There are multiple plausible explanations. From the point of view of a browser vendor, it makes a lot of sense to start with popular plugins that you know work and increase the selection over time...
Particularly given that there appears to be no technical reason why none of the other add-ons are allowed any more, it being merely a policy decision.
There are multiple plausible explanations. From the point of view of a browser vendor, it makes a lot of sense to start with popular plugins that you know work and increase the selection over time as you improve API support, because otherwise users might get a bad experience from broken plugins.
And looking at the announcement, it does sound like they will be opening it up more:
We’re kicking it off with the top 9 add-ons for enhanced privacy and user experience from our Recommended Extensions program. At the same time, we’re continuously working on offering more add-on choice in the future that will seamlessly fit into Firefox for Android.
Worried people will often pick the worst plausible explanation for an action they don’t understand, and it’s possible that they’re right, but but it’s important to remember that this is based on speculation.
There may be multiple plausible explanations, however the "getting a bad experience from potentially broken plugins" one you listed is a policy decision not strictly a technical limitation, and at...
There may be multiple plausible explanations, however the "getting a bad experience from potentially broken plugins" one you listed is a policy decision not strictly a technical limitation, and at least according to @dblohm7 (our resident Mozilla employee) that actually was the reasoning behind preventing unapproved add-ons from being installed. So the author is correct in that it was a policy decision... but at the same time, they are wrong about pretty much everything else that follows after that statement, IMO.
They recently started rolling it out for all users and while it does have extension support only a handful popular extensions can be installed and installing any other extension is not possible.
Just to be clear on why this is happening the way it is: Extension support in GeckoView is being added incrementally, more or less using a top-down approach. Getting uBlock Origin up and running was our topmost priority, so we initially focused on implementing the API surface specifically so that uBO could run. Then we started moving our way down the list of top recommended extensions, adding their required APIs as we went.
So (and I think you understand this, but many others do not) it's not that we don't want people installing non-recommended add-ons, but rather the issue is that the ones on the supported list are the ones that, at this time, we can guarantee have sufficient API support.
And it being a policy decision doesn't automatically mean it's a bad one. Just because something is technically possible doesn't mean it's a good use of time and effort.
And it being a policy decision doesn't automatically mean it's a bad one. Just because something is technically possible doesn't mean it's a good use of time and effort.
Yep, agreed. And from the fenix github it seems like the plan is to allow the nightly build to actually install arbitrary/unsupported addons, which contradicts most of this author's assumptions...
Yep, agreed. And from the fenix github it seems like the plan is to allow the nightly build to actually install arbitrary/unsupported addons, which contradicts most of this author's assumptions about Mozilla's intentions to "kill" addons support. See: https://github.com/mozilla-mobile/fenix/issues/14034#issue-683738881
So if someone really is determined to install them, they still will be able to if they install the nightly, once that feature is added to it.
Well, yes, technical limits can be the result of policy decisions, so it’s not a clear-cut distinction. Teams with limited time and resources have to decide what to support somehow. This is a...
Well, yes, technical limits can be the result of policy decisions, so it’s not a clear-cut distinction.
Teams with limited time and resources have to decide what to support somehow. This is a major problem with rewrites. If it’s a major design change, it’s not going to support everything the old code did right away, or maybe at all.
The question is whether the rewrite is worth it, and people are going to disagree about that. While I was working on GWT, Firefox deprecated support for the way we implemented our debugger, which was a major hassle for us and our users. We invented a different way to do debugging, which was in some ways better (since it’s cross-browser) but was in other ways worse. But if Firefox had frozen things for us then they would be even further behind.
Unfortunately, plugin API’s tend not to be as stable as web apis. You don’t get the same long-term support.
Yeah, I totally understand Mozilla needing/wanting to focus their effort more, and trying to ensure the best experience for the majority of their users. I also agree there is definitely crossover...
Yeah, I totally understand Mozilla needing/wanting to focus their effort more, and trying to ensure the best experience for the majority of their users. I also agree there is definitely crossover between policy/technical issues, and in this case it would probably be most accurate to say it was a policy decision that lead to them intentionally implementing a technical limitation. In any case, I still think the rest of this author's assumptions about Mozilla are likely very wrong; Addons aren't really going anywhere and eventually fenix will probably support most of the old ones.
While I understand the appeal of having as many add ons as your heart desires, the old Firefox for Android sucked. A lot. It was slow and had horrible UX. I really tried to use it as my daily...
While I understand the appeal of having as many add ons as your heart desires, the old Firefox for Android sucked. A lot. It was slow and had horrible UX. I really tried to use it as my daily driver but it was such a painful experience I just ended up back with chrome.
The new Firefox for Android, on the other hand, is actually pleasant to use (I've been using it since it was in beta). I would much rather have a usable, modern mobile Firefox that I can use today over waiting months or years just so super techy people can have all of their extensions. It doesn't seem like a useful expenditure of dev time.
And if you absolutely have to have all of your extensions, what's stopping you from going and installing an old APK of firefox yourself?
When was this? I've been using Firefox for Android for about 4 years now and it's a lot better than Chrome, which was always buggy for me in some way or another.
the old Firefox for Android sucked. A lot. It was slow and had horrible UX. I really tried to use it as my daily driver but it was such a painful experience I just ended up back with chrome.
When was this? I've been using Firefox for Android for about 4 years now and it's a lot better than Chrome, which was always buggy for me in some way or another.
I've been using Fennec for about the same amount of time felt the same way. It think (especially earlier) it could sometimes be a little slower to first render than Chrome was, but not in a way...
I've been using Fennec for about the same amount of time felt the same way. It think (especially earlier) it could sometimes be a little slower to first render than Chrome was, but not in a way that really interfered with the relatively small amount of mobile web browsing I do.
The new one breaks a bunch of UX habits (tabs, frequent sites, etc) and introduced a weird and irritating issue where the URL-bar autocomplete will sometimes eat the first few characters I enter (something to do with the way the suggestions are auto-selected, then the next character replaces all the selected text and it starts over).
All the issues are fixable, but it's annoying having what feels like a distinctly unfinished product replace a very polished and familiar experience.
I used a OnePlus One for five years, and I would always encounter weird problems with Fennec like it would get stuck loading pages if I had too many tabs open at once.
I used a OnePlus One for five years, and I would always encounter weird problems with Fennec like it would get stuck loading pages if I had too many tabs open at once.
The browser wars will continue. Maybe the vast majority will be satisfied with the Chromeopoly, and Firefox may fade to obscurity. But something will fill that void in the future. It might not get...
The browser wars will continue. Maybe the vast majority will be satisfied with the Chromeopoly, and Firefox may fade to obscurity.
But something will fill that void in the future. It might not get much mass appeal, but perhaps that will be what it takes to de-commercialize user-driven web content.
The issue here is as the web continues to get so much more complex creating a new rendering engine isn't easy. Without a big company behind it, it's unlikely there will be another one. If Firefox...
something will fill that void in the future
The issue here is as the web continues to get so much more complex creating a new rendering engine isn't easy. Without a big company behind it, it's unlikely there will be another one. If Firefox dies only WebKit based engines remain.
A browser that focused on modern features is also certainly viable. It doesn't have to implement all the old layers of cruft. I expect most websites would work fine without IndexDB, observe(), and...
A browser that focused on modern features is also certainly viable. It doesn't have to implement all the old layers of cruft. I expect most websites would work fine without IndexDB, observe(), and marquee.
For the sites that do break, fall back to a "legacy" browser that can handle that stuff. Not so different than current mitigation strategies.
Gemini is somewhat interesting for retro and embedded computing, so it’s going to be a very small niche. For a new protocol to become popular, I think it’s going to have to do something new and...
Gemini is somewhat interesting for retro and embedded computing, but I think they went a bit too far by getting rid of encryption, so it’s going to be a very small niche.
For a new protocol to become popular, I think it’s going to have to do something new and different. Simplicity isn’t enough.
What do you mean? Gemini has mandatory TLS, it is one of the major differences from Gopher.
Gemini is somewhat interesting for retro and embedded computing, but I think they went a bit too far by getting rid of encryption, so it’s going to be a very small niche.
What do you mean? Gemini has mandatory TLS, it is one of the major differences from Gopher.
I've only three add-ons installed uBlock, Privacy Badger, and Facebook container. Firefox has inbuilt support for screenshots so I no longer need Fireshot in firefox. It also has...
I've only three add-ons installed uBlock, Privacy Badger, and Facebook container. Firefox has inbuilt support for screenshots so I no longer need Fireshot in firefox. It also has picture-in-picture mode (Don't know if there was an add-on for this but I still use Google's pip extension in Chrome). Installing separate add-on doesn't make sense when the browser has same feature now. However, add-on support should never go away.
I currently have 10 add-ons: 1Password Auto-Sort Bookmarks Enhancer for YouTube Facebook Container Firefox Multi-Accounts Container Ghostery Google Container HTTPS Everywhere Reddit Enhancement...
I currently have 10 add-ons:
1Password
Auto-Sort Bookmarks
Enhancer for YouTube
Facebook Container
Firefox Multi-Accounts Container
Ghostery
Google Container
HTTPS Everywhere
Reddit Enhancement Suite
uBlock Origin
Honestly, I could do without half of them (I don't really need the specific container add-ons, they're more of a convenience) but they make the browsing experience better. A few are available on Chrome, but I'd much rather use a browser that's focused on making all web browsing good and not just making Google's web browsing better.
Firefox has betrayed its core userbase time and time again over the past few years. Despite being a company founded on principles of free software, it included proprietary software by default....
Firefox has betrayed its core userbase time and time again over the past few years. Despite being a company founded on principles of free software, it included proprietary software by default. Despite supposedly fighting the good fight against Google, they accepted a kickback from them. Is it the new CEO? I have no idea, and Firefox has been going down this path for a couple of years now. Remove choice from the user, gaslight the user, rinse and repeat. The most egregious example of this is their telemetry; despite being the browser for privacy minded people, it phoned home more than the Brave browser (https://arstechnica.com/information-technology/2020/03/study-ranks-edges-default-privacy-settings-the-lowest-of-all-major-browsers/). I've lost faith in Mozilla since the recent firing (you can't donate to fund browser development). And the "open web" doesn't mean much if Google controls the web standards that will be implemented on most websites anyways. Seems like avoiding the modern web in general is the best choice nowadays, the browser wars are a complete farce..
Politely asking, what are they supposed to do? Without Google's money they would run out of funding quickly. It's not like you're forced to use Google anyways. And how are you planning to avoid...
Politely asking, what are they supposed to do? Without Google's money they would run out of funding quickly. It's not like you're forced to use Google anyways.
And how are you planning to avoid the modern web? It's unavoidable. To attempt (emphasis on attempt) to stop or even limit using it isn't practical for many people. Banking, work, school, connecting with others, shopping, entertainment... the list goes on. It's not like even 1% of people boycotting the entire web is going to do anything. The rest of the world doesn't care.
The way I see it, my options are to use Firefox and try and support the open web or succumb to Chrome. The other options are just not practical—maybe even selfish. It's like voting 3rd party.
Agreed. And I think Mozilla "betraying its core userbase" is nothing at all. It's the vocal minority that loves to complain, but does nothing to help. You see this a lot in Linux too. This core...
Agreed. And I think Mozilla "betraying its core userbase" is nothing at all. It's the vocal minority that loves to complain, but does nothing to help. You see this a lot in Linux too.
This core userbase can yell at clouds or go to Chrome, because without doing these things Firefox can barely survive. The core userbase will not give any money.
They want some utopic ideal that does not work in this situation.
Technically yes, but Chrome/ium's Blink renderer was forked from webkit if I remember right, so they share the same ancestry even though they've diverged since.
Technically yes, but Chrome/ium's Blink renderer was forked from webkit if I remember right, so they share the same ancestry even though they've diverged since.
There are multiple plausible explanations. From the point of view of a browser vendor, it makes a lot of sense to start with popular plugins that you know work and increase the selection over time as you improve API support, because otherwise users might get a bad experience from broken plugins.
And looking at the announcement, it does sound like they will be opening it up more:
Worried people will often pick the worst plausible explanation for an action they don’t understand, and it’s possible that they’re right, but but it’s important to remember that this is based on speculation.
There may be multiple plausible explanations, however the "getting a bad experience from potentially broken plugins" one you listed is a policy decision not strictly a technical limitation, and at least according to @dblohm7 (our resident Mozilla employee) that actually was the reasoning behind preventing unapproved add-ons from being installed. So the author is correct in that it was a policy decision... but at the same time, they are wrong about pretty much everything else that follows after that statement, IMO.
source: https://tild.es/r7l#comment-5gk0
And it being a policy decision doesn't automatically mean it's a bad one. Just because something is technically possible doesn't mean it's a good use of time and effort.
Yep, agreed. And from the fenix github it seems like the plan is to allow the nightly build to actually install arbitrary/unsupported addons, which contradicts most of this author's assumptions about Mozilla's intentions to "kill" addons support. See: https://github.com/mozilla-mobile/fenix/issues/14034#issue-683738881
So if someone really is determined to install them, they still will be able to if they install the nightly, once that feature is added to it.
Well, yes, technical limits can be the result of policy decisions, so it’s not a clear-cut distinction.
Teams with limited time and resources have to decide what to support somehow. This is a major problem with rewrites. If it’s a major design change, it’s not going to support everything the old code did right away, or maybe at all.
The question is whether the rewrite is worth it, and people are going to disagree about that. While I was working on GWT, Firefox deprecated support for the way we implemented our debugger, which was a major hassle for us and our users. We invented a different way to do debugging, which was in some ways better (since it’s cross-browser) but was in other ways worse. But if Firefox had frozen things for us then they would be even further behind.
Unfortunately, plugin API’s tend not to be as stable as web apis. You don’t get the same long-term support.
Yeah, I totally understand Mozilla needing/wanting to focus their effort more, and trying to ensure the best experience for the majority of their users. I also agree there is definitely crossover between policy/technical issues, and in this case it would probably be most accurate to say it was a policy decision that lead to them intentionally implementing a technical limitation. In any case, I still think the rest of this author's assumptions about Mozilla are likely very wrong; Addons aren't really going anywhere and eventually fenix will probably support most of the old ones.
While I understand the appeal of having as many add ons as your heart desires, the old Firefox for Android sucked. A lot. It was slow and had horrible UX. I really tried to use it as my daily driver but it was such a painful experience I just ended up back with chrome.
The new Firefox for Android, on the other hand, is actually pleasant to use (I've been using it since it was in beta). I would much rather have a usable, modern mobile Firefox that I can use today over waiting months or years just so super techy people can have all of their extensions. It doesn't seem like a useful expenditure of dev time.
And if you absolutely have to have all of your extensions, what's stopping you from going and installing an old APK of firefox yourself?
When was this? I've been using Firefox for Android for about 4 years now and it's a lot better than Chrome, which was always buggy for me in some way or another.
I've been using Fennec for about the same amount of time felt the same way. It think (especially earlier) it could sometimes be a little slower to first render than Chrome was, but not in a way that really interfered with the relatively small amount of mobile web browsing I do.
The new one breaks a bunch of UX habits (tabs, frequent sites, etc) and introduced a weird and irritating issue where the URL-bar autocomplete will sometimes eat the first few characters I enter (something to do with the way the suggestions are auto-selected, then the next character replaces all the selected text and it starts over).
All the issues are fixable, but it's annoying having what feels like a distinctly unfinished product replace a very polished and familiar experience.
I used a OnePlus One for five years, and I would always encounter weird problems with Fennec like it would get stuck loading pages if I had too many tabs open at once.
Probably a year and a half, maybe 2 years ago. I switched to Fenix as soon as it was in a usable state. I will admit that it's entirely anecdotal
The browser wars will continue. Maybe the vast majority will be satisfied with the Chromeopoly, and Firefox may fade to obscurity.
But something will fill that void in the future. It might not get much mass appeal, but perhaps that will be what it takes to de-commercialize user-driven web content.
The issue here is as the web continues to get so much more complex creating a new rendering engine isn't easy. Without a big company behind it, it's unlikely there will be another one. If Firefox dies only WebKit based engines remain.
That's what they said about IE6. It might different this time, it might not.
A browser that focused on modern features is also certainly viable. It doesn't have to implement all the old layers of cruft. I expect most websites would work fine without IndexDB, observe(), and marquee.
For the sites that do break, fall back to a "legacy" browser that can handle that stuff. Not so different than current mitigation strategies.
Time to migrate to the small internet. Gopher and Gemini, here I come.
Gemini is somewhat interesting for retro and embedded computing,
but I think they went a bit too far by getting rid of encryption,so it’s going to be a very small niche.For a new protocol to become popular, I think it’s going to have to do something new and different. Simplicity isn’t enough.
What do you mean? Gemini has mandatory TLS, it is one of the major differences from Gopher.
You are right. Thank you for the correction. I don’t know why I misremembered that.
I don't think this will ever happen.
Now all the apps are basically web pages.
I've only three add-ons installed uBlock, Privacy Badger, and Facebook container. Firefox has inbuilt support for screenshots so I no longer need Fireshot in firefox. It also has picture-in-picture mode (Don't know if there was an add-on for this but I still use Google's pip extension in Chrome). Installing separate add-on doesn't make sense when the browser has same feature now. However, add-on support should never go away.
I currently have 10 add-ons:
Honestly, I could do without half of them (I don't really need the specific container add-ons, they're more of a convenience) but they make the browsing experience better. A few are available on Chrome, but I'd much rather use a browser that's focused on making all web browsing good and not just making Google's web browsing better.
Firefox has betrayed its core userbase time and time again over the past few years. Despite being a company founded on principles of free software, it included proprietary software by default. Despite supposedly fighting the good fight against Google, they accepted a kickback from them. Is it the new CEO? I have no idea, and Firefox has been going down this path for a couple of years now. Remove choice from the user, gaslight the user, rinse and repeat. The most egregious example of this is their telemetry; despite being the browser for privacy minded people, it phoned home more than the Brave browser (https://arstechnica.com/information-technology/2020/03/study-ranks-edges-default-privacy-settings-the-lowest-of-all-major-browsers/). I've lost faith in Mozilla since the recent firing (you can't donate to fund browser development). And the "open web" doesn't mean much if Google controls the web standards that will be implemented on most websites anyways. Seems like avoiding the modern web in general is the best choice nowadays, the browser wars are a complete farce..
Politely asking, what are they supposed to do? Without Google's money they would run out of funding quickly. It's not like you're forced to use Google anyways.
And how are you planning to avoid the modern web? It's unavoidable. To attempt (emphasis on attempt) to stop or even limit using it isn't practical for many people. Banking, work, school, connecting with others, shopping, entertainment... the list goes on. It's not like even 1% of people boycotting the entire web is going to do anything. The rest of the world doesn't care.
The way I see it, my options are to use Firefox and try and support the open web or succumb to Chrome. The other options are just not practical—maybe even selfish. It's like voting 3rd party.
Agreed. And I think Mozilla "betraying its core userbase" is nothing at all. It's the vocal minority that loves to complain, but does nothing to help. You see this a lot in Linux too.
This core userbase can yell at clouds or go to Chrome, because without doing these things Firefox can barely survive. The core userbase will not give any money.
They want some utopic ideal that does not work in this situation.
Well, there's also Apple and their WebKit browser.
Technically yes, but Chrome/ium's Blink renderer was forked from webkit if I remember right, so they share the same ancestry even though they've diverged since.