You are correct, that would be a great feature! It is being worked on, but it actually is not that easy to implement. Hopefully it'll be released soon.
Hiding the version number with an option sounds dangerous - how will people file bug reports? Replacing the version number likewise.
A GUI for about:config? No geek needs a GUI:P (proud Linux commandline user here;)
Anyhow, if you don't like some of the shiny features, you can disable them. Or you can create an addon that disables them. We could add an option to disable "shiny stuff", but we would have a hard time agreeing on what is shiny and what isn't (your list is different from mine, for example). But the point is that anyone can customize the browser however they want.
There is no advantage. Your version number is suppose to communicate something to the user. A major version change means you have something new and significant. A minor version means you are making a small tweak or minor refinement.
A major version number often signifies potentially incompatible changes, while minor version numbers keep stable APIs and functionality. In that context, Firefox and Chrome bumping the major version number is correct: These 6-week updates are quick, but they do change functionality in ways that break stuff (that is, websites render differently, potentially wrongly).
I do agree there are downsides to this numbering! But I am just saying it has a certain logic, and advantage.
Re:Plugins are tied to version in FF, but not Chro
on
Firefox 8.0 Released
·
· Score: 2
That is where we are headed with 'jetpack' addons. Those have an API like you said. But most addons today use the older interface, which doesn't work that way - and it will be a long time until most addons are rewritten to the new API.
Hi there, I'm a developer at Mozilla. Some responses to your comments:
The sanity of the Firefox team is under question as of late. From what I can remember:
* Incrementing the major version number with every slight tweak is annoying.
I understand that this annoys some people. But both Chrome and Firefox do it now, and benefits and detriments are well known. It's not a perfect approach, but it does have its advantages. I don't think both Google and Mozilla are 'insane';)
* Worse yet, the reasoning behind it is stupid. They just want their version number to be big, like IE.
The main reason for Chrome and Firefox doing this is to get improvements faster to users. Rapid releases allow that.
Mozilla also has the reason that it is following Google's lead. Google started with this version numbering scheme, and not inventing a new one is better for everyone - less confusion.
* Major feature creep: they keep talking about the browser as an OS, and 3D acceleration, and stuff that has no purpose in a browser.
That is a long discussion, for sure! But this is nothing to do with Firefox. All browsers are including 3D acceleration (well, except for IE) and other OS-like features. Google is even pushing native code in the browser (which I think is taking things too far).
* The long-standing issues about Firefox are being ignored: primarily memory and performance.
We are working very hard on those issues. If you try this release, I think you'll see significant improvements on both issues, and there are even more in the pipeline for the versions coming up afterwards.
How is this that contemptible? It's not immoral, unethical, or even evil. It's a painter saying no I like my painting with purple grass,
I think the reason this annoys so many people, is the expectation of openness. Google has an open bug tracker, and Chromium is open source. So people assume it is open in all ways, like say the Linux kernel is: The Linux kernel is not only open source, and has an open bug tracker, but it is also openly developed as a community open source project.
But Chromium is not a community open source project. It's run by a corporation, Google. It isn't developed completely in the open, and decisions are made from within Google.
There is nothing wrong with that, of course (although I personally do prefer fully open projects to partially open ones). But the problem is that the amount of openness that Chromium does have, leads people to assume it is fully open when it isn't, which ends up annoying a lot of people.
I understand the concept of RAM caching - it's not exactly rocket science. But how does Firefox/Opera/IE free up memory when the OS needs it? What is the mechanism by which the OS tells the browser to free memory?
It depends on the OS. But you can query many OSes to find out how much physical memory is being used, and when physical memory is running low, your browser can free up some caches. If done right, this will happen before any significant amount of swapping occurs, so the browser can benefit from free physical memory to speed things up, but won't hog it so other applications end up starved. (But the downside is that some users see large amounts of memory being used and assume that means a memory leak.)
In addition, some OSes give explicit indications of memory running out, I believe Maemo did this.
Now they are copying the game engine. (another feature that chrome already has)
Why innovate for yourselves if you can just copy everything that chrome has done already?!?
Neither Firefox nor Chrome has a full game engine. They both have WebGL though..
Paladin/Gladius is just a game engine for WebGL - not a new thing that will be part of the browser itself. So it will run in both Firefox and Chrome. There are plenty of such game engines, hopefully this new one will be useful too.
Between the version number / release cycle insanity and this, I think it's finally time to switch. What a shame, I've been using Firefox since it was called Phoenix. But the update today broke another extension, and building 3D into the browser is a sign of insanity that I thought we had wiped out with the demise of VMRL.
Dear Mozilla developers: If it's not something the majority of your users are going to actually use, it belongs into an extension or a plugin. Also, there are already several 3D engines with Firefox plugins, with years of experience in the field, because you don't build a good engine in a lazy summer. So with all due respect, what the fuck are you thinking?
I'm afraid you've fallen into the way-too-common bloatware trap: Not realizing when your product is feature complete and what it needs is polishing, not more stuff bolted on. There's enough CSS3 and HTML5 support still missing, for example.
Time to take a serious look at Chrome.:-(
Both Firefox and Chrome use the same 3D API: WebGL. Firefox is not adding any new browser components here. The Slashdot summary is a little misleading - Paladin/Gladius is just a game engine written in JS/HTML5/WebGL, not something new being included in the browser. So there is nothing being bolted on here. (Or are you saying that WebGL is unneeded bloat? That's a separate argument, but with Firefox, Chrome, Safari and Opera all supporting it or close to releasing support for it, that is a done deal I think.)
Regarding the update problems, I am a Firefox dev, and all I can say is that we realize there is a big problem, and are working very hard on it. There is no quick fix though. But the upcoming updates will improve things quite a lot, you will have far fewer incompatible addons. And yes, we fully realize that polish and so forth is much more important than new shiny things, and labs projects like this new game engine. We have far, far more people working on polish than any of those other things. But no Slashdot stories on those;)
Did Mozilla go hire some MBAs or something? That's the only rational explanation for this idiocy.
The userbase has rejected rapid release. They hate it. Users are leaving the browser faster then ever before ever since it started.
So Mozilla's response is... even faster releases?
Hi, I am a Firefox dev. The answer to your question is no: The answer is not faster releases. We are not currently planning to do faster releases, despite the Slashdot headline.
What is the link then? Someone - not sure if a Mozilla developer or not - posted the suggestion to make it faster. Since Firefox's development is open, anyone can post whatever they want whenever they want. There was some debate, most of it negative - as you would expect. Then someone posted it to Slashdot, where it was picked up.
So, no faster releases. What actually is the Mozilla response to the current situation: To fix the problems. We are working to make updates silent and break less addons. We've also made it so third parties can't install addons without your permission. All of this is in response to user feedback. Hopefully some of that stuff will be posted to Slashdot too;)
Lua is very Javascript-like already except it's very small, simple, clean, and fast. Much faster; LuaJIT is incredible.
LuaJIT is not much faster anymore.
LuaJIT gets to about 2-3X slower than the fastest gcc, while the latest JS engines (V8 with CrankShaft, SpiderMonkey with TypeInference) get to 3-5X slower than gcc. That's still a significant difference, but the JavaScript engines are also improving faster. In a year or so the difference will have vanished.
Those numbers are also a little misleading. They are mainly simple benchmarks, where LuaJIT's tracer is phenomenal. But if you take a complete program, with a lot of use of classes/inheritance/closures/etc., LuaJIT won't do as well - it hasn't been tuned for those things. In other words, on real-world code the difference would be nonexistent or even reversed. But what is representative 'real-world code' is debatable so it's hard to come up with numbers for that. (But you will often see it in practice in the field.)
What someone needs to do is actually fix the add-on code in firefox itself so that users don't have to jump through hoops for every release.
We are working as hard as we can on that. The reasons this is difficult have already been discussed here on Slashdot many times (but I am happy to elaborate if you want), and there are a lot of fixes already in place - for example, scanning of addons on addons.mozilla.org and marking them as compatible automatically. But there is no simple fix that will make everything just work, without a new model for addons (which Firefox has, the addon SDK/jetpack - but it is new and older addons don't use it).
Is this technology FOSS? Where can I get non-obfuscated sources for this? There isn't even a copyright notice or any information about the developers anywhere on the page...highly unusual.
I can't get this website to work, and have no idea what technology they use. But if you want an open source way to run Python in your browser, you can check out this demo (source code and build instructions are in the emscripten source code on github).
I think Linux is a good comparison: It also has no stable internal APIs, and no binary ABI, like Firefox currently does. For both projects, its important to do that to allow for quick progress.
Mozilla did announce the 6-week rapid release schedule ahead of time. As I said before, I agree there are problems with it - but it was announced publicly, on the mailing lists, blogposts, etc.
I believe that you should not have to update and maintain your non-jetpack addon every 6 weeks! If your addon is hosted on addons.mozilla.org, it should be auto-scanned for compatibility before each update, and marked as compatible if it is. And in the majority of cases it should be compatible, since 6 weeks isn't enough to change very much, unless your addon happens to use APIs that are currently very much in flux.
Regarding jetpack, I have not worked with it myself, so i really don't know much about it. But as I understand, it isn't meant to be able to do everything XUL can. Jetpack is purposefully very limited - like the Chrome addon API - in order to make it very stable and noninvasive (to the browser), so browser updates don't break stuff. So the APIs you are missing might be that sort of stuff - purposefully left out, as opposed to not yet developed - but again, I am not a jetpack user so I can't be sure.
I definitely understand your criticism. I do feel it is overly pessimistic though, even if you are correct on the main points you make.
But more than that, I do disagree with a main point: That you say users want stability and consistency over all else. To some degree it's true, they do want those things. But they also want other things as well. The fact that users want more features than the web currently provides can be seen in the success of Flash, for example, and in the mobile space, in that native iPhone and Android apps are much more successful than websites (or 'web apps'). So users do want those additional things, and when the web doesn't provide them, they go elsewhere.
The problem is that those elsewheres are closed silos. Some more than others, but none is truly open or standards-based.
We can keep the web where it is now, stable and unchanged, and innovate in those other places. There are upsides to doing that, but I strongly believe the downsides are much bigger: Our best hope for an open, standards-based future in computing is the web. The web isn't far from doing what those other platforms (Flash, iPhone native apps, etc.) can do, but it does need more work.
Sadly the price of doing that work and innovation on the web is that things are no longer as stable as they were. But it's the best way to get to where users want things to be: They want to be able to buy the devices they want and run the stuff they want on them. The web platform will be able to provide that, but not the iPhone platform, the Flash platform or even the Android app platform. That isn't a silicon valley tech dream, that's a universal dream.
AdBlockPlus would not work as a JetPack addon, yeah. That's also why Chrome's AdBlockPlus isn't as good - there is a benefit to accessing internal APIs. It's a tradeoff - will work for some addons, not for others.
You make a lot of good points. I agree that Firefox would benefit from focusing on a niche not occupied by Chrome. What would that be, though? Are you saying we should find it, or are you saying that there isn't such a niche?
From my perspective, Firefox fills the role of a 100% open source, 100% community developed browser. As an open source supporter, I believe the web needs such a thing. WebKit is open source too, but it is basically controlled by two massive corporations - Google and Apple, and it isn't developed as openly as Firefox. Also, Google and Apple include a lot of closed-source elements in their browsers, Google less so, but still - things from print preview to Flash itself are proprietary code included in Chrome.
I don't know if you'd call that a niche. But it's the reason I think Firefox is important and why I'm a Firefox developer. It's hard to use for marketing purposes, though;)
"I've found this bug in Firefox..."
"Do you run the latest version?"
"I don't know. I'm running the version my distro gives me."
"So which one is it?"
"I don't know. It won't tell me."
"Please update to the latest version."
"Well, I already have the latest version my distro gives me. If this is actually the latest version, I have no idea."
As the article states, the version number will still be available through the help menu. It will just be slightly less noticeable. It used to be in two menu elements, and after this change it will be in one.
That's the problem here. Firefox's ever changing APIs which are always breaking add-ons. The Chrome add-on API is much more limited and as such doesn't need to change as frequently or as drastically. How Firefox thinks they're going to succeed by becoming a crappier version of Chrome is beyond me.
Firefox has Jetpack addons, which are basically like Chrome's - a limited, stable API. Such addons should work even when the browser updates every 6 weeks. However, jetpack is fairly new, so most addons aren't written using it, and that means they can use internal APIs that do change.
In the long term, most addons will probably be Jetpack, with non-Jetpack addons being special cases that actually do need those internal APIs, and the developers of those addons will need to keep up with the latest Firefox version for them. But, meanwhile there is definitely a lot of inconvenience about this.
I agree the addon compatibility issue is a big problem, exactly as you said. Perhaps we should have waited until it was solved or much closer to being solved before doing rapid releases. It's hard to know in advance what timing is optimal for this kind of thing, and like everyone we make mistakes sometimes.
However I don't agree with this part of your comment:
I have a feeling that Mozilla lives in its own little bubble where competing on features and speed, setting the standards pace, and "pushing the web platform forward" seems critical. I'm sure it all feels great in Silicon Valley, but not all web development happens in cool startups champing at the bit for the latest canvas 3D gizmo. I'd hazard that there's even more spread across businesses large and small.
Before FF4, everyone here on Slashdot and elsewhere were saying that Firefox was losing users to Chrome because Chrome was faster. And I think that was true to a significant extent - users do notice speed, and it gets a lot of attention by power users who recommend browsers to normal users. So it was a correct choice to focus on speed for FF4, and as a consequence, today people are saying that Firefox's weak side is something other than speed - because the speed issue is largely resolved, we are very competitive there now. So I don't think any "Silicon Valley bubble effect" had an effect on that choice. Anyhow, most of our developers don't live in Silicon Valley;) (like most open source projects, we are very spread out)
Slashdot Mirror
You are correct, that would be a great feature! It is being worked on, but it actually is not that easy to implement. Hopefully it'll be released soon.
Hiding the version number with an option sounds dangerous - how will people file bug reports? Replacing the version number likewise.
:P (proud Linux commandline user here ;)
A GUI for about:config? No geek needs a GUI
Anyhow, if you don't like some of the shiny features, you can disable them. Or you can create an addon that disables them. We could add an option to disable "shiny stuff", but we would have a hard time agreeing on what is shiny and what isn't (your list is different from mine, for example). But the point is that anyone can customize the browser however they want.
There is no advantage. Your version number is suppose to communicate something to the user. A major version change means you have something new and significant. A minor version means you are making a small tweak or minor refinement.
A major version number often signifies potentially incompatible changes, while minor version numbers keep stable APIs and functionality. In that context, Firefox and Chrome bumping the major version number is correct: These 6-week updates are quick, but they do change functionality in ways that break stuff (that is, websites render differently, potentially wrongly).
I do agree there are downsides to this numbering! But I am just saying it has a certain logic, and advantage.
That is where we are headed with 'jetpack' addons. Those have an API like you said. But most addons today use the older interface, which doesn't work that way - and it will be a long time until most addons are rewritten to the new API.
The sanity of the Firefox team is under question as of late. From what I can remember:
* Incrementing the major version number with every slight tweak is annoying.
I understand that this annoys some people. But both Chrome and Firefox do it now, and benefits and detriments are well known. It's not a perfect approach, but it does have its advantages. I don't think both Google and Mozilla are 'insane' ;)
* Worse yet, the reasoning behind it is stupid. They just want their version number to be big, like IE.
The main reason for Chrome and Firefox doing this is to get improvements faster to users. Rapid releases allow that.
Mozilla also has the reason that it is following Google's lead. Google started with this version numbering scheme, and not inventing a new one is better for everyone - less confusion.
* Major feature creep: they keep talking about the browser as an OS, and 3D acceleration, and stuff that has no purpose in a browser.
That is a long discussion, for sure! But this is nothing to do with Firefox. All browsers are including 3D acceleration (well, except for IE) and other OS-like features. Google is even pushing native code in the browser (which I think is taking things too far).
* The long-standing issues about Firefox are being ignored: primarily memory and performance.
We are working very hard on those issues. If you try this release, I think you'll see significant improvements on both issues, and there are even more in the pipeline for the versions coming up afterwards.
I don't want PDFs to open in the web browser. I want to open them in Acrobat in another window. Let the browser be a browser and Acrobat be Acrobat!
That's fine, this is a Firefox addon: Just don't use it. Addons are optional.
Linus rejects stuff, sure. But the point is that the project is developed in an open community manner. That doesn't mean some things are not rejected.
How is this that contemptible? It's not immoral, unethical, or even evil. It's a painter saying no I like my painting with purple grass,
I think the reason this annoys so many people, is the expectation of openness. Google has an open bug tracker, and Chromium is open source. So people assume it is open in all ways, like say the Linux kernel is: The Linux kernel is not only open source, and has an open bug tracker, but it is also openly developed as a community open source project.
But Chromium is not a community open source project. It's run by a corporation, Google. It isn't developed completely in the open, and decisions are made from within Google.
There is nothing wrong with that, of course (although I personally do prefer fully open projects to partially open ones). But the problem is that the amount of openness that Chromium does have, leads people to assume it is fully open when it isn't, which ends up annoying a lot of people.
I understand the concept of RAM caching - it's not exactly rocket science. But how does Firefox/Opera/IE free up memory when the OS needs it? What is the mechanism by which the OS tells the browser to free memory?
It depends on the OS. But you can query many OSes to find out how much physical memory is being used, and when physical memory is running low, your browser can free up some caches. If done right, this will happen before any significant amount of swapping occurs, so the browser can benefit from free physical memory to speed things up, but won't hog it so other applications end up starved. (But the downside is that some users see large amounts of memory being used and assume that means a memory leak.)
In addition, some OSes give explicit indications of memory running out, I believe Maemo did this.
Now they are copying the game engine. (another feature that chrome already has)
Why innovate for yourselves if you can just copy everything that chrome has done already?!?
Neither Firefox nor Chrome has a full game engine. They both have WebGL though..
Paladin/Gladius is just a game engine for WebGL - not a new thing that will be part of the browser itself. So it will run in both Firefox and Chrome. There are plenty of such game engines, hopefully this new one will be useful too.
Between the version number / release cycle insanity and this, I think it's finally time to switch. What a shame, I've been using Firefox since it was called Phoenix. But the update today broke another extension, and building 3D into the browser is a sign of insanity that I thought we had wiped out with the demise of VMRL.
Dear Mozilla developers: If it's not something the majority of your users are going to actually use, it belongs into an extension or a plugin. Also, there are already several 3D engines with Firefox plugins, with years of experience in the field, because you don't build a good engine in a lazy summer. So with all due respect, what the fuck are you thinking?
I'm afraid you've fallen into the way-too-common bloatware trap: Not realizing when your product is feature complete and what it needs is polishing, not more stuff bolted on. There's enough CSS3 and HTML5 support still missing, for example.
Time to take a serious look at Chrome. :-(
Both Firefox and Chrome use the same 3D API: WebGL. Firefox is not adding any new browser components here. The Slashdot summary is a little misleading - Paladin/Gladius is just a game engine written in JS/HTML5/WebGL, not something new being included in the browser. So there is nothing being bolted on here. (Or are you saying that WebGL is unneeded bloat? That's a separate argument, but with Firefox, Chrome, Safari and Opera all supporting it or close to releasing support for it, that is a done deal I think.)
;)
Regarding the update problems, I am a Firefox dev, and all I can say is that we realize there is a big problem, and are working very hard on it. There is no quick fix though. But the upcoming updates will improve things quite a lot, you will have far fewer incompatible addons. And yes, we fully realize that polish and so forth is much more important than new shiny things, and labs projects like this new game engine. We have far, far more people working on polish than any of those other things. But no Slashdot stories on those
Thanks for the input, yeah, I do try to raise awareness of this stuff there.
Did Mozilla go hire some MBAs or something? That's the only rational explanation for this idiocy.
The userbase has rejected rapid release. They hate it. Users are leaving the browser faster then ever before ever since it started.
So Mozilla's response is... even faster releases?
Hi, I am a Firefox dev. The answer to your question is no: The answer is not faster releases. We are not currently planning to do faster releases, despite the Slashdot headline.
;)
What is the link then? Someone - not sure if a Mozilla developer or not - posted the suggestion to make it faster. Since Firefox's development is open, anyone can post whatever they want whenever they want. There was some debate, most of it negative - as you would expect. Then someone posted it to Slashdot, where it was picked up.
So, no faster releases. What actually is the Mozilla response to the current situation: To fix the problems. We are working to make updates silent and break less addons. We've also made it so third parties can't install addons without your permission. All of this is in response to user feedback. Hopefully some of that stuff will be posted to Slashdot too
Lua is very Javascript-like already except it's very small, simple, clean, and fast. Much faster; LuaJIT is incredible.
LuaJIT is not much faster anymore.
LuaJIT gets to about 2-3X slower than the fastest gcc, while the latest JS engines (V8 with CrankShaft, SpiderMonkey with TypeInference) get to 3-5X slower than gcc. That's still a significant difference, but the JavaScript engines are also improving faster. In a year or so the difference will have vanished.
Those numbers are also a little misleading. They are mainly simple benchmarks, where LuaJIT's tracer is phenomenal. But if you take a complete program, with a lot of use of classes/inheritance/closures/etc., LuaJIT won't do as well - it hasn't been tuned for those things. In other words, on real-world code the difference would be nonexistent or even reversed. But what is representative 'real-world code' is debatable so it's hard to come up with numbers for that. (But you will often see it in practice in the field.)
What someone needs to do is actually fix the add-on code in firefox itself so that users don't have to jump through hoops for every release.
We are working as hard as we can on that. The reasons this is difficult have already been discussed here on Slashdot many times (but I am happy to elaborate if you want), and there are a lot of fixes already in place - for example, scanning of addons on addons.mozilla.org and marking them as compatible automatically. But there is no simple fix that will make everything just work, without a new model for addons (which Firefox has, the addon SDK/jetpack - but it is new and older addons don't use it).
Is this technology FOSS? Where can I get non-obfuscated sources for this? There isn't even a copyright notice or any information about the developers anywhere on the page...highly unusual.
I can't get this website to work, and have no idea what technology they use. But if you want an open source way to run Python in your browser, you can check out this demo (source code and build instructions are in the emscripten source code on github).
I think Linux is a good comparison: It also has no stable internal APIs, and no binary ABI, like Firefox currently does. For both projects, its important to do that to allow for quick progress.
Mozilla did announce the 6-week rapid release schedule ahead of time. As I said before, I agree there are problems with it - but it was announced publicly, on the mailing lists, blogposts, etc.
I believe that you should not have to update and maintain your non-jetpack addon every 6 weeks! If your addon is hosted on addons.mozilla.org, it should be auto-scanned for compatibility before each update, and marked as compatible if it is. And in the majority of cases it should be compatible, since 6 weeks isn't enough to change very much, unless your addon happens to use APIs that are currently very much in flux.
Regarding jetpack, I have not worked with it myself, so i really don't know much about it. But as I understand, it isn't meant to be able to do everything XUL can. Jetpack is purposefully very limited - like the Chrome addon API - in order to make it very stable and noninvasive (to the browser), so browser updates don't break stuff. So the APIs you are missing might be that sort of stuff - purposefully left out, as opposed to not yet developed - but again, I am not a jetpack user so I can't be sure.
I definitely understand your criticism. I do feel it is overly pessimistic though, even if you are correct on the main points you make.
But more than that, I do disagree with a main point: That you say users want stability and consistency over all else. To some degree it's true, they do want those things. But they also want other things as well. The fact that users want more features than the web currently provides can be seen in the success of Flash, for example, and in the mobile space, in that native iPhone and Android apps are much more successful than websites (or 'web apps'). So users do want those additional things, and when the web doesn't provide them, they go elsewhere.
The problem is that those elsewheres are closed silos. Some more than others, but none is truly open or standards-based.
We can keep the web where it is now, stable and unchanged, and innovate in those other places. There are upsides to doing that, but I strongly believe the downsides are much bigger: Our best hope for an open, standards-based future in computing is the web. The web isn't far from doing what those other platforms (Flash, iPhone native apps, etc.) can do, but it does need more work.
Sadly the price of doing that work and innovation on the web is that things are no longer as stable as they were. But it's the best way to get to where users want things to be: They want to be able to buy the devices they want and run the stuff they want on them. The web platform will be able to provide that, but not the iPhone platform, the Flash platform or even the Android app platform. That isn't a silicon valley tech dream, that's a universal dream.
AdBlockPlus would not work as a JetPack addon, yeah. That's also why Chrome's AdBlockPlus isn't as good - there is a benefit to accessing internal APIs. It's a tradeoff - will work for some addons, not for others.
You make a lot of good points. I agree that Firefox would benefit from focusing on a niche not occupied by Chrome. What would that be, though? Are you saying we should find it, or are you saying that there isn't such a niche?
;)
From my perspective, Firefox fills the role of a 100% open source, 100% community developed browser. As an open source supporter, I believe the web needs such a thing. WebKit is open source too, but it is basically controlled by two massive corporations - Google and Apple, and it isn't developed as openly as Firefox. Also, Google and Apple include a lot of closed-source elements in their browsers, Google less so, but still - things from print preview to Flash itself are proprietary code included in Chrome.
I don't know if you'd call that a niche. But it's the reason I think Firefox is important and why I'm a Firefox developer. It's hard to use for marketing purposes, though
"I've found this bug in Firefox ..."
"Do you run the latest version?"
"I don't know. I'm running the version my distro gives me."
"So which one is it?"
"I don't know. It won't tell me."
"Please update to the latest version."
"Well, I already have the latest version my distro gives me. If this is actually the latest version, I have no idea."
As the article states, the version number will still be available through the help menu. It will just be slightly less noticeable. It used to be in two menu elements, and after this change it will be in one.
That's the problem here. Firefox's ever changing APIs which are always breaking add-ons. The Chrome add-on API is much more limited and as such doesn't need to change as frequently or as drastically. How Firefox thinks they're going to succeed by becoming a crappier version of Chrome is beyond me.
Firefox has Jetpack addons, which are basically like Chrome's - a limited, stable API. Such addons should work even when the browser updates every 6 weeks. However, jetpack is fairly new, so most addons aren't written using it, and that means they can use internal APIs that do change.
In the long term, most addons will probably be Jetpack, with non-Jetpack addons being special cases that actually do need those internal APIs, and the developers of those addons will need to keep up with the latest Firefox version for them. But, meanwhile there is definitely a lot of inconvenience about this.
Why would someone that doesn't like Firefox updating every 6 weeks switch to Chrome - the browser that invented the 6-week update idea?
Both of those browsers update every 6 weeks, and those 6-week updates do change the user interface, functionality, web API support, etc.
If you want a stable browser, your options are IE, Safari and Opera, and not Firefox or Chrome.
However I don't agree with this part of your comment:
I have a feeling that Mozilla lives in its own little bubble where competing on features and speed, setting the standards pace, and "pushing the web platform forward" seems critical. I'm sure it all feels great in Silicon Valley, but not all web development happens in cool startups champing at the bit for the latest canvas 3D gizmo. I'd hazard that there's even more spread across businesses large and small.
Before FF4, everyone here on Slashdot and elsewhere were saying that Firefox was losing users to Chrome because Chrome was faster. And I think that was true to a significant extent - users do notice speed, and it gets a lot of attention by power users who recommend browsers to normal users. So it was a correct choice to focus on speed for FF4, and as a consequence, today people are saying that Firefox's weak side is something other than speed - because the speed issue is largely resolved, we are very competitive there now. So I don't think any "Silicon Valley bubble effect" had an effect on that choice. Anyhow, most of our developers don't live in Silicon Valley ;) (like most open source projects, we are very spread out)