I didn't see anything about removing any extension support. The word "upgrade" also isn't anywhere on the page that I can see. So my guess is no, you're not reading it right.
Well I've heard that they are moving to a new extension api for the mobile release and it seems like ublock origin is the only thing thats supported for now per this webpage. I use that and umatrix so I'd like to know if uMatrix is going to be supported for mobile in the future... Does the developer for umatrix browse HN?
I don't know that uMatrix will be supported any time soon, but you can turn uBlock Origin into Advanced mode and get much of the same functionality that way.
See point 3, this is a complete redesign so they don't have support for any other extensions yet, compared to the existing Firefox on android. But what irks me is that they will only support "recommended extensions" in the near future with no indication on how to install "unsupported" extensions or how to get "recommended".
This article is about the final release to all Firefox users. It is not about the preview and nightly versions. They are going to break all extensions for all users (except for uBO).
From the article being discussed here <https://blog.mozilla.org/firefox/firefox-android-new-feature... >: "The biggest ever update to Firefox browser for Android is on its way. Later this spring, everyone using the Firefox browser on their Android phones and tablets will get the update. ... This update will initially include support for one of the most popular extensions on Android, uBlock Origin. Additional extensions will be supported in subsequent releases so you can customize and expand your mobile browsing experience even more."
>I don't know what else you can expect? That add ons will never be deprecated?
I thought we went through the whole round of destroying support for non-WebExtensions so that we wouldn't have to do this ever again.
This update will initially include support for one of the most popular extensions..."
Having support for exactly one means you don't support the others. They also say "Additional extensions will be supported in subsequent releases", which implies they aren't supported initially. Also supported by someone's link to
https://blog.mozilla.org/addons/2020/02/11/faq-for-extension..., thanks for that.
It's phrased in a way that avoids saying outright what they are actually doing (dropping support), though.
Yes. Because for the majority of users, who do not use extensions, it is an upgrade. And it comes with built-in tracking prevention and still supports ad blocking, surely the two the big reasons most people use extensions in the first place.
And they're going to restore extension functionality, so the power users who depend on extensions can keep the current version and wait if they wish.
> And they're going to restore extension functionality
I put the odds of that somewhere around the same as the odds that GNOME Screensaver will ever support any of the xscreensaver demos it claimed it would … which is to say, I'm not holding my breath.
> can keep the current version and wait if they wish.
Unfortunately, web browsers are one of the few applications where "keep the current version" is not a viable option. Web browsers are very exposed to every web site they visit (including code injected by ad networks), and once a new version is released, many serious vulnerabilities on the previous version are revealed (for instance, https://www.mozilla.org/en-US/security/known-vulnerabilities... is the list for Firefox). Keeping an old version can be a serious security risk.
God Damnit I Hoped They Wouldn't Do This. I love you Firefox but you make it so hard. I've come to depend on NoScript on mobile (the mobile web is a _nightmare_ with JS enabled by default) and now they're going to break it on me. The worst part is the part of this article about how their browser is not ready for extensions is labeled "Extension-ready". No it fucking isn't.
I've never tried the uBO script blocker, but NoScript's UI is really nice for performing common tasks related to blocking JS. Stuff like allowing temporary permissions per domain, or allowing all on the current tab. Does uBO have similar features?
You can try the brave browser on Android. Disabling and enabling js temporarily is easy. Even better, enabling js on a website in private mode does not persist. So you can disable all js by default, enable it on trusted sites via 1 click easy to use UI, and enable anywhere in private mode without worrying about forgetting to disable it later.
> enable it on trusted sites via 1 click easy to use UI
Does this just enable all scripts loaded by that site, or does it allow filtering by domain the scripts themselves are loaded from?
E.g. when browsing site.com you want to allow site.com (and often sitestatic.com or something), but still not allow doubleclick.net even though it's loaded by the same page.
Yes, in fact ublock origin has a very advanced global/local case where you can block resources globally or block resources locally and temporarily if you really want.
It's generally better to use ublock origin on it's own than to use ublock origin with umatrix or ublock with noscript. Mainly because ublock origin comes wiht neutralising anti-adblock scripts that would get blocked if you combine it with a separate script blocker.
Actually I think you're right, and in fact they've already done this for those of us who used Firefox Nightly on Android. I had both Nightly and Preview on my phone, and kept using Nightly precisely because it supported all my extensions (Cookie AutoDelete, Bypass Paywall etc). One morning I woke up to see that Nightly was "upgraded" to Preview, and my extensions (except uBlock) were disabled. From what I understood, it seems like the same is going to happen with the stable branch - first it'll be "upgraded" to Preview and only later Preview will get all extensions support.
Another bitter thing about this is that as much as the new Preview UI is slicker, it's actually browsing noticeably slower on my Google Pixel.
It was unpopular on desktop, but they had an argument as to why they couldn't keep the old extension model without crippling the browser. Firefox mobile already used the new extension type, and if there's some technical limitation that justifies killing existing extensions then I've not heard about it.
They're not killing it. Preview/Fenix is essentially a complete re-write of the mobile browser, and they just haven't added all of the extension hooks back into the new version yet.
(I work for Mozilla, but not specifically on mobile.)
The reason I've heard is that the new GeckoView architecture requires piping each and every WebExtension API through a Java layer (that naturally doesn't exist on desktop). So it's some amount of manual work per API entry. The definition of the MVP only included critical extensions like reader mode and uBlock. And there's still a fair bit of other work remaining for the MVP, so nobody can spare time for noncritical tasks.
Whether that's the right definition of the MVP, and how many resources we'll be able to spend on it post-MVP (especially post-layoffs)... your guess is as good as mine, I'm going by the same public info you have.
Extensions are a pretty important distinguishing factor of mobile firefox, though.
A work in progress that removes significant features that the existing release has is... not a great idea. I'm not sure what percentage of Firefox Android users have more extensions installed than uBlock Origin, but this is a recipe for losing users. And not just any users, but the more-technical power users that tend to be best at evangelizing products to their less-technical friends and family.
On top of the missing extensions, Firefox Preview (nightly) also crashes a lot. Been giving it a spin in the past two weeks but I don't feel it's ready to replace current Firefox.
I’m really interested about TRUE reason about skipping of extensions in next version ( you can check current MESS by simply checking of number of firefox apps from mozilla only (sic!)).
For the organization of mozilla size it is not technical reason (also considering that this will be downgrade). They have ample of resources, they solved much more difficult problems in recent year, they easily can allocate their resources accordingly, they know what users ask for (simply read this discussion), they know that mobile>desktop. Yet....
There is some other factor at play, and i simply would like to know the truth. Not zuck-truth about connecting people...
I've been using the nightly builds of Firefox Preview for a couple of weeks, and it supports the extension I need the most (uBlock Origin). The new preview is so much faster than the old Firefox for Android I probably won't need to buy a new phone for another year or two. I like the UI changes, too.
For me it did a few updates ago, now it's pretty much stable. I really do like this release and i can live with not all extentions being enabled on the nightly, but for releasing this to stable users i would think they should first fix the extentions.
For the last few days for me, Firefox Preview Nightly crashes almost immediately but Firefox Preview is still stable. My understanding is that Preview picks up changes from Preview Nightly every two weeks.
(Disclaimer: I work for Mozilla but not on our Android browser)
Yes, this is such a foolish move. I have six extensions on my Firefox Android install, and the fact that five of them will not work for who knows how long after initial release makes me want to switch to another browser. I just might. We'll see.
To be fair, current firefox is very slow and resource hungry, and works well only on top of the line phones.
For example, I have 2 android phones, 1 with 4GB and 1 with 8 GB of RAM. In the 8 GB one, I can use firefox with ublock origin, but it is a terrible experience on the 4GB phone (which has a slower processor too). The new firefox is much faster and runs snappily on both phones.
This is nonsense, I am running it on samsung galaxy s5 and s7 and both are running without issues. But with a small twist, both are free from google with microg lineage installed and xprivacy lua and netguard blocking all unneeded communication. Maybe you should think about removing all google apps to remove bloatware from your phone to save the resources for somethinv usefull instead od wasting your batter life (30% longer battery life) and cpu to feed google?
Anyway, I dont use firefox due to speed but as I dont want any of google products on my phone, regardless of how fast they are. It is not about speed. It is about having some dignity within survailance capitalism. And I am filtering everything trough mitm proxy so I dont need ublock. For the fingerprinting part... I wont update untill they fix this mess.
That doesn't match my experience at all. My phone has 4GB of RAM, my firefox has over tabs up (yes I'm a monster but it makes my point a lot stronger here), and other than taking a bit to open from a cold start it's perfectly fine performance-wise.
Ugh. Now I need to choose between disabling updates for Firefox, which is a pretty big security risk, and giving up UMatrix, which I also consider to be at least somewhat of a security risk.
It's such a questionable decision. I understand that addons take a while to support, I understand for most users it's still an upgrade since they don't use extensions, I understand that at some point you need real users instead of just beta-testers. But they're taking a huge step backwards for user freedom. It's a bad look for Mozilla.
I'm disappointed. I wish they'd develop the apps in parallel until they had actual feature parity. Even if the old version of Firefox got pushed to a separate app; I'd be fine uninstalling Firefox and reinstalling 'Firefox Classic' from Mozilla. I feel like they don't really understand what the community response to this is going to be.
How big a security risk is it, really? Firefox Mobile apparently has a 0.66% usage share, which is lower than desktop Linux, and desktop Linux anecdotally is pretty much a non-target for malware authors. I, for my part, blocked my FF Mobile autoupdates upon reading this and am not going to lose a lot of sleep over the possibility that some malicious website would choose to target me over the Chrome/Safari users.
(I would also suspect, without concrete evidence, that an average default mobile browser user has more valuable data on their phone - Firefox seems like the browser of choice for the same sort of refuseniks who would still do most of their online banking and what-not on PC only rather than installing their bank's app)
I agree, particularly if the OS itself is up to date, but it's kind of irrelevant—if GP wants the web to keep working, they'll have to upgrade relatively soon.
>They're going to launch a browser update that removes support for most extensions
It's not like they're "removing" it, they rewrote half the browser and not all of the APIs are hooked back up yet. They're going to be added back eventually.
It was a preview release, but they've already started rolling it out to the regular Nightly channel and according to this very blog post we're discussing it's supposed to have a "public release later this spring".
Given that any changes have to work their way through the Beta channel, too, there's not much time left for landing changes in Nightly that will still make it into the first Release version... so the first official Release version that will be pushed out to everybody will indeed be in an unfinished state.
It is kinda of a complete rewrite as far as I know. This is using GeckoView, lots of things changed and moved places. And yes, I don't think this should be pushed to stable before extension support is in place but apparently the metrics tell a different story.
I like this browser and usually have it set as my default on my phone. I like that I can set links to open in private mode by default quite a bit.
My wishlist: 1. collections syncing to my Firefox account and desktop Firefox 2. a menu option to move a page from private mode to normal mode where I'm logged into things; right now I copy/paste the url 3. (sadly) an easy way to open the current page in Chrome.
The biggest issue I've run into is some app sign-in flows that roll through the browser or webview break when using Firefox private by default. I switch Chrome to the default temporarily to work around this.
> "[...]everyone using the Firefox browser on their Android phones and tablets will get the update."
Well, everyone who has at least Android 5... which excludes about 10% of the android landscape[1].
Most of these people probably still haven't upgraded because they a)can't afford it or b) use their devices until they're broken and will not buy a new device simply because it's newer.
I understand the move, but still, Mozilla has to communicate this in the right way and not "Everyone will have to update and if you can't, sorry for you, not our business"
The current expected lifetime of a smartphone (and laptop for that matter) is about 2-3 years. After that the batteries start to degrade. when the batteries start to degrade most devices start to down clock to preserve battery life. And even if they don't down clock they stop being as useful as their intended purpose.
It doesn't mean that you should toss your phone away, but from a security perspective your operating system is very much a legacy if it didn't get any updates for 6 years. Luckily on Android your can often flash an alternative ROM when official updates are over.
In software terms, yes, you should avoid running anything internet-connected that hasn't been updated in that long.
The problem is, of course, manufacturers who refuse to support their phones with updates for more than a couple years. Many 6-year-old phones work just fine (some needing battery replacements at some point), but any 6-year-old phone that hasn't had updates in 3-4 years sadly just shouldn't be used if you value security at all.
Smartphones, no (although with batteries being almost universally not replaceable now, six years is usually a difficult prospect without some delicate surgery), but a particular version of an OS, I think six years is plenty reasonable.
The root of the problem is that phones have so much proprietary hardware with closed source drivers that prevent the end-user from keeping the OS up-to-date like we can on PCs. If I wanted to, I could put a completely modern version of Linux on a 15-year-old desktop and it'd work fine, assuming the hardware's still fine. But with phone OSs you're completely at the mercy of your phone's manufacturer for OS releases, and so "six year old smartphone" necessarily means "smartphone missing probably at least 2-3 years of security updates and multiple OS versions behind the latest". It doesn't have to be that way, but that's the world phone manufacturers have created for us, and they have no incentive to change it.
You shouldn't be running outdated software on an internet connected device. I don't think any company updates their Android phones for that long. Samsung supports their devices for 3 years. Google might be a bit longer for the Pixel devices but they did drop their old Nexus line quicker. This isn't to say that the hardware needs to be given up on. If you want to use a community developed custom ROM like LineageOS then you can use a device longer.
I don't think any high end devices have user replaceable batteries and the battery is likely to be very degraded after 6 years of use.
If you plan to use an android phone for a long time (>2 years as manufacturer very rarely push updates beyond year 2), you should at least try to purchase a phone with active community support. For example, LineageOS wiki provide a list of such devices here: https://wiki.lineageos.org/devices/
That data is getting close to a year old at this point. "Each snapshot of data represents all of the active devices during a 7-day period ending on May 7, 2019"
Also, everyone on a Kindle Fire. My understanding is that since Firefox isn't in Amazon's app store, users need to download and then sideload it which means not getting auto updates.
I've been considering getting one and that's one of the things that's held me back (I use Firefox as my browser on my current tablet I'd be replacing)
Glad to see support for uBlock Origin landing, finally. Although to be honest, I've been using FF Preview for about a month now, and I haven't observed any ads to speak of. The tracking blockers may be enough on their own.
Sounds good. Does lastpass work on this version of Firefox? Lastpass is a really popular password manager, but last I checked it didn't work on Firefox on Android (Lastpass works fine on Firefox on desktop).
Honestly, if you don't use extensions, it seems to be pretty great. (I say, as a heavy user of extensions who's fairly annoyed right now.) It is snappy and polished.
To be honest, if a userland application is capable of kernel panicking a phone, that's a bug in the kernel, not in the application.
There's also a bug in Android (official and LOS) which I've encountered many times related to location detection in certain always-on GPS apps, causing a systems freeze and a reboot. This sucks, but it's a bug with Android, not with the application itself.
The Android/LOS peeps should really take a look at this though. I don't expect this to be the only app that can cause this and with some severe bad luck the crash might even be exploitable.
If they fix the frequent crashes, and even more frequent inability to render a page at all (just a white screen, even after refreshing), then I'll definitely give it a go. I'd love to switch back, but had to retreat to Brave for the time being...
I have this issue on the Galaxy S10e (stock). I get the white screen of death after a while. I uninstalled it because my phone was swapping out the second to last app when I had Firefox Preview installed, even when not using it. (Maybe it was being the default web view inside other apps?) The phone seems fine with it gone as I use regular Firefox.
I just tried this again after being back on Firefox Beta on Android for a few months. It's really gotten quite nice, and they added back the reader mode which I use a lot to simplify long pages on my phone. The only thing I occasionally use that's not there right now is the ability to open a site from a tab open on my desktop.
> This update will initially include support for one of the most popular extensions on Android, uBlock Origin. Additional extensions will be supported in subsequent releases
... if it supports exactly 1 extension, it is absolutely not "extension-ready"!
I'm guessing the wording was meant to suggest extension ready vs extension supporting. As in, the codebase is capable of adding full extension support, no rearchitecture needed.
I'm on nightly on my phone so have had this experience for a while. I very much dislike the new homescreen with collections instead of the 2x4 of most frecently visited sites.
Collections would be ok if it kept the old UI of a 2x4 grid. Instead I get to see 2 sites at first, and have to scroll to see more. If I close a tab, I'm not able to use the bottom 1/4 of the screen for ~2 seconds while the undo popup hangs out. This means even more scrolling. This decluttering really sucks imo.
The other annoying thing is that the upgrade starts up with an empty collection screen. The upgrade should have created a default collection populated with my 2x4 grid for me. Come on Mozilla, stop giving me paper cuts like this.
Ah, the old "we know what's best for our users" mentality. Take something good away in favor of a worse UX, with no config setting to switch back, even though old and new could co-exist.
I definitely understand them not wanting to maintain two parallel settings here. I think this failure sits squarely on the project manager who decided to couple the new behavior (collections instead of frecency) together with a new ui. I'd bet dollars to donuts that they didn't test the changes independently, and they're probably tightly linked in the code.
I imagine collections were the driving force here. If the new behavior (collections) were shipped separately, I imagine they would have certainly migrated installations to have a default `recents` collection that users could then easily delete (or migrate and maintain manually)! What a nice situation that would have been as everyone would have gotten exactly what they wanted.
If the new UI were tested separately, I can't imagine it would have gotten positive feedback. But maybe I'm just an old grump and most people like it better? It seems like shipping the UI change as a separate feature would have easily identified that the grid view was something a lot of people liked.
To make such changes, commercial products I've worked on did these steps:
1. test with some users in 1:1 sessions
2. incorporate the feedback (or consider dropping the alternative, if most all feedback was negative)
3. implement metrics
4. launch to 0.5 - 1% of users. See metrics AND feedback.
5. incorporate feedback
6. gradually launch to more
7. at some point, remove the old way
I agree it's hard to do these with volunteer work. That's how some hard issues are never solved, but Gnome gets new apps or rewrites on existing apps with almost every release -- that's the fun work. It's volunteered, so take it, or leave it, or DYI, or pay somebody.
PS: I've reached out to a few Mozilla staff via LinkedIn with a proposal for a new commercial product, but nobody replied. Anyone has a contact there?
Proposal: A Google Suite alternative for businesses, provided by Mozilla both as a cloud service and on-premise. Dual licensing, AGPL and Enterprise:
https://docs.google.com/document/d/17czxzh20c8ouhrX-iZPS2BH8...
Basically, Mozilla. Having Mozilla behind it would help with marketing. It would help Firefox too: Google has browser + office, MS has browser + office, both work to decrease Firefox usage.
Also, it would be available as a cloud offer, which Nextcloud doesn't seem to have.
This is bad news for me. I'd like at least a 4x4 or 6x4 grid to be able to pin more sites. The Highlights below the grid are not very useful. Four of them take as much space as 4 grid lines and are only 4 sites.
To be honest if I could disable collections I would. If find the feature utterly useless and take screen space for stuff I'd consider more useful, like my list of mobile bookmarks (which I could define add my default pane on Fennec).
I really wish they'd get around to implementing saving credit card info, that's one reason I go back to chrome now and then.
and there are some bugs with opening Firefox, clicking the address field on a tab that is loading and writing a search and pressing enter. That search is somehow lost if the timing of loading what was already in the tab is unfortunate. Maybe this never happens on a brand new phone on a high speed connection, but for me it is the biggest nuisance with Firefox mobile.
I'll keep using Mobile Firefox, and my experience with the new version is indeed favorable, but I admit that I'm frustrated and confused that this is being pushed to stable users before complete extension support is implemented. Thankfully uBlock Origin is supported or I'd be looking for a new browser, but my less-important extensions aren't so lucky.
At best I could imagine that Mozilla is privy to internal metrics showing that some overwhelming proportion of users have uBlock Origin as their only extension, which I could absolutely believe. Still, what sort of timeframe will it take to support the remaining addons, and why couldn't the new browser have baked on the nightly channel for a while longer until extension support was completed?
I think the metrics actually show that most users don't use extensions. But of the group that do use extensions, I'd guess an adblocker is the top one.
The question is - _an_ adblocker or _the_ adblocker (ublock)? As a long time FF Android user I prefer uMatrix and will wait with upgrade, hoping they fix this quickly. Otherwise... Not sure what I'll do. This is bad. :-(
As far as I know, uMatrix is not a (dynamic) ad blocker. That is, if some requests to subdomain.domain.net are for ads and some of the requests are for important page content, uMatrix isn't capable of distinguishing between the two. A real ad blocker uses block lists compiled (at least in part) by other people that can distinguish between two different URLs on the same domain.
Fair point, except - I am more concerned with tracking than the ads themselves. Which means I want to block Google fonts anywhere, and only allow domains when I want to. So no, I prefer uMatrix approach.
The top comments are so negative. I find the steps towards making uBlock origin directly available, and extensions available in subsequent releases laudable. I have been using Firefox mobile for at least 8 years and it's been getting faster and better with every new release.
Well, this is quite telling of Mozilla's marketing department; the link to the developer edition at the bottom is being blocked by my PiHole. It's sad to see Mozilla's open values being wasted by whatever marketing/data people they've hired to switch to massive telemetry and tracking for their own products. Their related articles are all over the place about protecting yourself from tracking companies but Mozilla themselves are as bad as any company on their own blog site.
I hope it'll be a while until the launch. I like the direction they're going with the collections and the UI. And sure, it's faster already, but without extensions I might as well just use Chrome or Brave or Edgeium for all that matters.
The preview also has several weird freezes and crashes during the minimal use I give it.
They also packaged their own certificate store like usual so my custom root certificate doesn't work. Unlike the current version of FF, they haven't added UI to add certificates to it yet. In other words, people relying on intranet access with corporate certificates will have to switch to Chrome. This is a breaking issue for me; I'm not willing to mess around with implementing the weird API my DNS provider uses to use Let's Encrypt's wildcard certificates just so an update for Firefox lets me access my intranet.
> We do a staged roll-out before publishing APKs. As far as I'm aware neither are at 100% which is a requirement for posting the builds. The APK is available from automation in the meantime if you want: https://firefox-ci-tc.services.mozilla.com/tasks/index/proje...
I'm using Firefox Preview regularly for a few month now, and it really is a huge step up in performance. Javascript and CSS-heavy sites now behave as they should, and browsing feels smooth overall. However the UI changes are a big step back and the main reason why I've not completely switched over from the regular Firefox (now I won't have that choice it looks like) :
- The address/search bar is much much inferior. Much less relevant results now come up first. When typing the domain name of a site I visit daily said site will come up low on the list, needing to scroll, while some autocompleted address I've never visited once will be on top.
- Extensions are not supported. A single supported extension does not count.
- The thumbnails are removed from the tab list. Not a huge loss but I had gotten used to it
- Popups each time you close a tab use up unnecessary screen space. Also they often just stay there.
- The address bar is now at the bottom. Change for the sake of change.
I don't know if all of these issues will transfer to the new Firefox, but I wish they would have brought in the engine changes only. Current UI is fine.
I know you say a single supported extension does not count, but they did include the one extension that was an absolute deal breaker for me if it didn't exist.
They -don't- support extensions yet, but they did make it approachable for people like me who simply won't browse the web without an adblocker.
Agreed, to be honest uBlock Origin is the only extension I'm currently running on mobile, and "deal-breaker" is exactly how I'd describe browsing without it. I'll probably give this new build a try.
> They -don't- support extensions yet, but they did make it approachable for people like me who simply won't browse the web without an adblocker.
But from reading this they will ship the update to everyone before all extensions are ready? That really sucks for me, because I also use a password manager and a bunch of userscripts on my current mobile ff.
I'm not happy with this update. They say it is "extension-ready", while they really mean it's "ready for extensions we approve". This isn't the first time they've been wishy washy about true extension support on mobile.
The whole point of extensions is they allow anyone to add functionality, not just selected parties. It is not clear at this point whether or not Firefox will make true extension support a priority.
This new Firefox for Android is shaking up to be far more limited than Chrome's Web Store which is almost as notorious for removing content as Apple's App Store.
Weirdly, I have felt this is a huge improvement over the Chrome-style address bar that hides when you scroll down. I always found it very distracting for reading, and it doesn't seem to annoy me as much when it's on the bottom (even though I think they brought back the collapsing behavior in a recent version).
> Chrome-style address bar that hides when you scroll down
Is this a Chrome thing? I thought it was just a user-hating web design thing.
A lot of news sites seem to have a topbar that becomes invisible when you scroll down and reappears when you scroll up. I can't stand this. I want to scroll a particular line of text to the top of the text pane. I don't care where the top of the text pane is. I don't care whether the topbar is visible or invisible. But adjusting the topbar's visibility in response to scrolling makes it actually impossible to scroll the text to where I want it to be.
OP is referring to the address bar (part of the browser chrome), and not the "top bar" common on news sites, which is an html element (part of the page).
On the other hand, having a phone you can reach every corner of while using it one-handed is such a nice thing (and regularly discussed here on HN as such) that it wouldn't be too surprising if OP does prefer the bar on top for exactly that reason.
Also, my experience with buttons on the bottom with larger phones is that frequently you end up trying to extend your thumb and hitting a button with your palm, so YMMV anyway.
I agree. It's also very minor, but one feature I miss from the old Firefox is being able to scroll the address horizontally. Now you have to tap on it, which opens the keyboard.
- Extensions are not supported...
Edit: maybe I misread the article. Do they mean that they're going to replace the old Firefox with Preview in it's current state? It would definitely be horrible if they pushed a downgrade that removes extension support (even if they plan to add more support later).
AFAIK they're adding the extensions API used by all extensions, but they're going incrementally by trying to support APIs used by popular extensions first. They even said they plan to add more extensions soon. Ideally, once the extension API is solid they'll allow you to install from the addon store (like old Firefox). It would be awful if they only allowed you to install they're recommended extensions.
> The thumbnails are removed from the tab list...
That was a nice feature. The icon beside tab names is pretty big, maybe they could add an option to make it even bigger and show a tab thumbnail.
> Popups each time you close a tab use up unnecessary screen space...
Agreed, although I like being able to undo if I accidentally close one. I much prefer the way the old Firefox did it (basically the same thing, but a slimmer banner touching the bottom of the screen). Hopefully they improve this somehow.
> The address bar is now at the bottom...
As someone with a large phone I appreciate this, but if you don't like it you can change it back to the top in settings.
The most important feature for me ist the tab queue, which allows one to open tabs in the background. My whole mobile web-workflow depends on this feature and they did not implement this, which makes Fenix inferior for me despite the improved performance.
> I'm using Firefox Preview regularly for a few month now, and it really is a huge step up in performance. Javascript and CSS-heavy sites now behave as they should, and browsing feels smooth overall.
Everyone keeps saying this, but I haven't experienced it myself, and haven't seen any objective evidence for it. On objective benchmarks that I've tried, Preview is maybe 5-10% faster at best. Subjectively I haven't noticed a difference.
They also slowed down the GUI animations to the point where it's visibly slow to me (and maybe a little janky too?). I prefer the older Firefox for Android for that reason alone. (Then again that may just be a personal preference: I switch the Android animations to double speed in the developer options.)
Totally agree with you on the rest of your complaints. They all ring true for me, individually and together.
For me the killer feature continues to be reader mode. It's a breath of fresh air to be able to view the content separated from the inevitably poorly thought out, user unfriendly overlays that take up two thirds of the viewing area.
190 comments
[ 3.0 ms ] story [ 225 ms ] threadMozillas own FAQ on the subject, with many more questions in the comments there: https://blog.mozilla.org/addons/2020/02/11/faq-for-extension...
If you don't want things to break occasiinally just use normal Firefox not a pre-release
> At that time, other add-ons will be supported (although there’s a chance that some of the ones you currently have installed will not be supported.)
I don't know what else you can expect? That add ons will never be deprecated?
>I don't know what else you can expect? That add ons will never be deprecated?
I thought we went through the whole round of destroying support for non-WebExtensions so that we wouldn't have to do this ever again.
This update will initially include support for one of the most popular extensions..."
Having support for exactly one means you don't support the others. They also say "Additional extensions will be supported in subsequent releases", which implies they aren't supported initially. Also supported by someone's link to https://blog.mozilla.org/addons/2020/02/11/faq-for-extension..., thanks for that.
It's phrased in a way that avoids saying outright what they are actually doing (dropping support), though.
And they're going to restore extension functionality, so the power users who depend on extensions can keep the current version and wait if they wish.
I put the odds of that somewhere around the same as the odds that GNOME Screensaver will ever support any of the xscreensaver demos it claimed it would … which is to say, I'm not holding my breath.
Unfortunately, web browsers are one of the few applications where "keep the current version" is not a viable option. Web browsers are very exposed to every web site they visit (including code injected by ad networks), and once a new version is released, many serious vulnerabilities on the previous version are revealed (for instance, https://www.mozilla.org/en-US/security/known-vulnerabilities... is the list for Firefox). Keeping an old version can be a serious security risk.
Fuck this is so frustrating.
Does this just enable all scripts loaded by that site, or does it allow filtering by domain the scripts themselves are loaded from?
E.g. when browsing site.com you want to allow site.com (and often sitestatic.com or something), but still not allow doubleclick.net even though it's loaded by the same page.
It's generally better to use ublock origin on it's own than to use ublock origin with umatrix or ublock with noscript. Mainly because ublock origin comes wiht neutralising anti-adblock scripts that would get blocked if you combine it with a separate script blocker.
https://www.reddit.com/r/firefox/comments/706xrr/umatrix_vs_...
Here is a good tutorial on how to use ublock origin in advanced mode.
https://www.youtube.com/watch?v=2lisQQmWQkY
A classic example of the baffling transparent corporate bare-faced lie. Nobody is going to believe it's ready, so why on earth say that?
Another bitter thing about this is that as much as the new Preview UI is slicker, it's actually browsing noticeably slower on my Google Pixel.
The reason I've heard is that the new GeckoView architecture requires piping each and every WebExtension API through a Java layer (that naturally doesn't exist on desktop). So it's some amount of manual work per API entry. The definition of the MVP only included critical extensions like reader mode and uBlock. And there's still a fair bit of other work remaining for the MVP, so nobody can spare time for noncritical tasks.
Whether that's the right definition of the MVP, and how many resources we'll be able to spend on it post-MVP (especially post-layoffs)... your guess is as good as mine, I'm going by the same public info you have.
Extensions are a pretty important distinguishing factor of mobile firefox, though.
Can't things be a work in progress?
The issue is when you push a work in progress as a replacement for a finished solution
For the organization of mozilla size it is not technical reason (also considering that this will be downgrade). They have ample of resources, they solved much more difficult problems in recent year, they easily can allocate their resources accordingly, they know what users ask for (simply read this discussion), they know that mobile>desktop. Yet....
There is some other factor at play, and i simply would like to know the truth. Not zuck-truth about connecting people...
This discussion only shows what people who like to spend time online talking about implementation details of HTTP user agents ask for.
It does not tell us what most users want.
(Disclaimer: I work for Mozilla but not on our Android browser)
I don't see them listed in about:crashes though, so I'm unsure how to follow up about them.
From there GitHub project it doesn't sound like the only thing that won't meet feature parity in the new version.
That being said I am using nightly preview and am really enjoying it and would recommend it.
And if you still want all those extensions you could always fork the current version like waterfox.
Anyway, I dont use firefox due to speed but as I dont want any of google products on my phone, regardless of how fast they are. It is not about speed. It is about having some dignity within survailance capitalism. And I am filtering everything trough mitm proxy so I dont need ublock. For the fingerprinting part... I wont update untill they fix this mess.
Of course I usually have JavaScript disabled.
It's such a questionable decision. I understand that addons take a while to support, I understand for most users it's still an upgrade since they don't use extensions, I understand that at some point you need real users instead of just beta-testers. But they're taking a huge step backwards for user freedom. It's a bad look for Mozilla.
I'm disappointed. I wish they'd develop the apps in parallel until they had actual feature parity. Even if the old version of Firefox got pushed to a separate app; I'd be fine uninstalling Firefox and reinstalling 'Firefox Classic' from Mozilla. I feel like they don't really understand what the community response to this is going to be.
See: https://github.com/gorhill/uBlock/wiki/Blocking-mode:-medium...
https://old.reddit.com/r/firefox/comments/706xrr/umatrix_vs_...
(I would also suspect, without concrete evidence, that an average default mobile browser user has more valuable data on their phone - Firefox seems like the browser of choice for the same sort of refuseniks who would still do most of their online banking and what-not on PC only rather than installing their bank's app)
It's not like they're "removing" it, they rewrote half the browser and not all of the APIs are hooked back up yet. They're going to be added back eventually.
A lot of corporate communication wizards might be required to dress this up as something else than direct removal.
For most users this is a distinction without a difference.
It's a Preview release for a reason. Mozilla aren't exactly pulling a KDE 4 here.
Given that any changes have to work their way through the Beta channel, too, there's not much time left for landing changes in Nightly that will still make it into the first Release version... so the first official Release version that will be pushed out to everybody will indeed be in an unfinished state.
Breaking people's experiences and worfklows in an extreme way is demonstrably not high on their list of concerns.
My wishlist: 1. collections syncing to my Firefox account and desktop Firefox 2. a menu option to move a page from private mode to normal mode where I'm logged into things; right now I copy/paste the url 3. (sadly) an easy way to open the current page in Chrome.
The biggest issue I've run into is some app sign-in flows that roll through the browser or webview break when using Firefox private by default. I switch Chrome to the default temporarily to work around this.
Well, everyone who has at least Android 5... which excludes about 10% of the android landscape[1].
Most of these people probably still haven't upgraded because they a)can't afford it or b) use their devices until they're broken and will not buy a new device simply because it's newer.
I understand the move, but still, Mozilla has to communicate this in the right way and not "Everyone will have to update and if you can't, sorry for you, not our business"
[1] https://developer.android.com/about/dashboards/
The problem is, of course, manufacturers who refuse to support their phones with updates for more than a couple years. Many 6-year-old phones work just fine (some needing battery replacements at some point), but any 6-year-old phone that hasn't had updates in 3-4 years sadly just shouldn't be used if you value security at all.
The root of the problem is that phones have so much proprietary hardware with closed source drivers that prevent the end-user from keeping the OS up-to-date like we can on PCs. If I wanted to, I could put a completely modern version of Linux on a 15-year-old desktop and it'd work fine, assuming the hardware's still fine. But with phone OSs you're completely at the mercy of your phone's manufacturer for OS releases, and so "six year old smartphone" necessarily means "smartphone missing probably at least 2-3 years of security updates and multiple OS versions behind the latest". It doesn't have to be that way, but that's the world phone manufacturers have created for us, and they have no incentive to change it.
I don't think any high end devices have user replaceable batteries and the battery is likely to be very degraded after 6 years of use.
I've been considering getting one and that's one of the things that's held me back (I use Firefox as my browser on my current tablet I'd be replacing)
Brave:
https://news.ycombinator.com/item?id=22419284
I also use Firefox Focus for links from external apps.
[1] https://github.com/mozilla-mobile/fenix/issues/5663
There's also a bug in Android (official and LOS) which I've encountered many times related to location detection in certain always-on GPS apps, causing a systems freeze and a reboot. This sucks, but it's a bug with Android, not with the application itself.
The Android/LOS peeps should really take a look at this though. I don't expect this to be the only app that can cause this and with some severe bad luck the crash might even be exploitable.
> This update will initially include support for one of the most popular extensions on Android, uBlock Origin. Additional extensions will be supported in subsequent releases
... if it supports exactly 1 extension, it is absolutely not "extension-ready"!
No user-facing advantage yet, sadly.
Collections would be ok if it kept the old UI of a 2x4 grid. Instead I get to see 2 sites at first, and have to scroll to see more. If I close a tab, I'm not able to use the bottom 1/4 of the screen for ~2 seconds while the undo popup hangs out. This means even more scrolling. This decluttering really sucks imo.
The other annoying thing is that the upgrade starts up with an empty collection screen. The upgrade should have created a default collection populated with my 2x4 grid for me. Come on Mozilla, stop giving me paper cuts like this.
I imagine collections were the driving force here. If the new behavior (collections) were shipped separately, I imagine they would have certainly migrated installations to have a default `recents` collection that users could then easily delete (or migrate and maintain manually)! What a nice situation that would have been as everyone would have gotten exactly what they wanted.
If the new UI were tested separately, I can't imagine it would have gotten positive feedback. But maybe I'm just an old grump and most people like it better? It seems like shipping the UI change as a separate feature would have easily identified that the grid view was something a lot of people liked.
To make such changes, commercial products I've worked on did these steps:
1. test with some users in 1:1 sessions
2. incorporate the feedback (or consider dropping the alternative, if most all feedback was negative)
3. implement metrics
4. launch to 0.5 - 1% of users. See metrics AND feedback.
5. incorporate feedback
6. gradually launch to more
7. at some point, remove the old way
I agree it's hard to do these with volunteer work. That's how some hard issues are never solved, but Gnome gets new apps or rewrites on existing apps with almost every release -- that's the fun work. It's volunteered, so take it, or leave it, or DYI, or pay somebody.
PS: I've reached out to a few Mozilla staff via LinkedIn with a proposal for a new commercial product, but nobody replied. Anyone has a contact there? Proposal: A Google Suite alternative for businesses, provided by Mozilla both as a cloud service and on-premise. Dual licensing, AGPL and Enterprise: https://docs.google.com/document/d/17czxzh20c8ouhrX-iZPS2BH8...
Also, it would be available as a cloud offer, which Nextcloud doesn't seem to have.
At best I could imagine that Mozilla is privy to internal metrics showing that some overwhelming proportion of users have uBlock Origin as their only extension, which I could absolutely believe. Still, what sort of timeframe will it take to support the remaining addons, and why couldn't the new browser have baked on the nightly channel for a while longer until extension support was completed?
I would not be surprised if "_the_ adblocker" was actually Adblock Plus.
https://data.firefox.com/dashboard/usage-behavior
I hope it'll be a while until the launch. I like the direction they're going with the collections and the UI. And sure, it's faster already, but without extensions I might as well just use Chrome or Brave or Edgeium for all that matters.
The preview also has several weird freezes and crashes during the minimal use I give it.
They also packaged their own certificate store like usual so my custom root certificate doesn't work. Unlike the current version of FF, they haven't added UI to add certificates to it yet. In other words, people relying on intranet access with corporate certificates will have to switch to Chrome. This is a breaking issue for me; I'm not willing to mess around with implementing the weird API my DNS provider uses to use Let's Encrypt's wildcard certificates just so an update for Firefox lets me access my intranet.
> We do a staged roll-out before publishing APKs. As far as I'm aware neither are at 100% which is a requirement for posting the builds. The APK is available from automation in the meantime if you want: https://firefox-ci-tc.services.mozilla.com/tasks/index/proje...
- The address/search bar is much much inferior. Much less relevant results now come up first. When typing the domain name of a site I visit daily said site will come up low on the list, needing to scroll, while some autocompleted address I've never visited once will be on top.
- Extensions are not supported. A single supported extension does not count.
- The thumbnails are removed from the tab list. Not a huge loss but I had gotten used to it
- Popups each time you close a tab use up unnecessary screen space. Also they often just stay there.
- The address bar is now at the bottom. Change for the sake of change.
I don't know if all of these issues will transfer to the new Firefox, but I wish they would have brought in the engine changes only. Current UI is fine.
This can be changed in Settings > Toolbar > Top.
They -don't- support extensions yet, but they did make it approachable for people like me who simply won't browse the web without an adblocker.
But from reading this they will ship the update to everyone before all extensions are ready? That really sucks for me, because I also use a password manager and a bunch of userscripts on my current mobile ff.
The whole point of extensions is they allow anyone to add functionality, not just selected parties. It is not clear at this point whether or not Firefox will make true extension support a priority.
This new Firefox for Android is shaking up to be far more limited than Chrome's Web Store which is almost as notorious for removing content as Apple's App Store.
This was just recently fixed! The "undo closed tab" toast now hides itself immediately when you switch to a new tab.
Is this a Chrome thing? I thought it was just a user-hating web design thing.
A lot of news sites seem to have a topbar that becomes invisible when you scroll down and reappears when you scroll up. I can't stand this. I want to scroll a particular line of text to the top of the text pane. I don't care where the top of the text pane is. I don't care whether the topbar is visible or invisible. But adjusting the topbar's visibility in response to scrolling makes it actually impossible to scroll the text to where I want it to be.
Also, my experience with buttons on the bottom with larger phones is that frequently you end up trying to extend your thumb and hitting a button with your palm, so YMMV anyway.
I agree. It's also very minor, but one feature I miss from the old Firefox is being able to scroll the address horizontally. Now you have to tap on it, which opens the keyboard.
- Extensions are not supported...
Edit: maybe I misread the article. Do they mean that they're going to replace the old Firefox with Preview in it's current state? It would definitely be horrible if they pushed a downgrade that removes extension support (even if they plan to add more support later).
AFAIK they're adding the extensions API used by all extensions, but they're going incrementally by trying to support APIs used by popular extensions first. They even said they plan to add more extensions soon. Ideally, once the extension API is solid they'll allow you to install from the addon store (like old Firefox). It would be awful if they only allowed you to install they're recommended extensions.
> The thumbnails are removed from the tab list...
That was a nice feature. The icon beside tab names is pretty big, maybe they could add an option to make it even bigger and show a tab thumbnail.
> Popups each time you close a tab use up unnecessary screen space...
Agreed, although I like being able to undo if I accidentally close one. I much prefer the way the old Firefox did it (basically the same thing, but a slimmer banner touching the bottom of the screen). Hopefully they improve this somehow.
> The address bar is now at the bottom...
As someone with a large phone I appreciate this, but if you don't like it you can change it back to the top in settings.
The most important feature for me ist the tab queue, which allows one to open tabs in the background. My whole mobile web-workflow depends on this feature and they did not implement this, which makes Fenix inferior for me despite the improved performance.
If you also depend on this feature, please chime in on Github: https://github.com/mozilla-mobile/fenix/issues/470
Everyone keeps saying this, but I haven't experienced it myself, and haven't seen any objective evidence for it. On objective benchmarks that I've tried, Preview is maybe 5-10% faster at best. Subjectively I haven't noticed a difference.
They also slowed down the GUI animations to the point where it's visibly slow to me (and maybe a little janky too?). I prefer the older Firefox for Android for that reason alone. (Then again that may just be a personal preference: I switch the Android animations to double speed in the developer options.)
Totally agree with you on the rest of your complaints. They all ring true for me, individually and together.
I already switched from Nightly to the Release version to continue using extensions.