Firefox 4: Path to Shipping
Damon Sicore
We've worked tremendously hard on Firefox 4, and it's time to ship it. I'm seeing the same burst of excitement and activity that we've seen in the endgame of every release. Over the past several days, component leads have again reduced their blockers by identifying hard blockers and those we can live without. We've around 160 hard blockers remaining, and historically it has taken us six weeks to reach RC once we have 100 blockers left. We must press hard now.
To Finish:
1) We have to reach Release Candidate status as quickly as possible, ideally finishing the hard blockers by the beginning of February and shipping final before the end of February. We'll need your help to balance these targets against the need to build a high quality product.
2) Bug counts demand another beta. We'll drive the beta bugs to zero and ship another beta. If we can't get them to zero in reasonable time, we'll repeat, deliberately. It depends on how quickly we can drive down the list of hard blockers that need beta feedback. This is our top development priority, since it pushes the rest of our schedule.
3) We need *everyone* to help in testing. Specifically: Do not disable Flash, Silverlight, or other major plugins as we need as many people testing these as possible. Windows users: We need to know if you are affected by hardware acceleration causing crashes or other issues. Don't just assume that someone else has filed a bug already. Make sure. Ask someone if you don't know how. This is very important.
MOST IMPORTANT: We must ship the best possible product we can. If a blocker needs more time, tell release drivers and component leads immediately. If you disagree with a blocking call, say so loudly. Do not be timid. This is your product, we need you to own it.
I know you're all tired and stressed. You all do incredible work every day, and you've built an amazing product. Stay focused. Be nice to each other. Firefox 4 is gonna kick ass, and you should be fiercely proud of it.
All my best,
Damon
Steven Roussey
Being a web developer mostly, I haven't had much of a chance to use Fx4, but it looks great and I can't wait to start trying some of the new stuff out!
Thanks!
-steve--
Kevin Dangoor
> This is somewhat self serving, since I volunteer on Firebug, but it would
> be nice to have at least one beta of Firefox that had working debugging
> tools. Then whatever mass of issues users face can be reported and hopefully
> fixed in the RC. Two beta cycles would be nice, but beggars and choosers and
> all that. ;)
>
I agree that it would be good to get at least one beta out. I have been
watching the Firebug-related bugs that I'm aware of and they look to be
getting attention. Honza and I talk regularly about what's needed for
Firebug and Firefox 4+Firebug 1.7 does look to be shaping up.
All of which to say, "yep! we're working on it!"
Kevin
--
Kevin Dangoor
work: http://mozilla.com/
email: kdan...@mozilla.com <k...@blazingthings.com>
blog: http://www.BlueSkyOnMars.com
Steven Roussey
> watching the Firebug-related bugs that I'm aware of and they look to be
> getting attention. Honza and I talk regularly about what's needed for
> Firebug and Firefox 4+Firebug 1.7 does look to be shaping up.
> All of which to say, "yep! we're working on it!"
Yes, John has said that important fixes have come very rapidly lately. :) My only worry is that with all of five people testing it (and that doesn't include me yet, since I don't have a build of that special branch), that when real people test it, something will come up that is too late to fix/change by the time the report comes in. And then we will have something like the last entire year with no closure variables in the debugger. And that kinda killed my use of Firebug even though I'm a contributor to it!
Anyhow, I don't want to veer off topic. I'm merely suggesting that if there was going to be one or two releases before final at the end of February, that there be three in that same time period. ("I'm *only* suggesting"... yeah I know ;)
Then all I need to do is to get Google to fix Chrome so it doesn't hang on the getfirebug.com website by the time Firefox 4 comes out. It has all the makings of a conspiracy theory if I didn't know better...
-steve--
Steven Roussey
> watching the Firebug-related bugs that I'm aware of and they look to be
> getting attention. Honza and I talk regularly about what's needed for
> Firebug and Firefox 4+Firebug 1.7 does look to be shaping up.
> All of which to say, "yep! we're working on it!"
Yes, John has said that important fixes have come very rapidly lately. :) My only worry is that with all of five people testing it (and that doesn't include me yet, since I don't have a build of that special branch), that when real people test it, something will come up that is too late to fix/change by the time the report comes in. And then we will have something like the last entire year with no closure variables in the debugger. And that kinda killed my use of Firebug even though I'm a contributor to it!
Steve Fink
> Yes, John has said that important fixes have come very rapidly lately.
> :) My only worry is that with all of five people testing it (and that
> doesn't include me yet, since I don't have a build of that special
> branch), that when real people test it, something will come up that is
> too late to fix/change by the time the report comes in. And then we
> will have something like the last entire year with no closure
> variables in the debugger. And that kinda killed my use of Firebug
> even though I'm a contributor to it!
I don't disagree. Although Firebug looks like it should be totally
usable though far from problem-free in beta9, the long outage period
leads to a potential for some unpleasant bugs to come crawling out into
the light at the last minute when people finally start testing it for real.
https://bugzilla.mozilla.org/show_bug.cgi?id=623351 is probably the
worst for Firebug of what didn't make it into beta9. Is your closure
problem the same as https://bugzilla.mozilla.org/show_bug.cgi?id=609668
or are there further problems?
Steven Roussey
> worst for Firebug of what didn't make it into beta9.
Crashing is bad (though that bug looks familiar, i think it has been around a long time?). Then there is stepping, etc., where I see problems and just wait for another day. But it doesn't look like much of your work on that issue has made it to the beta. Is it in beta9?
> Is your closure
> problem the same as https://bugzilla.mozilla.org/show_bug.cgi?id=609668
> or are there further problems?
No, FF4 has that fixed! Just noticed that this morning, actually. That bug should be closed (though somewhere there ought to be a test, since that bug has resurfaced before, and the FB test only covers half of the issue -- that part I can fix -- but it is really the wrong place, you should have your own tests for it).
I'm prolific at finding bugs in FB, but I will be traveling quite a bit in Feb., so I'll be passing the buck on that one... :0
More betas == more passing the buck.
I did say my suggestion was a little self serving, did I not? ;)
-steve--
PS: Got me excited about FF4, so I started to add some minor fixes to FB as I read through the "Firefox 4 for developers" doc. :)
Steven Roussey
> worst for Firebug of what didn't make it into beta9.
Crashing is bad (though that bug looks familiar, i think it has been around a long time?). Then there is stepping, etc., where I see problems and just wait for another day. But it doesn't look like much of your work on that issue has made it to the beta. Is it in beta9?
> Is your closure
> problem the same as https://bugzilla.mozilla.org/show_bug.cgi?id=609668
> or are there further problems?
No, FF4 has that fixed! Just noticed that this morning, actually. That bug should be closed (though somewhere there ought to be a test, since that bug has resurfaced before, and the FB test only covers half of the issue -- that part I can fix -- but it is really the wrong place, you should have your own tests for it).
Eli Gabay
FF4 is pretty irrelevant at this point regarding your market share.
You broke everything that worked, and so the masses will move on to Chrome, which is hassle free and works with more sites, and is updated all the time.
FF4 will still be used by web developers, such as myself.
But I no longer recommend FF4 for the average user, who cant stand waiting more than 2-3 seconds for the browser to load.
You guys lost track of the goal, and working on a bloated browser, based primarily on your pride.
Good luck.
Eli Gabay
I have been using FF nearly since it came out, especially firebug, an indispensable tool for web developers. And I have been using the FF4 beta since day 1.
So you broke firebug, FF4 is still full of bugs, AND still has the same frustrating memory and performance problems as FF3. The UI is still ugly, the interface clunky. Why did you not just copy Chrome's interface. Since Day 1, they did right everything you did wrong over the years.
Your pathetic pride is what makes FF suck now.
The average user barely knows what a browser is. Most of them will ask you how to use the internet on your computer if you hide the IE icon. So by shooting your loyal tech users in the foot, they are gradually stopping to spread your browser by word of mouth, which was your viral channel.
RIP FF
Rob Campbell
Thanks for the excellent feedback!
All the best,
Rob
PS, to make my point clear, you suck. Cheers.
> _______________________________________________
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
Eli Gabay
But really, wake up.
Many people, including myself, would really like to see this browser get ahead.
But you guys have to start with first making priorities.
FIX your memory issues.
FIX your application-loading speed.
FIX Firebug.
LEARN to copy what others do better than you.
Mike Beltzner
cheers,
mike (dev.planning moderator)
Steve Fink
>> https://bugzilla.mozilla.org/show_bug.cgi?id=623351 is probably the
>> worst for Firebug of what didn't make it into beta9.
> Crashing is bad (though that bug looks familiar, i think it has been around a long time?). Then there is stepping, etc., where I see problems and just wait for another day. But it doesn't look like much of your work on that issue has made it to the beta. Is it in beta9?
Steven Roussey
Then I withdraw my request. People will get to test with beta9, beta10, and RC, and that sounds like it should be enough. Not that a beta11 squeezed in there wouldn't be welcome...
-steve--
Steven Roussey
Then I withdraw my request. People will get to test with beta9, beta10, and RC, and that sounds like it should be enough. Not that a beta11 squeezed in there wouldn't be welcome...
-steve--
Guillermo
I'm pretty excited with FF4 launch, but I'm wondering if you are being too much optimistic. If it takes six weeks to finish 100 hard blockers, how are you going to correct 160 in only two weeks? Is not too much pressure for developers?
Congrats on your work, I expect FF4 to be a great browser!
Greetings
Guillermo Julián
Messier
- Slow on Ubuntu 10.10, even with all add-ons disabled
- Was causing me problems with my webmail provider.It keeps telling me
that my browser is not supported. Is there a way to "fake" the version
and/or name of my browser so I can access my email with no warning ?
l4m3rx
I guess this is not the place to write this , but i'll do it
anyway ... ff4 is great (i'm using nightly build from his baby years
when it was just 3.7) , And now when I see it almost done ...it's
great. Please don't listen to the masses , and the ideas about doing
firefox more chrome like (yak!). So anyway, the things I'll really
think you should do some more work in later version (4.1 maybe) is
doing some more work on the plugin container , maybe spawning thread
for every flash .... the other thing that will be really good (well ,
it will be only for the linux users, but i'm one of them! :P ) is to
take advantage of the kernel dynamic cgroups, I think that some kind
of active grouping of the plugin conteinter and main firefox stuff
will be great,... oh and the last think that comes to my mind is ,
maybe inplementing firebug or something like that in firefox , most
of the web developer are using firefox mainly 'cose of firebug , and
it's not really a bad way to go ...
Anyway great job ! Keep up the good work, and again , 10x for not
going chrome like :]