryan Rix -
shmuel siegel -
Thomas Janssen -
Tom Callaway -
Tom Lane