This marks the last Android release with extensions/add-on support. There will not be a 69 release of this edition of Firefox for Android ("Fennec").
The new remake of the mobile browser ("Fenix") does not have extensions support and it is on an indefinite and easily ignorable backlog. It is explicitly not being worked on this quarter. At the end of this quarter is the scheduled release of Firefox 69. Unless there's no release at all of Firefox 69 upon the Android platform, that's the time when Mozilla I think will deceptively try to drop support for extensions/addons on android.
Bug to have Add-ons pages declare Fenix extensions as unsupported.
"version <= 68.0.98 is legacy Fennec
version >= 68.0.99 is Fenix (or some other GeckoView-powered browser)"
https://bugzilla.mozilla.org/show_bug.cgi?id=1548428
Hi. I work for Mozilla. Firefox for Android is not dead. You are free to continue using it. It will continue to support add-ons, and it will receive regular security updates and bugfixes for at least the next year.
We are focusing all new feature development in Firefox Preview, but we're continuing to support Firefox for Android as an Extended Support Release while we build Preview.
And, for what it's worth, we do understand how important extensions are to folks, and we are taking that into account in our planning.
To be honest that is one of the most Fortune 500 style responses I have ever read coming out of someone involved in Mozilla. I mean what does 'we are taking that into account in our planning." actually mean?
I was about to say that this was one of the most Mozilla-style responses I've read in a while. But I agree with you too. What a sad state of affairs for an organization I once trusted.
It's corporate speak for "we decided to pull the plug but don't want to admit to it until we can get enough people using the new browser who will hopefully like it enough to stay".
I can't make any specific commitment, but there's a lot of time between now and Firefox for Android going EOL (and we haven't committed to specific timing for that, either.)
However, we do have concrete plans for supporting adblocking as a built-in feature in the near future (https://github.com/mozilla-mobile/fenix/issues/96), which should make Preview more palatable to folks who rely on add-ons for that, even if only as a stopgap.
More fundamentally, when considering major features like add-on support, we want to take the time to ensure that we're getting our design right at all relevant levels: Firefox Preview, the underlying GeckoView library, and the Gecko engine itself. And though development on Preview moves quickly, there's still a lot of design and development work yet to be done.
> However, we do have concrete plans for supporting adblocking as a built-in feature in the near future
reading this almost makes me cry because that is the same bullshit google is telling us with chrome.
Firefox is not supposed to block ads! Firefox only has to provide an interface for addons so they can do whatever(!!!) they want (for example blocking ads or installing malware on your system).
Please don't sacrifice our freedom for a wrong understanding of security (like that addon-signing you pulled lately).
I can't agree with this more. I have already moved off of Firefox on the desktop. If this is the destination of Firefox mobile, there is absolutely no use case for it.
Most users will only use an ad blocker that is built-in. You should consider how elitist it is to confine ad blockers to users who know and trust certain extensions vetted by their community, but not the remaining extensions which have been a minefield.
Having a built-in ad blocker doesn't hinder an app from also accepting extensions which block ads.
Originally I wasn't even concerned about an adblocker being built-in. I
was just afraid that it will be used as an argument against more
powerful extensions.
But you brought up another point: I think it is in the nature of an
adblocker, that it MUST be "elitist" to be actually useful. If every
browser does "adblocking" by default, the ads will just adapt. That
will be as useless as the HTTP-Header "dont-track-me".
Good, let ads adapt. Then the adblockers adapt once more. And so on, and so forth. If ads were delivered by the same source/domain (ie. it wasn't outsoured to an ad network who also do profiling) people would have less issues with them.
Not wanting my workflow broken and replaced with inferior software is "elitist"? I don't think so. I hope your attitude isn't prevalent inside of Mozilla. You can say that breaking existing extensions isn't necessary, yet that's exactly what has already happened once and seems like it will happen again.
Yeah, a lot of it is elitist. In other parts of this thread people are complaining about Vimperator being broken by the switch to WebExtensions.
The switch to WebExtensions allowed all the excellent performance improvements we have been seeing in Firefox over the last year. So it improved the performance for everyone, and broke a feature that a small minority used. That minority is a highly specialized, highly trained, bunch.
If you think that features that benefit a small elite group of users should trump improvements for all, then I think that is an elitist attitude.
Also, here there is absolutely no talk about changing extension APIs. So I don't see why extensions that work now shouldn't continue working in FF Preview once Extension support is added. The situation is pretty unlike the WebExtension situation.
So yeah, hold their feet to the fire to make sure that things don't regress without need. But also look at what they are trying to do with merging the FF Focus stuff in, and getting a built-in adblocker. They are doubling down on privacy for all. This should be commended.
People are not less deserving of stable software or good UX by virtue of being highly specialized and highly trained. These users, as much as anybody, have every right to be upset when their software experience is degraded, so I reject your premise unequivocally. If opinions like yours are prevalent inside Mozilla, then the organization has gone cancerous and I fear it's days of producing praise-worthy software are drawing to a close.
Consider the following: over the years I have convinced a dozen or two users in non-technical careers to switch to firefox. Most of them do not install extensions, at least for themselves. However, without extensions, I would not have recommended firefox to them in the first place. Firefox's rise in popularity is owed to the word of mouth campaign carried out primarily by people with technical/foss careers or inclinations. Why would I, or anybody, suggest a browser that I do not enjoy using? To suggest software to somebody is to go out on a limb for that software, because if the person you are suggesting it to has a bad experience with that software, your personal reputation will take the hit for recommending it. So why on earth would I stick my neck out to promote Firefox when Firefox is no longer software that cares about the needs of users like myself? If Firefox has adopted a "socially progressive" model of disregarding the needs of technical users because such users have more software-privilege, then they have taken the wind out of their own sails[sales].
(I put the words 'social progress' in scare-quotes because I also reject the premise that infantilizing software is in fact socially progressive. In truth it's socially regressive, since it's effectively technical people deliberately reducing the exposure non-technical users have to technical problems that might challenge them to learn more. The true impact of infantilizing consumer software is to pull the ladder up behind us. Reinforcing and widening the dichotomy between technical and non-technical users is inherently socially regressive.)
Extensions can be a good thing for sure. But they need to be done right. Hacking out some hurried implementation of extensions is how malware and security advisories happen. It is elitist to suggest that the desires of the 3% trump building a good browser first.
Also, the Internet and most technology happens to break existing workflows. We shouldn't assume disruption is bad if the result is an improved version.
> However, we do have concrete plans for supporting adblocking as a built-in feature in the near future
I also work on an open-source project, and I understand where you're coming from WRT decisions that are unpopular on the outside, but make sense when viewed from the inside. But as a user of Firefox, having already gone through one round of breaking extensions I quite liked with no apparent benefit to me as a user, I'm really not enthused about going through another round of this. Adding a built-in ad-blocker is fine, I guess, but I also use NoScript on mobile. Is that going to die when Firefox for Android goes EOL? What benefit am I going to gain by giving up another one of my core add-ons?
I know you can't answer that here and now, but please, seriously consider how this looks from the user's perspective. I love Firefox and I would like to continue to do so.
I think you need to make it clear to your colleagues that you need to strongly reconsider any possibly that you EOL browser a before browser b supports plugins (specifically ublock origin and noscript) as a lot of your core users will simply not use a browser without them. They're not a "nice to have" - they're what they need to even consider using the browser. Otherwise they might as well be using chrome, or one of any number of other browsers.
>However, we do have concrete plans for supporting adblocking as a built-in feature in the near future
I use Firefox because I enjoy the freedom it grants users do modify it as they see fit. If the play here is to turn adblocking into a walled garden feature so you can monetize it either on the user or advertiser site I might as well go back to Chrome.
Not exactly. I understand "EOL" to mean that all support has ended. An EOL'd project is dead, and if anything breaks no one is going to fix it for you. That is not the case for Firefox for Android.
Instead, think of Firefox for Android as a Long-Term Support (LTS) release, like you might see with projects like Ubuntu, Node, or Django.
There are no new features coming to Firefox for Android, but we will keep the 68 branch alive and maintained to ensure that everything keeps working and is secure on Android. We will do this for at least another year, to give Firefox Preview more time to develop and mature.
On desktop, we do something similar with Firefox's Extended Support Release ("ESR") versions (https://www.mozilla.org/en-US/firefox/organizations/). Every year, we pick a version of Firefox that we commit to supporting for a very long time. Last year, it was Firefox 60. So even though Firefox 68 is available, we're still backporting patches and fixing bugs in Firefox 60. This year, Firefox 68 is the ESR version, so we'll support until at least late next year.
One year is a very short time. The fact is that we can't rely on having support to all our add-ons and browser experience in the near future (next 2-3 years) means we have to start thinking about what to switch to. Also, I wouldn't feel comfortable recommending firefox anymore for that same reason. When the time comes to decide whether to switch to fenix or chrome, fenix will need to really impress. It's a bold move from mozilla.
Anyway, I use firefox on Android because of the ad blocker and to avoid AMP. If the new firefox works similarly in that regard, then I might end up switching to that instead.
Thank you (and the other Mozillians) for trying to engage the feedback! Please remember when planning that extensions need time to be fixed, which means there must be significant overlap between when extension support in Fenix shows up and when Fennec actually goes EOL.
Unfortunately, Firefox for Android _beta_ is dead (it makes no sense to do ESR for beta), and the profiles are not shared, so I'll be stuck with an old Firefox.
I haven’t been able to customize keyboard shortcuts[1] on desktop since 2016, which extensions allowed then. Doesn’t look like a priority to me, or that you understand the importance.
[1] I mean in a fully general sense, not “oh it’ll take effect after each tab fully loads”.
If you can't have plugins on the browser (for Android) then you might as well use chrome (which is not planned for EOL and is faster and secure). My "grrr I hate google" tendencies are satisfied by not logging into any google stuff on chrome and using duckduckgo whilst on chrome. I accept that chrome is - privacy issues notwithstanding - faster and better (supported) than firefox, and I put up with firefox because of the addons, but without addons it's just a slower chrome.
This sounds a bit like FUD since Firefox on Android is faster than Chrome for me. Your wording also implies it's insecure, which I don't think is a correct evaluation.
That said, I do find Mozilla's behaviour on this quite saddening.
Wait what? Extensions are the biggest reason I'm using Firefox on my mobile and the only way I got people to switch to it. How can they just be removing them for any period of time?
Extensions is the ONLY reason to use Firefox on Android. It's not good browser compared to Chrome on slower Android phones, apart from being customizable.
To me it's even the single biggest reason to use Android over iOS as there's no full featured browser on iOS.
> Extensions is the ONLY reason to use Firefox on Android.
Well, no. I don't want to use a browser that is developed by an advertising company, and I want to encourage the development of more than one web rendering engine. But yes, killing extensions would be a major regression in Firefox's usability.
The "Linux" kernel even in something like LineageOS is heavily forked and in practically every case relies on userspace binary blobs for critically-important functionality. It's a far cry from what you can run on PC's, although postmarketOS is moving towards that goal. BTW it's not like Android itself is to blame for this, since other varieties of embedded Linux (by and large) are the same deal; the buck stops with SoC- and embedded-hardware manufacturers.
I've given up on custom ROMs. There's always only an extremely limited set of devices that are supported, and those devices are EOL rather too quickly, forcing us to use custom builds from someone not directly affiliated with the project.
To be clear, not one of the Android devices I've purchased have ever had official support. I'm using a Samsung a8 at the moment.
I hope the custom rom community finds a better mechanism for providing better device coverage.
Custom ROMs are rarely developed for Samsung devices these days because Samsung locks the bootloader and makes everything extremely difficult. Buy a device that is more developer-friendly like a Nexus, Pixel or OnePlus and you'll see a much larger selection of custom ROMs.
I'm personally a big fan of iOS, but coldpie's (and others in this thread) needs seem to be in third party browser capability that doesn't exist in iOS so with that, I wouldn't recommend it to them. I don't think it is a matter of spending money on competitor's devices so much as competitor's devices don't meet their needs.
iOS doesn't allow anywhere near the level of customization that Android does, nor does it allow 3rd party browser engines at all, however there's Sailfish OS and maybe the Librem 5 soon.
True, me neither. But politics aside, it's the only technical reason, for me. I'd rather switch to iOS and Safari than Android and Chrome, which I will consider.
To be fair, the new Firefox Preview (aka fenix) seems to go a long way towards changing that. I agree that they damn well better release with extension support (and I think they will), but it's pretty reasonable to get the core browser stable first.
Give Kiwi Browser a shot on Android. It's a fork of Chrome with all the google tracking removed and fully supports Chrome Add-ons. I love it and it's fast (I was a long time Firefox on Android user but it just got to slow to use).
>No. The browser is not fully open source like open in open source. It is a common myth and misconception about Kiwi Browser.
>Damn, 1 commit and nothing else but keeps updating and releasing apks? Looked into the issues and scarcely seen replies.
>I think this repo is just to show that "hey, I provided the so-called source code (even though not fully open-source as said above) so trust me and install me" thing.
The way you share quotes sounds like you personally have a problem with the browser :D
Kiwi was started as a fun+independent project, then more and more people asked access to some source to be able to develop their own hacks (new tab page, extensions, bottom toolbar, import/export bookmarks, AMP, etc) and they are shared when the devs ask politely.
At the end of the day, if you use Samsung, Kiwi, Fennec or Yandex, what matters is to understand why the developers are doing their project, what do they get from it (not necessarily money), and what are the influences around. The source-code is one indicator, and, unless you actually have reproducible builds (and only Fennec has them from all the mentioned browsers) you have to use your intuition.
Well, that's what I say, Fennec is your only option. For now, there are no Chromium-based browsers on F-Droid (except the WebView-based browsers, which ironically, runs on proprietary software as well in the background).
If you find one truly open-source and reproducible Chromium browser, I'm intrigued, but as far as I know, they all pretend to be on F-Droid (= fake external F-Droid repo) but are not done by F-Droid team (which is sad).
The closest I found was https://gitlab.com/thermatk/Unobtainium but still depends on proprietary code as far as I know.
Yeah, no kidding. Seems pretty significant, like pretty much a non-negotiable reason to switch to another browser if there is one with extension support.
Hopefully the ESR version of Firefox will support add-ons on Linux/Android and will be supported (security fixes, I don't care about other changes) until any replacement browser supports add-ons on Linux/Android.
I've honestly looked, and while there are several versions of Firefox in the play store. None of those I've tried seem to support extensions. Can you tell me where to find one that does?
Sorry but this seems conflicting with what callahad replied above
https://news.ycombinator.com/item?id=20394503 when answering "Will Preview be guaranteed to support extensions before Firefox for Android goes EOL?":
> I can't make any specific commitment, but there's a lot of time between now and Firefox for Android going EOL (and we haven't committed to specific timing for that, either.)
> However, we do have concrete plans for supporting adblocking as a built-in feature in the near future (https://github.com/mozilla-mobile/fenix/issues/96), which should make Preview more palatable to folks who rely on add-ons for that, even if only as a stopgap.
> More fundamentally, when considering major features like add-on support, we want to take the time to ensure that we're getting our design right at all relevant levels: Firefox Preview, the underlying GeckoView library, and the Gecko engine itself. And though development on Preview moves quickly, there's still a lot of design and development work yet to be done.
My understanding is that there's a good chance that Firefox 68, which supports extensions, will reach EOL before 69+ gets support for extensions.
I am reading this differently: it seems likely that Fenix supports extension before Firefox 68 goes EOL, but they don't want to make any announcement yet (maybe, because they don't really know what is going to happen).
I agree with you however. We don't have any guarantee at this point.
Please, Mozilla: don't EOL Fennec before supporting extensions on Fenix. This would be a huge mistake. Extensions are useful beyond adblocking. Chrome not supporting extensions on mobile is not a good reason to consider that Fenix would be good enough without support of extensions. Chrome is not good enough.
If Fenix is not ready early enough, please consider releasing a version 69 of Fennec and then later versions until Fenix supports extensions.
> Linked bugs about the user agent seem irrelevant to this issue.
They're relevant because that's what suggests that they aren't planning for a Fennec 69: addons.mozilla.org is currently interpreting version 69 to mean Fenix, so if Fennec gets updated to 69 they'll have broken extension support anyways. If Mozilla is abandoning Fennec at version 68 before Fenix is even close to having extension support, they probably can't be trusted to not push out Fenix to users prior to implementing extension support. Their messy transition to WebExtensions certainly provides precedent for Mozilla making this kind of fuck-up.
Fennec is moving to an Extended Support Release model so we can focus on Preview. This means that the engine will remain at major version 68 for the next year, but it will still receive security updates and bugfixes. And it will continue to support add-ons.
Why hasn't Mozilla put out a clear statement that Fennec is EOL'd and that the replacement definitely won't be pushed out until it's ready (including support for extensions)? It would go a long way with users if Mozilla could just clearly state that Firefox for Android will be skipping versions 69 through at least ~73 due to Fenix not being ready for public consumption. Stating that extension support is not part of the MVP and failing to set a clear deadline for getting that feature back is not reassuring.
Honestly, it seems a bit weird to say that, on the one hand, you've made a statement to mostly that effect but then in another comment say that Mozilla can't commit to anything.
Plainly speaking, is it a possibility that Firefox on Android will lose extension support for some time in the future?
Support for add-ons will not be removed from Firefox for Android ("Fennec"). However, Fennec will eventually get discontinued (but not in the next year), because its code is honestly a mess.
So the question is whether or not Fennec's replacement will support add-ons when Fennec finally hits End of Life. And I know this sucks as an answer, but we're not yet ready to make public commitments on that front. That's not because we have bad news hidden up our sleeves, it's just... we're still working and designing and figuring this stuff out.
If you want to know more about the architectural mess we got ourselves into, and how we're extricating ourselves from it, I highly recommend reading through this internal presentation on the topic: https://docs.google.com/presentation/d/1MzU9q2wCwojC0kb1eVfm...
Thank you for the additional info. What this tells me is that Mozilla does not see add-ons for Firefox as essential as I do, which is indeed troubling for me.
I would like to point out you've not mentioned the state of Preview aka Fenix at all, only that the soon-to-be-ESR-only Fennec will continue to support add-ons.
uBlock Origin is the only reason I use Firefox on Android.
Mozilla broke extensions on desktop Firefox before and lost many users. Can't help but wonder what the hell are their product managers thinking? How can I recommend Firefox if they keep breaking it with little regard for users?
Is there another browser that supports uBlock Origin for Android? I'm not talking about biased half baked built-in ad blockers, I want the full open source extension.
I was skeptical of the move to Firefox Quantum originally because of the extension-breaking, but after it happened, I can see the motivation: after upgrading, Firefox got a lot faster and less prone to memory leaks for me, at least on macOS. As I understand it, the performance improvements in the new Quantum engine were incompatible with the way the old extension API worked, so a new one needed to be introduced. To me the tradeoff was worth it, because the increasing sluggishness of FF compared to Chrome, pre-Quantum, was making it harder to stick with.
I don't think anyone denies the performance improvements and other advantages of the Quantum project. The claim is that the transition wasn't handled well, as replacement APIs for the deprecated addon system weren't in place until several releases after Quantum debuted, and several of the more deep-reaching abilities addons had were deprecated completely. This broke many people's workflows, destroying trust.
This also happened a short time after Mozilla previously broke addons with the e10s changes, requiring many (most?) of them to be rewritten.
Basically, people don't trust Mozilla to handle transitions well any more, even when they happen to be worthwhile transitions.
This was my exact experience as well. I had to kill/restart FF every few days because it would randomly start eating CPU (this was on Win 10). Because of this I used Chrome until they disabled back navigation via backspace and forced alt-left instead. This pissed me off enough to try FF again, which was right around when Quantum came out. I've never looked back.
Did they "break" - ie some bug caused a problem - or did they change the framework add-ons used after giving months of notice? Most users use no add-ons, and wouldn't notice the change, and many of those who do were aware of this ahead of time even if the add-on developers were slow to wake up to the reality.
Exactly, months. Likes that's enough time to completely rewrite your extension using an half-baked replacement.
To name a few, Cookie-Autodelete [1] had to wait 7 months to get an API to clear the LocalStorage; noscript [2] lost tons of functionality and will probably never recover;
> Did they "break" - ie some bug caused a problem - or did they change the framework add-ons used after giving months of notice?
It's not clear what you're asking. Mozilla changed the add-on API, and the new API does not have all the features that the old API had. Two add-ons that I liked very much died during the switch.
Do you have the data for this? I know I live in a techie bubble, but it sure doesn't fit with my experience or the experiences of others I know who talk about it.
For browsers this seems like a bit too long without updates. I can imagine maintaining such project is demanding but hopefully the project can gain more attention now with the new version of Firefox breaking extensions on Android.
Last I checked Kiwi is not open source. There are bits and pieces of code lying around in the Github repo, but not the full source code.
The browser seems to get regular enough updates on play store, has decent ad blocking (at least good enough for my purposes) and to date has the best night mode out of all android browsers I've tried.
No white flashes like I've faced in Firefox for Android. The dark mode applies to the whole of the browser UI.
I use Bromite, Firefox for Android or Fenix mostly equally during the day and Kiwi at night.
Though the fact that it is not open source may put some people off.
> We're certainly aware of how significant ad blocking extensions are. This release required a great quantity of features with only a six month timeline until now.
> We already support a very limited set of the WebExtensions API to offer features like Reader Mode. Rest assured that more features will land in the coming months.
I originally commented there and expressed how they were underestimating the importance of ad blocking. One of the issues I pointed to as evidence (https://github.com/mozilla-mobile/fenix/issues/96) has since been re-prioritized to Q3.
More "features will land", not necessarily extensions.
They have limited internal extensions API because several of their features were implemented with bundled extensions, but it doesn't talk about extensions/addons themselves.
I don't understand why this comment is currently ranked so highly. It is based on pure conjecture, goes against what Mozilla has publicly stated, and assumes Mozilla is working in bad faith to deceptively drop add-on support.
Firefox 68 is a long-term support release. Even if Fenix exits preview without extension support, nothing stops you from continuing to run Firefox Android 68.
This is a lot of FUD. Making a browser is hard, and making a modular browser is harder. It's clearly on their list of things to do, but it's behind all the other "making a browser" things.
They've made no commitment to extensions, but clearly are targeting this new replacement browser for version 69 in a few months.
What they haven't done is given a clear statement. Nothing about if their replacement is going to drop extension support. But every response from Mozilla has been a sidestep and that I cannot overlook.
> Nothing about if their replacement is going to drop extension support.
That's because we are not yet ready to make public commitments. Honest. We're building the app and its underlying embedding library (GeckoView) from scratch, and as with all new greenfield projects, we're still working on getting the foundation right. It's simply too early to forecast how that specific design and development work will play out.
>> Nothing about if their replacement is going to drop extension support.
>That's because we are not yet ready to make public commitments. Honest.
Surely you can understand how this isn't good enough for most addon users, right? You've already announced Firefox for Android is going to be deprecated. And you're "not ready to make public commitments" about whether the replacement is going to support extensions?? That's obviously not acceptable to addon users.
If Mozilla's policy is that they simply don't care what users think, then whatever, it doesn't matter what's acceptable to addon users. I'm writing this under the presumption that at least someone in the Mozilla org does care about this.
The fact that you won't commit to saying your core advantage on Android will stay has to mean to me, as a user, that I need to be finding an alternative that does have it.
Fenix a.k.a Firefox Preview will not be Fennec 69. There are a lot of decisions the project hasn't made yet, but that one's official. It's an easy decision, too, given that Fenix won't have full feature-parity in three months and maintaining backwards compatibility to super old app versions is not worth it. The expectation is that users will want to migrate to the new store entry in the coming year or two.
As Ricky would say, "It doesn't take rocket appliances."
I don't understand Mozilla's position regarding extensions. First, they released a new desktop version with way less extension support, gimping the largest reason I use FF for - vimperator and it's successors.
Now they are going to kill extensions in Android FF, which will kill the only non-ideological reason my family is using it - for the adblocking extensions.
> First, they released a new desktop version with way less extension support,
Switching the webextensions was necessary to continue improving browser performance. It's my perception that they tried to accommodate extension authors with new apis to allow as many extensions as possible to be ported.
> Now they are going to kill extensions in Android FF, which will kill the only non-ideological reason my family is using it - for the adblocking extensions.
I'm guess enough ad-based companies with deep pockets finally bullied Mozilla hard enough.
Looking at Mozilla IRC at the time, the reason wasn't bribery, but something way more basic.
Did you ever support an old SOAP server made in '90, that had a lot of bugs, that was slow as hell made of molasses, and was vulnerable to several security exploits?
Now imagine you have millions of users demanding, speed, security and those XML add-ons , I mean servers.
Basically Mozilla decided to ditching the old system in favor of the newer one, was the best choice at the time.
To me the writing has been on the wall ever since the community said "Hey these webextensions can't do nearly as many things as the old ones" and Mozilla's response to that was "Go fuck yourselves."
So now I have to do what I do on my desktop: Chromium for casual browsing and an old non-crippled version of Firefox for getting things done. Thankfully I only buy phones I can root and block ads system-wide.
From Firefox's dropping extension support to Windows' removal of granular update controls the tech industry's inexorable slide away from user control makes me want to cry tears of impotent rage.
> Mozilla's response to that was "Go fuck yourselves."
Do you have a source for that or are you just making things up?
My observation was that Mozilla consistently responded to complaints from extension authors by implemented new apis for their webextension implementation to allow extensions to be ported. That's pretty far away from "Go Fuck Yourselves."
> Mozilla introduced many new WebExtensions APIs in Firefox 57 [Quantum], such as the openerTabId property for tabs. The opener information is highly relevant for Tree Style Tab. [0]
Which is handy but not relevant to my question of "why can't an extension do this" with the implied "so I don't have to fiddle with copy-pasting text off the internet into configuration files I'm not entirely familiar with using." Not to mention that by default Firefox 69 disables loading of userchrome.css and you can bet your britches they're doing that in future anticipation of removing the feature entirely just like they did with everything else they've stripped out.
You can enable/disable webrender in about:config if you want to try it out (gfx.webrender.all). I've switched to it on my laptop (HD Graphics 620, Gentoo Linux). So far so good.
It's a shame it hasn't made it over to Mac yet - graphics card options are a lot more limited there so you'd figure it would be easier to support. But I suppose it's a much smaller market share.
There is a known severe performance issue on MacBooks with retina screens and integrated Intel graphics which this is suppose to be the solution to. Unfortunately this doesn't seem to have driven macOS to be a priority for the rollout.
macOS never seems to be a priority at Mozilla, which kind of sucks. Another power management issue is video decoding. Even when I force YouTube to play .mp4 instead of VP9, my battery use is still 2-3x higher on Firefox than Safari, despite H264 acceleration being easily available via VideoToolBox, and VideoToolBox has been available since 2011 (Lion) or 2012 (Mountain Lion).
I also love the new Firefox Preview on Android, which I believe is the first time that it uses the "Quantum" improvements. I can't really decide if I like the speed or the new design more, though. I like that the private mode can be permanently enabled/disabled with a tap and how tabs are now presented in it. Still looking forward for extensions to be supported.
I love Firefox. First thing I install on any new computer. Super happy they keep pushing the web forward. Not a fan of the chromium quasi monopoly (not that the folks behind it are bad, just that any monoculture is problematic).
The whole chromium project gives out the 'death star' vibe for me. I'm sure the engineers who are working on it are good people but the end result seem lead to destruction of open web as we know it by the Gempire.
You can actually grab non-google Chromium. This site keeps track of the latest stable and nightly builds and how to download them directly: https://chromium.woolyss.com/ You won't receive automatic updates, or be able to login to your google chrome account, etc. The theme is slightly different in color but it works exactly like normal Chrome. Chromium is the open source project that Chrome is based on, originally it was a web browser built by the KDE team for Linux.
I disagree that Chromium is evil. Personally, I use Firefox because I like it better. Though, on Android I use Brave which is based on Chromium and has ad blocking, privacy, and anti-fingerprinting built-in. Just a really quick easy way to get all of that in a simple no fuss package.
Interesting thing I noticed recently. I have weaned off of Google in all but calendar and Android and when I initially weaned off, my phone began pinging me to use Assistant more often. Could have been a coincidence, but I don't believe so.
Making a browser that breaks extensions used by millions of people (Vimperator comes to mind) and making extension development harder is not pushing the web forward.
Vimium-FF, Tridactyl, and Vim Vixen adapted. It's true the APIs that allow this kind of thing lagged behind for a while, but there was enough interest to eventually develop those APIs.
FWIW, there are quite a few things that haven't been added to the API, most importantly the ability to accept key events from pages where our content script isn't running.
In fact, very little has been added that has made our life easier (the main one being the search API) since we first started developing Tridactyl. Features have actually been taken away from us - notably the ability to operate on PDFs.
Not to detract from your point, though - most of the API was already there and fine in September 2017.
>FWIW, there are quite a few things that haven't been added to the API, most importantly the ability to accept key events from pages where our content script isn't running.
Speaking as an end-user, I find the eagerness with which web developers re-purpose key events to be extremely aggravating. If I hit ctrl-F, I want to use Firefox's native full page text search, not jump to the in-site search box.
I'd like to see Mozilla start treating key event capture as an opt-in-required permission like site notifications, location, camera, and microphone. Currently the only way to blanket-deny permission to capture key events is by disabling javascript. That's not ideal.
We have a thing in Tridactyl for this, actually ("blacklistkeys" if you're a user).
You roll your own by writing a userscript (e.g. Tampermonkey) that adds an event listener to keydown which runs key.cancelBubble = true and key.stopPropagation(). See line 343 https://github.com/tridactyl/tridactyl/blob/master/src/conte... (starts "leavegithubalone").
Vimperator was used by millions? I think it had about 10.000 users total.
Reality is, most popular add ons got ported very quickly, and the subsequent speed and reliability improvements for Firefox that would not have been possible while keeping the old extensions around have more than validated their decision.
The Recommended Extensions program will go live on July 11. Extension updates are reviewed by Mozilla employees before they are published.
Two of my extensions are also participating, reviewers took special care in how user data is handled and how accurate the privacy policy is. Overall the program seems to be a positive development that will give us a curated list of safe browser extensions to use.
I hope their extension review process has improved over the past 2 years.
I used to have an extension live after it being reviewed, sent them an update a few weeks later, and they took down the old version without warning because the original reviewer didn't do something correctly (from what I can remember). I then had to wait a week before they could review the update and put it back live again by which point, a large portion of the userbase had disappeared and all our marketing efforts gone to waste for an entire week.
Never again would I deal with Mozillas reviewing processes.
This bit me. I store my bookmarks as a file under revision control. I have a "bookmarks.js" file, and which is loaded via a local index.html file. There is some javascript magic for filtering, searching etc.
As of today's upgrade opening file:///home/skx/bookmarks.html no longer loads the JS/data. Breaking the system:
For what it's worth, I'm still able to open and use Doxygen HTML docs without running a web server. Looks like it loads both CSS and JS. So I'm not sure what the changenote actually means.
That makes sense, because so much security nowadays depends on keeping origins separate. It's hard to tell whether any given file:// URL belongs in the same origin as another file:// URL. Better treat each file:// URL as its own origin.
CSS/JS imports in the HTML are ignored. Hyperlinks are links not file loads in the page. It means you can't load e.g. access a file via script (e.g. a JSON file in the directory, directly at least).
if it's the same as chrome the document would be able to load resources trough link and script, only xhr would not work, in case of a simple page statically loaded it should keep working but stuff like phaser.js would not be able to load assets
Currently blocked by Microsoft Defender SmartScreen, probably because the temporary (sub)domains were used as malware domains at some point.
Either way, exposing local files to an internet proxy just to open them in a browser is overkill. Lots of script runtimes come with their own lightweight HTTP servers nowadays. For example, I use Python's with "py -m http.server".
> New reporting feature in about:addons allows you to report security and performance issues with extensions and themes.
> Redesigned extensions dashboard in about:addons provides easy access to information about your extensions, including data and settings access required by each extension.
Find high quality, secure extensions via the Recommended Extensions program in about:addons, which now displays user count and ratings for each extension.
> "Recommended” badges for these extensions also appear on AMO. More extensions will be added over time.
I welcome the new changes to the extension ecosystem. For too long extensions were unsupervised and malicious code was allowed to run (remember the Stylish fiasco[0] where your browsing history was siphoned off?).
App stores and extension ecosystems need to be policed with a lot more rigour and code needs to be inspected so that the extension does what it says in the description and nothing more. No ulterior motives. No 'monetizing' of user data, and no surreptitious phoning home to a command and control server with your browsing history.
Does the new system really stop those shenanigans though? I see notifications for what major areas an addon requires. I don't get notifications if an addon tries to reach out to the internet. Has anyone seen such a notification?
Mozilla do actually do a lot of manual code review, especially for new addons. That was one issue with the big extension migration; it was taking forever for new style addons to be approved.
But more than anything... BigInt! Finally! JavaScript has real integers on Chrome and Firefox now! (The TC39 proposal is at Stage 3: https://github.com/tc39/proposal-bigint)
We should go back to system color (usually medium grey) as the default background for webpages. Not sure why more recent browsers have replaced it w/ white, this makes for pretty poor usability/accessibility.
Similarly, I wish the app itself enabled Dark Mode on Mojave. File save/etc dialogs are awfully white in a darker environment.
(More broadly, I wish Apple would allow things like file dialogs to force to dark mode. I know you can set NSRequiresAquaSystemAppearance to NO, but it tends to mess up a bunch of other random stuff too)
1) Videostreams are way smoother using Chrome, at least on Linux.
2) I've enabled Emacs Input in Gnome. When I enter something in the address bar, both Firefox and Chrome show suggestions for websites below. In Chrome I can type Ctrl+n / Ctrl+p to navigate between the suggestions, because it respects my keyboard schema.
In Firefox this opens new browser windows :( and I've found absolutely no way to change this behaviour.
No, I'm using gnome-shell which uses mutter as wm.
But using gnome tweaks you can enable Emacs Input, which if you are used to emacs key bindings, is really handy navigating in gnome-terminal or, as mentioned web browser address bars.
1) Regarding WebRender: Maybe I'm a bit out of the loop, but aren't they working on Servo? It seems, WebRender is a Gecko thing. I'm confused.
2) I'm also confused about the state of Gecko Embedding. As far as I know many are jumping to Chromium instead of Gecko. What's the plan there? Is Chromium better/easier to embed? Will I be able to embed Servo at some point?
Servo is an experimental browser project to explore ideas/techniques: it's never been intended to be a replacement for/embedded into Firefox.
What was planned and is happening is that a lot of the exploratory work happening on Servo is being integrated into Firefox in parallel. WebRender is one example of this work: it's (effectively) a Servo component that is now in Firefox.
Another example of the above is Stylo/QuantumCSS, Servo's CSS component. This post on Stylo/QuantumCSS gives a good early insight into the overall project to bring these components to Firefox https://hacks.mozilla.org/2017/08/inside-a-super-fast-css-en...
> many are jumping to Chromium instead of Gecko. What's the plan there? Is Chromium better/easier to embed? Will I be able to embed Servo at some point?
- Chromium is much easier to embed than Gecko
- Servo can be embedded now (and uses the same/a similar system to Chromium), but as mentioned above, Servo as a whole isn't stable.
- Making Gecko/Firefox easy to embed would be great, but at this stage I think this is largely a matter of focus, resources, priorities.
- On a related note, there is the GeckoView[0] which is a project to build an easily-embeddable Firefox for Android
I've been running Firefox full time on all my computers for a few months now. It's been really good. I've tried to switch away from Chrome in the past, but I've always had to come back due to performance issues or incompatibility with something. But it's basically 95% on par with Chrome now (missing a few things like syncing search engines, etc)
Meaning the search engines that appear at the bottom of the URL bar (respectively the separate search bar in case) and of course also the default search engine that is used when you load something that isn't an URL in the URL bar.
no it's not something is definitely different somewhere, maybe a different config/default or reset, everything is snappier and my cpu is reaching 95+ degC while it never reached 80 before 70.
Or maybe it's a different rendering library / os / gpu driver interaction.
"Cryptomining and fingerprinting protections are added to strict content blocking settings in Privacy & Security preferences."
my first reaction is it's really sad to see this. looks like something Google blackmailed them into adding. maybe Im missing some aspect to this.. but why is mining crypto bad but showing ads is not?
I rather mine coins/ether/whatever for the website owners than see more ads
moving away from the web being funded by advertisement to a web funded by distributed computing is what a lot of people have been looking forward to, no?
Crypto-mining scripts are mostly silent; if your computer doesn’t have a loud fan to spin up, you’re hardly able to even tell one is running other than on its eventual impact on battery life.
As such, they’re a favoured tool of script-kiddies who deface websites. They take over a site, and then drop a crypto-miner onto the site to make themselves money, otherwise leaving everything intact. Sometimes even the site owner doesn’t notice anything has changed for quite a while. Meanwhile, an unaffiliated third party is now making money off of their website on the backs of their users.
The whole adtech industry is built around arbitrary javascript execution for tracking purposes, so we've also seen crypto mining scripts served through ad networks. For example: ads on YouTube mining Monero https://arstechnica.com/information-technology/2018/01/now-e...
This is being used by anybody who can sneak a javaascipt into your page or an iFrame that you embed, which has become too easy to do. When it comes down to it, IMO, site owners are responsible for what ads and scripts they allow to be served via their page, but I think I'm in the minority on that opinion.
Ad networks (the more reputable ones anyway) are playing a cat and mouse game of allowing a Turing complete js environment and trying to prevent its use for doing particular pieces of math. I'm glad Firefox is joining in as well, but I'd be happier if we didn't have this problem to begin with.
If they wanted to show me which websites are impacting my battery life that'd be fantastic! That's a great idea
There seems to be a conflict of interest when a browser singles out one of the biggest threats to the online ad industry while being funded by that same industry
(~90% of Mozilla's money is from Google according to Wikipedia)
While we're on the topic, does anyone have a good guide for adding a crypto miner to a simple static website? (like Githubpages)
The vast majority of javascript crypto mining is not there to displace advertisements. It is added to the website by hackers or rogue sysadmins for unscrupulous reasons.
If I add a certificate on Android (for example for WPA2-Enterprise), do I need to disable security.certerrors.mitm.auto_enable_enterprise_roots if I don't want those with access to the private key to be able to MITM me?
You don't need to, actually. It's one of the few unwalled gardens left on the Internet :). You can ask questions, post answers, and even propose edits to other people's posts without having an account.
Edit: to clarify, I would have answered your question if I knew. I don't, but I think those sites give good answers.
I switched from Chrome after Google moved to disable ad blockers. Do we see any evidence of a broader trend towards Firefox? Or should we be happy in our little corner.
Wish there was a log in Firefox to determine how/why uBlock Origin keeps getting uninstalled across all my machines which are synced with my Firefox account.
I have a generic Intel card on a laptop and forced WebRender to be on via config and wow - it really feel very snappy. Congrats, Firefox team! Fantastic results.
I was absolutely mozilla's target audience when firefox was introduced. Now they remove control and capabilities from us, and I keep getting told here on HN I'm not their target audience, and should accept that they are making this for "most users" who need extension signing, auto updates, sandboxing, and in general NERFING THE POWER from me. Who the hell are they building for if not techies? Who do they think is going to evangelize for them if not techies?
I'll be downvoted to Hell for this, but I don't care. With each subsequent release of Firefox, I'm consistently happy that I've moved to Pale Moon. The nonsense they're adding and the problems that Mozilla has to "fix" seem more often than not to be brought on themselves with the rapid-release and the move away from being Firefox to being "almost Chrome". The move to multi-process (which has introduced memory leaks), the move to WebExtensions (which nuked add-ons that worked reliably since FF2.0), and the general roadmap make me almost want to stop using a web browser.
Mozilla has been pushing anti-features like their survival depends on it. I too have moved to forks. I use icecat and waterfox now. Tired of being abused because they assume too much. Just be a stupid browser and stop being the internet police. You're not making the web better, you're just driving your long time users (we hackers) away, and news flash, no one is there to replace them! Normal users use default tools.
The new remake of the mobile browser ("Fenix") does not have extensions support and it is on an indefinite and easily ignorable backlog. It is explicitly not being worked on this quarter. At the end of this quarter is the scheduled release of Firefox 69. Unless there's no release at all of Firefox 69 upon the Android platform, that's the time when Mozilla I think will deceptively try to drop support for extensions/addons on android.
Bug to have Add-ons pages declare Fenix extensions as unsupported. "version <= 68.0.98 is legacy Fennec version >= 68.0.99 is Fenix (or some other GeckoView-powered browser)" https://bugzilla.mozilla.org/show_bug.cgi?id=1548428
"Or do nothing because Fennec 68 and Fenix 68 will overlap for only 6–8 weeks with a preview audience." https://github.com/mozilla/addons-frontend/issues/7963
Firefox 69 releases 2019-09-03 at the end of Quarter 3. https://wiki.mozilla.org/Release_Management/Calendar
Extensions support are in the "Reserved Backlog" category. Also locked to prevent any more comments about it. https://github.com/mozilla-mobile/fenix/issues/574
Extensions support is not in Q3 backlog https://github.com/mozilla-mobile/fenix/projects/19