Hacker News new | past | comments | ask | show | jobs | submit login
How you can contribute to Firefox 57 success (braniecki.net)
318 points by Vinnl on Sept 2, 2017 | hide | past | web | favorite | 291 comments



I just tried the Nightly on Windows and I'm very impressed. Much faster and better looking.

Long time Firefox user and have been dreading the extensions going away, but it looks like Nightly has resolved some of the problems I was using extensions for (i.e. duplicate tab context menu) and I know some of the more prominent extensions (like NoScript) are being worked on.


I can't even express how awesome it feels to hear it. Particularly, because the "duplicate tab context menu" landed just yesterday and I happened to write the final patch [0] :)

[0] https://bugzilla.mozilla.org/show_bug.cgi?id=455722


That and native containers knocks out two extensions for me. Three if you count Testpilot.

By the way, is multi-process going to make it more feasible to introduce a more robust profile feature like Chrome?

I have it working mostly on Windows but it still feels pretty rudimentary.

That's the only major feature for me that Chrome is really just so much better than Firefox.


What do you mean with "more robust"?

You are talking about the actual Firefox Profiles and not about Container Tabs, right? https://support.mozilla.org/en-US/kb/profile-manager-create-...

You can also get to an alternative, somewhat more Chrome-like profile manager by typing "about:profiles" into the URL-bar.


Thanks, I didn't know about "about:profiles"! Might be a reasonable substitute for Chrome's functionality.

I'm a FF user since the days of Phoenix but the profile management is something I've relied upon Chrome for in recent years.

That "create as many arbitrary profiles as you like!" feature is awesome for testing some kinds of web applications. Last year, I worked on an in-browser chat project. At some points I really needed to have 4-5 different users logged in simultaneously so I could send messages back and forth and track down some bugs.

Another common scenario: suppose you're developing, testing, or supporting a web application where the user experience differs depending on a user's permissions. Can be a lifesaver to have multiple browser windows, each using a different browser profile + app login, simultaneously. Rather than repeatedly logging in/out with a single browser window.

FF's Container Tabs are an interesting feature. They could get some adoption. Similar underlying technology I guess, but different (more restrictive) UI meant to guide users down a specific path. That's cool; hopefully it can be extended at some point to allow a set of arbitrary accounts to be created by devs, testers, support folks, and others.


I'm using container tabs for privacy reasons, i.e. I keep Facebook and Twitter in their own tabs and browse separately.

But the ability to keep work and home separate - separate bookmarks, extensions, etc. is really valuable. And although it's possible with Firefox, it can still be a little quirky having to setup custom Windows shortcuts, etc.

I'll check out about:profiles. Not sure I've ever seen it.


I haven't been a serious FireFox user since Chrome hit the streets nearly 10 years ago. I've gone back to try it many times. I've really wanted to like it, but it's always felt really slow and stiff. I just installed the Nightly and I totally agree – I'm impressed! It feels like modern browser again.


I switched to FF nightly after reading a comment here in HN. I had tried FF stable with electrolysis and still went back to chrome. FF nightly is SERIOUSLY fast and snappy, like ridiculously so. I recommend it to everyone.


just tried it on macOS and I have to agree it feels very well made.

Some issues though: instead of going for 'native', they still seem to feel like Firefox is a little planet of its own. It doesn't use macOS-style menus, nor transparency. Not that Chrome uses all macOS style elements, but it at least feels like it has had some effort put into making it feel at home on macOS. There's also some weird defaults / ignorance of convention: every browser launches a private window with cmd+shift+n, Firefox uses cmd+shift+p. There's still a separate search field next to an URL&search field. Probably a few more that I missed.


Glad you like it!

On the issues - what you listed is the eternal dilemma of cross-platform software. Balancing cross-platform behavior vs. platform-compliance.

There are choices that are just "Firefox-specific" like the shortcut to open private window. Other items like "Gecko menus vs MacOS menus" are tradeoffs of productivity. Having one set of menus for all platforms is easier to maintain than separate for each.

Those choices are hard to make perfect. I've been recently fixing the drop-down menu lists styling [0] to allow websites to style some of it, while keeping it looking native when they don't. It's pretty tricky to get it right :)

[0] https://bugzilla.mozilla.org/show_bug.cgi?id=1386015


I always favor native over non-native, but the FF's non-native bits are tasteful enough for me not to care. Do I wish FF used native tabs on macOS? Sure, but FF's tabs are functional enough and visually innocuous enough for me to simply not really care.

And, wow does FF57 "feel" fast. Not sure if it's because I have Stylo enabled, or what, but the UI feels faster now. I'm not sure if I've felt that way about any other FF release. And I have used them all.


Oh, I fully understand there are some trade-offs, its just something that always makes it like something feels 'off' when I use cross platform apps. There's small touches that help a lot though.. a big one is UI elements transparency, like the Chrome window bar. Firefox actually had this until it was inexplicably ripped out in 43 (even more inexplicable when Firefox _does_ adapt to GTK styling on Linux).

Let me be clear that these are minor annoyances and I respect the effort you guys do and am pretty amazed that you can make ground on a browser backed by a behemoth like Google.

I'm also pretty stoked for when you guys will have fully Servo'd Firefox, or have released Servo as a standalone product. When I tried Servo it was already oh-so-smooth :)


> There are choices that are just "Firefox-specific" like the shortcut to open private window. Other items like "Gecko menus vs MacOS menus" are tradeoffs of productivity. Having one set of menus for all platforms is easier to maintain than separate for each.

Also if you're a user who don't care about any particular platform and want just to use your trusty Firefox browser, it's annoying to learn different shortucts for the same thing in the browser on windows, linux, and macos.


As long as you're implementing your own platform independent menus, then why don't you implement pie menus, like I implemented for the HyperTIES hypermedia browser at the University of Maryland Human Computer Interaction Lab in the 1980's? (For another more recent example, see the pie menus in The Sims from 2000.)

Demo of The Sims Pie Menus: https://www.youtube.com/watch?v=-exdu4ETscs

Designing to Facilitate Browsing: A Look Back at the Hyperties Workstation Browser

By Ben Shneiderman, Catherine Plaisant, Rodrigo Botafogo, Don Hopkins, William Weiland.

Human-Computer Interaction Laboratory, A.V. Williams Bldg., University of Maryland, College Park MD 20742, U.S.A.

http://www.donhopkins.com/drupal/node/102

http://drum.lib.umd.edu/bitstream/handle/1903/362/CS-TR-2433...

Pie menus to permit low cognitive load actions:

To avoid distraction of common operations such as page turning or window selection, pie menus were used to provide gestural input. This rapid technique avoids the annoyance of moving the mouse or the cursor to stationary menu items at the top or bottom of the screen.

3. PIE MENUS TO PERMIT LOW COGNITIVE LOAD ACTION

The Hyperties browser uses pie menus as accelerators, to make commonly used commands quickly and easily available. A pie menu is a type of pop up menu whose selections are laid out in a circle around the menu center (18). The menu pops up centered on the cursor, so that each selection is adjacent to the cursor but in a different direction (Figure 1 and 6). A selection is made by moving the cursor in the direction of the desired selection, and clicking. Experienced pie menu users can make selections from familiar menus quickly and reliably without even having to look at the menu, because the menu selection depends on the direction between the two mouse clicks that invoke and select from the menu. The distance of cursor motion does not effect the selection, but the further away from the center the cursor is, the more precise the control of the selection is.

The browser has a control panel at the bottom of the screen, with buttons showing the names of available commands, to turn the page, return to the previous article, show the index, etc. When users are browsing a document by pointing and clicking on highlighted text links in the main contents window, they move the cursor down to the bottom of the screen to press buttons in the control panel, and back up to continue browsing. The permanent display of those controls is important for the novice and occasional users. On the other hand, pop up menus reduce the distraction of moving the cursor by making these commands available wherever the cursor currently is. This reduces perceptual and motor load. Pie menus are arranged with their items in easy to remember directions. For example the BACK page turning commands are to the left (and the NEXT page is to the right) (Figure 6). This arrangement facilitates gestural input and encourages development of muscle memory. Experienced users can make gestural selections from these menus so comfortably and rapidly that it is often unnecessary to display the menu. This is called "mouse ahead display suppression", and its point is to reduce the perceptual distraction.


> There's also some weird defaults / ignorance of convention: every browser launches a private window with cmd+shift+n, Firefox uses cmd+shift+p.

Wasn't Firefox (or even Mozilla) the first to have this feature and possibly even before there was any alternative?

If so, did this keyboard shortcut change in the new FF?

If not, then the 'convention' is whatever FF has always done, which the others then broke..


Or there was never a convention..


Actually more browsers use Ctrl+Shift+P. Only chrome and its derivatives use Ctrl+Shift+N. And P makes more sense too.


Humorously (at least to me) Safari has a at least a couple of standout non-native UI quirks.

One is the location of the red/yellow/green titlebar buttons. Like FF, Safari eschews a traditional titlebar and puts those buttons level with the main browser chrome. (IMHO I think this is a good trade-off to make for a browser; I'm not complaining)

Two is Safari's scrolling behavior. It seems to use its own scrolling behavior, separate from the standard scrolling behavior provided by MacOS. I've tried a few apps that modify system-wide scrolling behavior, and they work everywhere but Safari. They make Safari's scrolling insanely twitchy, with a single click of the mousewheel causing Safari to scroll 10x or 100x faster than anything else on the system.


In Firefox, cmd+shift+n is "undo close window" and has been for a very long time. Possibly even predating private browsing mode...

Other browsers don't have to have "undo close window" as a feature at all, so had the shortcut free for private browsing.


I use an old thinkpad x201 for everything, and chrome works great. Firefox flickers during scrolling(disabling hw accelaration doesn't help) and no alsa support in the builds(no i will not install pulseaudio just for firefox).


Please Firefox developers : always keep in mind that power users are loyal to Firefox for some specific reasons. One of those reasons is that Firefox is, by far, the browser with the most customizable interface. The "Customize..." feature is why I personally stick to Firefox :

http://i.imgur.com/AH1ahF3.jpg

I really don't care which browser is #1 in speed tests. I don't care about integrations with social networks. I simply want a browser that I can configure the way I want!

Another suggestion : when you change something, even if you are 100% sure everybody will like it, please at least leave an "about:config" option that can revert that modification.

Sadly, I realized that, when Firefox is updated, I'm not trilled anymore to see what new features are included, I'm instead anxious to see what has been changed/removed!


The fundamental problem with the legacy Firefox extension API is that its incredible power comes from putting incredible constraints on Firefox internals. Every stable API commitment involves some constraints, but the constraints from legacy Firefox extensions were out of control. They were killing the browser.

The "legacy" Firefox extension API is powerful because it reaches its tentacles into every corner of the internals of the browser. Keeping that API stable involves freezing a large amount of internal architecture and behavior.

Firefox needs to evolve if it's going to live, and it can't evolve in the ways that it needs to if so much of its internals are effectively frozen. It's a painful drag on development, slowing things down and preventing necessary changes.

I'm not even going to get into the problems with the API itself except to say that it had many, including widespread dependencies on synchronous behavior.

Mozilla's sacred-cow-level commitment to preserving legacy extension support is a big reason why Chrome beat Firefox on so many technical levels for so long. It ate up a ton of developer time, held off or killed critical architectural changes, was the cause of a lot of crashes, hangs, memory leaks, security problems... Thank goodness that's about to be over.

I'm sorry that you'll miss it but this is a really important step. Mozilla is doing a great job with WebExtensions, I suspect you'll be happy with how things shake out in the end.


Incredible short sighted decisions on Mozilla side, destroying their legacy plugin & addon platform. It's like Microsoft would switch off WinAPI support and kill thereby their platform. Firefox is all about "legacy API", Windows is all about "legacy API (WinAPI)" - almost no one gives a shit about WebExtension or UWP, there is no advantage over better competition Chrome / iOS or Android. Good luck with that.

Wouldn't it be better to offer a legacy addon API shim - old addons would still work even on new Firefox.


I don't see how it's short sighted. Long time Firefox user here and the extensions I depend upon Just Work (TM). The few remaining ones like NoScript are being worked on. I find it absurd when users of a handful of extensions claim "it's all thanks to us that you're successful, you'd better keep our sacred cow alive or we'll make everyone switch to Chrome". Here's the thing, I'm a power user too, and in the past I've installed modern browsers for friends and family who didn't know better. But by now, most of these casual users use Chrome because it's fast. The way to appeal to these users is to make a browser that's lighter and faster. That's how I can make a pitch to my friends to use Firefox again - "hey why don't you just use the browser that loads quicker and uses less memory?" That's something they'll listen to, rather than "hey why don't you use this incredibly modifiable browser that allows you to disable JS if you want to?"

Look, I get it. You have your extensions and you don't want to see them go away. But you're a tiny niche and even other power users don't use these extensions that you find essential. Certainly the vast majority of users rarely install extensions at all. Mozilla is doing the right thing and an entire thread praising the performance of the latest version vindicates their approach.


Why not just give the NEW browser that is incompatible but faster a different brand name? Why is the NEW browser not based on Servo? The potential new users don't know Firefox anyway, and the current user that still use Firefox despite Firefox-tech-standstill for like 5 years (multi-process is coming next year, we promise) are power user who customized it heavily (and deactivated phone-home/telemetry). Microsoft made the same mistake, same rabbit hole.

Where is the next Servo based browser? Instead of refactoring the 20year old mess (with COM and XML), the should focus 99% on Rust and Servo, and finally release a stable build with a lightweight HTML5 UI (Vivaldi).


> Why not

Well, you answered your own question. Clearly they have a userbase that will stay with them out of inertia or loyalty. These users can now benefit from the improvements in the latest release. If they released a new browser, few people if any would try it out because of inertia. The lack of users would kill the product pretty quickly.

Of course, some users who rely on old addons will be left behind, but they have the option to use Firefox Extended Support Release, which will get security updates. Eventually the extensions situation will sort itself out and these few users can move to the main release.


> Where is the next Servo based browser?

That's Firefox. Many of the components in Firefox 57 are taken from Servo.


>I really don't care which browser is #1 in speed tests.

You're in the minority then. The majority of users care, a lot, about speed, and Firefox is visibly slower than Chrome, probably one of the main reasons it has lost so much share.

Personally I care about both, that's why I am so conflicted about this update. But it's disingenuous to assume this is an unwarrented regression. You cannot get better in terms performance while keeping the extension system as it is. This reworking is necessary for Firefox to even compete wit Chrome. I understand the dillema but this is far from clear-cut one way or the other.


    The majority of users care, a lot, about speed
Yeah. I care about browser speed a lot; I'm one of those guys always looking for settings to tweak and measuring results and so forth. I spent I-don't-even-want-to-know-how-many hours a week using a browser so even small improvements become noticeable quality-of-life upgrades.

But, still:

1. Benchmarks are important, but aren't always the best reflection of actual performance, either real or perceived. Subjectively, Safari on Mac and Edge on Windows "feel" fastest to me even though benchmarks say otherwise. Within reason, that's what typical end users care about: "feel" vs. benchmarks. (FWIW, Firefox is my main browser; I'm not telling anybody to use Safari or Edge)

1a. Chrome benchmarks the best, but it also uses the most memory due to its multiprocess model, so for users on memory-constrained devices (particularly with spinning HDDs, where swap's going to be really slow) I'm not convinced reality always lines up with benchmarks here

2. The user's network connection is still usually the biggest factor.

3. I don't think actual site content's getting much heavier. All this extra browser performance is just getting soaked the hell up by heavier and heavier ads. So I can't get as excited about browser performance as I used to.


Just wanted to say that, annecdotally, I switched from chrome to firefox not because of speed, but stability, four years ago. Chrome would reliably crash on my laptop after opening three tabs, while firefox would let me open dozens without trouble. I guess it was an OOM issue, but I never really cared. Firefox worked, chrome didn't.

The added extensibility, while nice, was mostly an afterthought. The only extension I'll miss is cliget, and hopefully it can be added back eventually.


Why do you want to customise so much? I know there are people like you but maybe like the Firefox developers I don't understand it. I use software as it comes. I almost never change any options. I don't even change my desktop wallpaper. I think making developers add options for everything creates bloat. Just use the defaults!


I'm not quite sure if you are being sarcastic or not, but one of the things I value more in software is costumizability. I don't want someone else to make my choices for me and force me in a given mold. No browser, no software, can be made perfect for everybody. Different people want different things, so a good software must always offer users the option to change it to fit their use case.

For instance neither "classic" top tabs nor a side list/tree is the "best" or "right" way to design a tab selector. Some users will prefer one, others the other. You should give them choice.


I'm not being sarcastic - I genuinely don't understand why people value being able to tweak such trivial things and why they want to spend their precious time and effort doing it.

I think building support for a choice about something like tab placement is a waste of time and code - I don't understand why people can't adapt to one way or the other and then we all benefit from simpler code with fewer bugs.

I don't think we need to decide if top or side tabs are better, because I can't see how it matters much. Just pick one, hard code it, and everyone get on with the things that matter.


It does matter to me. When I'm working I often have a large amount of tabs open. Side tabs let me view more tabs at one,which top tabs can't since they don't let me view more than ~10 at once. Furthermore, screens are wider than taller, while webpages scroll up and down. Makes much more sense to have all the real estate you can get vertically, while the tabs go on a sidebar.

I spend several hours every day on a browser, switching tabs hundreds of times. I am measurably and subjectively more efficient with side tabs than top tabs. Why is this something that doesn't matter? It's not trivial, I literally cannot browse without side tabs without getting annoyed. It's absolutely not a waste of time and code if it's important to people.


There are always things that people want to customize, and this is where Firefox differs from Chrome. An example is the recent change in Chrome that make the Backspace key no longer go back a page. Chrome just flatly forced users to adapt it, with no configurable option anywhere, whereas if Firefox ever adopts this, I have peace of mind that I can always go to about:config and change it back.


In case you didn't know: https://chrome.google.com/webstore/detail/go-back-with-backs...

Not really a "configurable option" but very close to it, it achieves the same goal and has been released along with the Backspace modification.


Tree Style Tabs.


Multirow tabs


> Just use the defaults!

Just customize!


Adding support for customisation has a cost - more code, more things to test. If people just used defaults we could have simpler code, shipped more quickly, with fewer bugs.


So?

Not supporting images or JavaScript would simplify things as well...


So supporting configurable options takes resources away from implementing useful features. I'd rather we had more useful features, and more bugs fixed, than silly options for fussy people to rearrange where their tabs are in their browser.


> silly options for fussy people

That's just your opinion—which I guess the Firefox devs and most users don't share.


> So supporting configurable options takes resources away from implementing useful features.

Customizability is a very useful feature for many people.


I've just had my "last straw" moment with Firefox after over a decade of using it. Several addons broke with their developers saying, in effect, F-it, and then it lost all of my saved logins. So, yep, F-it. Firefox has been moving toward becoming a clone of Chrome for years. I just skipped to the chase and started using Chrome. Congratulations, Mozilla. Nice work.


Instead of going to Chrome, I think you should consider moving to the better alternative to Firefox, Pale Moon. It has all the same mindset of decent Firefox, XUL compatibility (and an oath to never use WebExtensions), and other features.


I was interested in Pale Moon, but you just convinced me to skip trying it. Sandboxing is top important, and WebExtensions are rather important there.


Back in ~2015, there were several Firefox developers who visited a local hacklab where I used to hang out. I was aware of a bug in the current version of FX at that time, and I knew how to fix it. I wanted to get their help to create a patch for the bug and get it submitted.

After about two hours of trying, we were unable to produce a working build from a tagged release point in their version control - even before applying any changes.

I'm hoping the new tooling works better.


I'm sorry you had this experience.

I'm also happy to say that we actually did improve our build chain. I have a build env for Firefox on Mac, Windows and Linux and it all works quite smoothly. The only issue is that full compilation takes time (30 minutes on my laptop).

Fortunately we also now have Artifact Builds which allow you to hack on the front-end JS code without having to rebuild the C++ backend [0].

[0] https://developer.mozilla.org/en-US/docs/Mozilla/Developer_g...


You should download the code today and try to build.

https://developer.mozilla.org/en-US/docs/Mozilla/Developer_g...

It's just a ./mach build after checking out the code and running the bootstrap script.


I'm curious: Which operating system?

The Linux x86_64 build experience has worked very well on Ubuntu for many years. These days you don't even need to copy and paste the apt incantations: ./mach bootstrap takes care of that.


It was on a Mac, which I admit probably isn't where the core of development lies.


Many gecko/firefox devs work on Macs, so it is very surprising to hear that, unless your are running a somewhat old and unsupported version of MacOS (you need 10.7 according to https://developer.mozilla.org/en-US/docs/Mozilla/Developer_g...).


Oddly my impression is that their focus is more on Mac and Windows than on Linux. I swear that whenever i see them presenting something, it is invariably done using a Mac.


one of firefox engineers is working on https://janitor.technology, which allows to build firefox in an online ide


Had a similar experience trying to debug some UI problems I was facing a few years ago. Building and contributing to Chromium is a breeze, in comparison, that is as long as you have a steady and fast Internet connection.


Reading the comments, I get that many users are upset about the XUL extension APIs being dropped, but what I don't get is that many of them plan to switch to Chrome.

a. You're getting a less powerful set of APIs in Chrome than in Firefox, even with WebExtensions being used by both.

b. You're supporting a browser by a major corporation that is already number one anyway and has a IE-style chokehold on the web, giving them even more power is not the best solution, helping Mozilla make the new APIs as close in power to the old ones would be a better alternative in my opinion.

Again, it's everybody's personal choice and am not saying what anyone should be doing, just found it curious that many consider switching to WebKit/Blink to be the solution, where they won't get more power and give Mozilla even less of a say in terms of web standards and just having a decent alternative to WebKit.


Dear Mozilla developers! I've been using Firefox since v1.0. And I want you to know that by ditching the old plugin architecture you are alienating your faithful users, myself included.

I'm sorry, but I decided to stick with Firefox 55 for now, possibly switching to Palemoon later. I need my old extensions to work: Status4Evar, EdgeWise and Classic Theme Restorer. I can't stand the current FF UI and I will not give up on this.


While we're throwing anecdotal statements around...

Dear Mozilla developers!

I've been using Firefox as my daily driver nearly non-stop since it was Phoenix and I'm totally cool with you dropping the legacy stuff.


I'm the same. Been using since Phoenix 0.3 and am stoked about these changes, and the thought that Rust, and the major perf and safety improvements that come with it, are beginning to find their way into Gecko. Amazing work from a team that many had dismissed as irrelevant.

Prove them wrong!


Dear Firefox developers,

I’ve been using Firefox as my standard browser since 0.9.3 (except for a year or so when it was run-a-browser-from-a-USB-disk-or-use-IE and Firefox was just too terribly slow, so I used Chrome—by the next year Firefox had fixed its game and I could switch back), and for the last few years Nightly has been my standard browser.

I get why you’re dropping the legacy stuff and I’m looking forward to what’s possible after it’s gone. When the announcement came out in November last year that WebExtensions would be the only way come 57, I was intensely sceptical that it’d be ready within a year (no one was using it at that time) and prepared for a deal of pain: I’d just have to live with it, come November. (It’s not like I have any better options in mainstream browsers; they’re all hopeless with even only twenty tabs.)

Well.

You’ve proven me wrong, and I’m delighted. I was able to replace all my legacy extensions with WebExtensions ones with minimal loss of functionality (I miss the ability to style the chrome with Stylish, not that I was actually using it for anything other than shaving half pixels off here and there on my high-DPI screen). My Dad (also a programmer) is still grumbling about some status bar thing that broke a short while ago, but we went through his extensions and came to the conclusion that it’s really not that bad—the extensions that stop working at 57 were ones he didn’t use, or ones where there are already good alternatives available or a clear plan for the changeover.

(The fact that in several cases I did have to swap out one extension for another was a mild nuisance, but as a Nightly user I’m doing this a few months before everyone will have to, so I expected it.)

All in all, I’m happy with how things have gone and impressed at how well the migration is going. All that I really want now is to be able to hide the horizontal tab strip in favour of my vertical tab bar. And it looks like https://bugzilla.mozilla.org/show_bug.cgi?id=1332447 is almost there. (It’d also be good if file: URLs didn’t mess with said vertical tab bar something awful, but as I haven’t filed any bugs about that yet or investigated to find existing reports I’m not allowed to complain.)

Good job!


I'm totally cool with dropping legacy stuff that has a proper replacement.


Then you must hate all of your software. It's impossible to always continue in the direction that happened to be chosen more than a decade ago. At some point, you will run into problems, which in the case of Firefox were severe problems in performance, security, maintainability and complexity as well as stability of the extension API. You can't fix all of these problems without breaking things, without doing things differently.

And yes, it is to some degree subjective whether this is worth it, but at the end of the day, you, as someone who uses lots of extensions and complex extensions, are only a very small fraction of the Firefox user base. Most users have no extensions or just the obligatory ad blocker. They don't benefit at all from keeping onto the old architecture with all of its problems, which they are suffering under.

And then, yeah, Mozilla has to at some point piss you off in order to help the majority of their user base. It's not like all other browsers had such a complex extension API and Mozilla is just being lazy by dropping it.


Dear niche community using legacy extensions: We're sorry, but legacy extensions are holding back Firefox development in every way meaningful to the vast majority of users. In order to continue to exist as a viable option for an open web, Firefox must move forward. - (What I imagine they'd say)

Of the extensions you listed, the largest (Classic Theme Restorer) is only used by 333,000 people while the smallest (Edgewise) is only used by 218 people.


Your arguments may convince Mozilla not to port these features to WebExtensions, but what are you trying to tell me? To go fuck myself? What you're missing is there are many niches like these, and many people using many niches. It might still be a small percentage of users overall, but we're the geeks and we sit right in the center of the culture. Get rid of us, and the care-about-Mozilla temperature goes down significantly.

Mozilla should find a way of moving forward while keeping the code open enough to modding. AFAIK, features offered by these extensions that I use are not implementable in WebExtensions even in principle, and that is the main problem. Break the compatibility if you must, but make it possible for someone who cares enough to step in and rewrite these extensions to the new architecture.


The vast majority of Firefox users either don't use extensions or only use an ad blocker. I downsized to the same myself quite a while ago for performance and stability reasons in Firefox despite being one of the niche users that used to keep 10 extensions installed. Yes, there are quite a few different niches of users that use specific legacy extensions. But the legacy extension system is a mess and has been holding Firefox back for a while. Mozilla is going to fully embrace WebExtensions and keep moving it forward to enable as much functionality from old extensions as possible to serve several of those niches.

I'd argue that many of the users that choose Firefox over Chrome do so for the fact that Firefox is 100% open source and isn't tied to an advertising giant trying to capture tons of extra private data from you. This won't change that. It will mean you can't change the layout of the entire Firefox window and add tons of new custom buttons and things like that. But most users don't care about that.

For the niche that can't stand using a UI they can't fully customize, they're free to use a slower and less secure browser that continues to support XUL and the associated extensions while still keeping up to date with at least the slower Firefox ESR branch of renderer underneath. Pale Moon, for example.


Legacy extension system vs WebExtentions is actually a false dichotomy. Mozilla still can develop a low-level API in front of its new and improved subsystems. Yes, that will require more development and will be harder to maintain. But the benefits it would bring may arguably be worth it.


Mozilla is already planning on extension WebExtensions beyond what Chrome does with it (I believe they even implemented some of it already) and are planning on adding even more to enable the functionality of some of the more popular legacy extensions (bookmark trees, tab management, etc).


There is an umbrella issue that aggregates proposed improvements to WebExtensions [1]. If you look at RESOLVED WONTFIX issues linked in that bug, you will find a lot of them are rejected based on political reasons. For example: [2]. So there will never be anything close to feature parity with XUL extensions.

[1] https://bugzilla.mozilla.org/show_bug.cgi?id=1215059

[2] https://bugzilla.mozilla.org/show_bug.cgi?id=1246706


For every user of Classic Theme Restorer, there are probably a dozen more Firefox users who complain about the unnecessary UI changes but aren't aware that CTR exists to fix some of the usability regressions.


> only used by 333,000 people

doesn't sound like an "only" to me.


Firefox has 80 million daily active users, so that's an 'only' in context. Keeping XUL around and all of its messiness to support a small niche of users at the expense of what most users want is part of why Firefox has lost so many users to less customizable, (arguably) less capable competing browsers.


There are a boatload of assumptions in that statement that are pretty unwarrented. The vast majority of why chrome is popular is because google.com and youtube.com push it nonstop, not any other imagined reasons.


You're forgetting that people who want to use the features older versions had but not anymore are likely the same people who have been using Firefox for a long time. These are the same people who drive Firefox adoption by being experienced long-time users giving advice to the young'uns. If you lose that core user base, you'll lose the rest of 80 million to Google/Microsoft ads in no time.


likely _some_ of the same people..

Plenty of those long-time users giving advice to the young-uns are more than happy yo sacrifice a small part of firefox for big improvements elsewhere.


I've been using Firefox since 1.5, and I understand completely why they have to drop their legacy extensions framework to modernize the browser. In recent years I found myself using Chromium more and more because of its responsiveness. Firefox 57 is really close to where it needs to be to let me drop all my use cases of Chromium I have now. Yes, it means 4 of the ~12 extensions I've used for years stop working. Most of those have a web extension version being worked on or aren't worth the competitive disadvantage of supporting a 15 year old addon API.

I've seen plenty of my favorite software projects have to make "hard breaks" to modernize. KDE broke everything with Qt5 / Plasma 5, but the payoff is now they are the premier high DPI Linux desktop. Python broke everything with 3, but features enabled by the changes have made Python 3 the best scripting language ever in my book.

Firefox would be doomed to a slow death with its legacy extensions API the same way PHP and Perl languished for years with the inability to push a major version. A slow death is not a downfall anyone wants.


The issue is not that things break, the issue is that the WebExtensions API is missing a lot of features so a lot of addons can't be ported.


It's missing functionality because the old API allowed deep, deep hooks into FF's internals.

Per Mozilla, aside from security problems and other nightmares, these deep hooks allowed developers to write plugins that were incompatible with a multiprocess browser architecture where each tab/window can have its own sandbox and remain responsive even if other tabs go down.

And that is one of the killer features that has caused FF to bleed massive amounts of user share to Chrome in the last decade: Chrome almost always "felt fast" in situations where FF ground to an embarassing halt because something was happening in another tab/window that was causing the entire browser to become unresponsive.

I don't know what precisely is missing from the new WebExtensions API but I have to imagine that Mozilla will work to expand its capabilities as much as possible over time...


WebExtensions is also missing functionality because Mozilla set a release date before figuring out what APIs would be needed or how migration would work.

The Add-on SDK was supposed to fix the "deep hooks" problem and is mostly multiprocess-compatible. As far as I know, most currently maintained XUL extensions are multiprocess-compatible, too.


I rely mainly on one add-on: Vimperator. It's a big big productivity enhancer and it's the main reason for sticking with Firefox for so long. Even in a period when Firefox was slower and buggier than the alternatives.

It's sad they will deprecate the old extension interface and make Vimperator a no-no. I understand the reasons behind WebExtensions, specially those related to security, and reckon the benefits of the multiprocess paradigm, but the flexibility of the old extensions API was a big advantage in my pov.

No, Vimperator won't be ported (I wish it could), because WebExtensions is a much more limited API. And no, Vimium and VimFx doesn't come even close to it. They are not alternatives.


I hate to add so little to the discussion, but I am on the exact same boat, and I am quite anxious about the upcoming change.

If there is anything to make either Mozilla hear this use case more, or contribute anywhere to make Vimperator a possibility for the future, I would be happy to do my part.


fwiw they seem to be actively trying to support vimperator like webextensions in this bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1215061

What does vimperator do that vimium and vimfx can't?


Though I haven't used it, it seems to radically make firefox into something like Qutebrowser. Not just the key bindings but also the UI which is not possible under new api for Web extensions while Vimium and vimfx only give vim key bindings in the browser (Not the full experience for those who are used to vimperator).


I've been using Firefox since ~Phoenix 0.1 and I don't care at all about the legacy plugins.

It would be interesting to see what percentage of their long time users use more than 1 or 2 add ons (I have 4 installed and only really use 2 of them).


In my case, the critical extension for me is Tab Mix plus (770k users) for multi-row tabs. It's really hard to use a browser without this feature because I've been using it for so long.

The real question is for me is how many users haven't moved onto Chrome -- which is obviously faster and more stable -- because of 1 or 2 add-ons.


i've used ff since the phoenix days (0.6? do not remember). this particular laptop now has 55.0.2 with Cookie Monster, HTTPS Everywhere, NoScript, Policeman, Show my Password, Tab Mix Plus. some other extensions are installed but not used.


How many of those users really care about firefox and dont just use it for inertia reasons? (already installed, etc)

Frankly if they arnt using extensions they are going to have a vastly better user experience on chrome, and google has infinity manpower to always ensure thats the case.

So what is firefox gaining by becomming a crappier chrome? Why the hell wouldnt I just use chrome at that point, because I feel so great about mozilla, who regularly brushes me off when i try to take advantage of things unique to firefox?


I use mozilla nowadays on my low end laptop because chrome chokes on it most of the time (just plain hangs and stops working with more than 10 tabs which is all the time for me). FF 55 is much snappier with most of my important extensions and doesn't choke as much. So I moved from chrome to FF after 55. Hope they continue to make it better.

P.S FF is now not a crappier version of Chrome but Chrome is crappier if you don't have powerful hardware or RAM (mine's hardly 2 GB and everything sucks a lot though its only 4 years old)


One more voice for the above. I will be stopping at Firefox 56 until I decide what works best for me. I rely on several plugins daily that are about to be made obsolete. Likewise I revert Firefox back to a very classic UI interface that is very compact and familiar. I have avoided Chrome all these years because I can't stand its UI. Now I'm also about to loose the ability to make FF feel familiar along with many of my required productivity add-ons.

I'll ride out FF 56 until I decide where to go next but honestly the many times in the past it has felt FF has ignored the requests of long term users, coupled with lack of differentiation from Chrome, I'm not sure I have much reason to stick around.

Kudos to the team for all the hard work improving the FF core. But you're about to abandon that which kept long term users like myself here through all those rough years. Makes me sad.


I'm going to be switching to the extended support release[1] (based on Firefox 52 but with security patches through June 18th 2018).

I don't mind WebExtensions, and cross-browser compatibility is a nice goal, but a half-baked system that breaks the most important distinguishing feature of Firefox (extensions that can modify any part of the browser) is ridiculous.

[1] https://www.mozilla.org/en-US/firefox/organizations/all/


I'm already on ESR, but my reason is I need java plugin for my company OEBS (it supports web start, but it takes ages for our admins to patch anything). At the same time, keeping legacy plugins working is a big bonus, I love Classic theme restorer.

Still undecided what to do after june 2018. I guess I'll just try firefox and chrome side by side and move to chrome if there is no real difference. It is probably more convinient to migrate to chrome, since it is on the phone by default too.


I don't agree.

Yes it's a change but it serves a good purpose - to standardize the extension development and making sure that 1 extension can be running on multiples browsers e.g. Firefox, Opera, Edge, etc..


WebExtensions is the lowest common denominator.

Yes, it's a good purpose to have a standard base. It's bad, though, that nothing besides the blessed parts are officially customizable anymore.

Previously, extensions had messed up with anything they wanted (they could override most of the browser's core). That had allowed third parties to do incredible things that weren't ever considered by the browser developers.

With WEs, though, the only things that browser developers had specifically thought about are possible. All for user's convenience, so they can switch to Chrome, Edge or Opera.


It's the same thing as plugins, unsigned drivers, etc. etc. If the platform gives unfettered control, that is nice in that it gives power users flexibility, but it comes at the cost of seriously jeopardizing the experience of non-power-users who don't understand the intricacies of the software. They expect the platform to protect them by default. This applies to the whole package -- security, performance, and overall experience.

Vendors have learned that if they fail to provide this protection, the average user will blame them directly, not the extension that makes memory usage bloat 6x. The solution is to enforce stringent controls, including limiting modifiable parts of the experience to some well-defined areas.


>All for user's convenience, so they can switch to Chrome, Edge or Opera.

Nah, it's the other way(s) around - it's for [longterm] developer convenience; so users don't have to switch to another browser because an add-on they need isn't available in FF.


Unfortunately I don't think so. The only draw keeping FF alive for years has been its extension system. They gave up the unique UI flexibility, now they are giving up the flexible extensions. When Google makes it so easy to find and install Chrome everywhere, and sync it with your mobile device, now that the extensions world will be on "par" FF is even less distinct than it was and thus less reason to use it. This is the beginning of Firefox's final dive.

In a few more releases I really won't have any concrete reason to recommend FF over Chrome anymore (and no, 99% of users outside of this forum don't care about memory usage and keeping 100 tabs open at the same time)


This too, but to the lesser extent. That's because Firefox had much more powerful extension system than other browsers.


Then they should bring back the old UI we came to love. I want my blue title bar on Windows, my menu bar, my status bar, my square tabs, etc, etc. Also, I can't imagine living without EdgeWise, so they would have to incorporate that, as well. Other people use different sets of features, which are also not supported in the new plugin model. How can Mozilla ever hope to plug all these holes (pun intended)?


> I want my blue title bar on Windows, my menu bar, my status bar, my square tabs, etc, etc.

So like this [0]? The title bar isn't blue, but neither is any other title bar on Windows 10.

To achieve this on Nightly I opened the Customize view, switched the theme to "Light", checked the "Title bar" checkbox, and enabled "Menu bar" under the "Toolbars" dropdown. Tabs have squared edges by default.

[0] https://vgy.me/KvNgxS.png


No, like this: https://vgy.me/VBLBYH.jpg And when the FF window is not focused, the title bar is grey. Like it's supposed to be.

It's not part of a plugin, though, I had to patch something manually (I never remember what and have to re-google it every time).


You get square tabs in 57 i think.


> my square tabs

Square tabs have been back for a while now. The Compact Light and Compact Dark themes are included in Firefox by default. Set the theme in Add-ons Manager -> Appearance.

I use Compact Light.


Actually, it still does support legacy extensions and probably will do for a while. There are `browser/features/*.xpi` files in the browser install directory. Even in 57, they're all "legacy" extensions, and the only WebExtension there is `screenshots@mozilla.org.xpi` (even then, still a "legacy" install.rdf-type file, with embedded WE).

And I've just copied the legacy `https-everywhere@eff.org.xpi` non-WE addon there - and it worked ;)

I'm sure they're going to plug this "hole" shut, but it's still there. As well as patching omni.ja.


If it keeps working in FF 57 and won't break after every minor update, I might upgrade. I was actually thinking that providing some kind of officially unsupported backdoor for determined users like us might be a way out of this situation.


No, it is going to break after every minor update. Most legacy extensions already don't work anymore in Firefox Nightly, even if you tell it to load them [0], simply because Mozilla has very much already used their new freedom to throw out tons of legacy code [1] and to refactor what's left.

The Mozilla-supplied legacy extensions continue to function, because Mozilla is updating them as they make changes to the internals of Firefox. So, I imagine maintaining such a legacy extension is now as much work as it is to just directly contribute and maintain the feature in Firefox's codebase.

[0]: "extensions.legacy.enabled" in about:config. This won't be shipped in Stable, though, and will likely be taken out from Firefox Nightly soon or with the release of Firefox 57.

[1]: https://bugzilla.mozilla.org/show_bug.cgi?id=1347507


I'm not involved in FF development, but I find it exceedingly unlikely that this will keep working for much longer than a few releases after 57. The whole purpose of all of this is to rip out all the XUL baggage to decrease maintenance cost and make it easier for the developers to focus on actual features and fixing bugs that actual people encounter (as opposed to keeping legacy code paths working where no one has a clue if they are even used anymore).


Ripping out XUL baggage is fine, it doesn't matter which language UI elements are described in. Yes, it would require to rewrite stuff, but that's manageable.

The important part is ability to replace (overlay) chrome resources, patching/extending the browser code. I hope that's going to stay for a while.


The internal APIs will eventually change (break). They already did on major releases - one can find lots of "does not work on Firefox NN+ anymore, please update" comments all over AMO. (By design, this should be significantly rarer occasion with the WebExtensions)

Now there is no requirement to keep them stable even across the patch-level releases. Whenever things would break faster or at the same pace is unknown.


I know it's not a good way out, but it's something we could at least in theory work with.


I used to do that until I discovered the Compact theme (which is now installed by default, previously only available on the developer edition without installing an add-on).


Love the compact theme. Glad it made its way into the release channel.


I've been using firefox since the very early days and I am excited about ditching the old plugin architecture.

Oh, and good news, the ui you dislike is going away.


The main problem with legacy extensions is that they won't work with multiprocess firefox by design. Multiprocess is the only way forward, because of the enormous speed and stability benefits.

Unfortunately Mozilla has rushed the transition, and there hasn't been enough time to add important APIs and for developers to update their extensions. Firefox 57 is on the verge of being released, and there are still many addons which aren't webextension compatible.


Isn't v55 already multiprocess? How are these extensions working right now?


many of the apis were shimmed in such a way so that they'd work, but could still slow down the browser. Those shims will disappear.


If you have any legacy extensions installed multiprocess is disabled. If you don't have any, it is turned on.


Not correct. Some legacy plugins work fine: the add-ins store relied on user testing and flagging to confirm which still worked.

https://developer.mozilla.org/en-US/Add-ons/SDK/Guides/Multi...


> I can't stand the current FF UI and I will not give up on this.

Then the simple way forward is to make your own browser, isn't it? Use an embeddable engine which does all the hard work for you and focus on making the UI the way you insist it must be. It really could be a single developer project.

So why not just do that?


>>alienating your faithful users, myself included.<<

I wonder what the telemetry says about those plugins. If it shows a minority, than Mozilla many not give it a thought at all.


The telemetry probably shows that all the power users opted out of telemetry.


It's the other way around I think, telemetry isn't (wasn't?) on by default and I enabled it on purpose. And I'm a power user, that's why I don't use chrome (1k tabs here).


Telemetry isnt the whole story, especially if it is being interpreted by somebody looking to justify decisions.


I completely left Firefox over this and other bullshit.


Sadly, I don't think the devs have much of a voice anymore. Corporate and upper management have taken over, as shown by the various SJW and usability debacles.

I'm always concerned of this happening to other large-scale independent FOSS projects. Special interests tend to take over without some benevolent dictator.


The devs still have a voice in the sense that they can always quit and fork. Or, in case of Firefox, join the Palemoon team.


Joining Palemoon is highly unlikely - there's bad blood between the two communities.


Hit the download button for nightly, only to start reading the comments here and realizing that a bunch of addons will stop working. Checking my list, FireGestures will stop working, Self-Destructing Cookies apparently already stopped working this version (broke localStorage deletion), and a more obscure one (QuickJava) is also legacy. The latter is the only one I could find that adds easy on/off toggles for Javascript, CSS, Flash and Proxy. They want to receive telemetry to see how people use nightly? Well, here's my datapoint: can't use it altogether.

I might as well switch to MSIE as update to 57.


This sheet may help you find replacements: https://docs.google.com/spreadsheets/d/1TFcEXMcKrwoIAECIVyBU...

FireGestures -> https://addons.mozilla.org/en-US/firefox/addon/foxy-gestures... SDC -> https://addons.mozilla.org/en-US/firefox/addon/cookie-autode...

It'll take a bit for the new generation of extensions to fully replace the old one, but it's happening.


I'm not sure they will. Vimperator etc. has a lot of dependencies some of which will never come to be. But the parts that could, like a proper keyboard api hasn't even been reviewed for months. The community complained about the lack of apis, they made a webextension experiment, which is apparently the way you ask for api extensions and it's pretty much just being ignored(the thread is 2 years old [1][2]). The customization thing mentioned also applies since the plugin creates a custom status bar and hides all sorts of other stuff.

I'm not sure what browser I'm going to use to be honest. All the hotkey browsers are buggy as hell since they use qtwebkit, webkitgtk, qt-webengine and whatnot. Chrome has extensions like cvim which sort of work, but they suffer from issues like not being able to properly hook keys and properly notify the user about the state you're in making me frequently make mistakes on what I want to do. As the sibling said a poor imitation of what was possible before.

If firefox just immitates chrome, what's the point in having it? How about trying to eat some of IE's lunch by making enterprise customization a first class citizen? But then again not sure if that matters. And yes I'm aware of GPO[3], but calling that first class citizen is disingenuous at best.

[1] https://bugzilla.mozilla.org/show_bug.cgi?id=1215061

[2] https://github.com/Koushien/keyboard-shortcut-api

[3] https://addons.mozilla.org/en-us/firefox/addon/gpo-for-firef...


Did you try Pale Moon?


I indeed looked for alternatives and they're just not there yet.

The more I read, the more APIs turn out to be missing or buggy, making it impossible for add-ons to be made (if the developer is active in the first place). I just don't understand this. Firefox was built on the extendible principle and they're now going back to "v0.2 beta" in that regard.

This process should have taken years to mature the new APIs before dropping support for the old method.


So, look at this graph: http://netmarketshare.com/report.aspx?qprid=1&qpaf=&qpcustom...

Notice that major bump in August 2016 and after? Firefox 48 was released on the 2nd August 2016. This was the first release to ship the new multiprocess architecture. And this new architecture also already required breaking all extensions.

They could not have shipped it this early without knowing that the switch to WebExtensions will cut off the multiprocess-incompatible extensions.

And they needed to get multiprocess shipped. Had that graph continued as in the half year before that, we would now have negative user numbers.


Downthemall was the only reason I was still using firefox. The "replacements" are extremely poor imitations, disappointing but I guess all good things must come to an end.


From a link on your sheet, I found that Tab Mix Plus multi-row tabs has over 700k users! This is the killer feature for me and Firefox and I won't be upgrading without it.


Based on the number of signatures this petition is garnering, I'd say you're not alone in your sentiment: https://www.change.org/p/mozilla-save-mozilla-firefox-s-best....


TLDR; could someone point me to a tutorial or an explanation on how to properly install FF, update & save the open tabs without making bookmarks (that last point is and edge case we can skip if it's difficult), and having the current icon, on a nix system that has explanations for 5 year olds?

I would love to help but being an amateur I've encountered a few pain points, unless the issue is my search skills are lacking. A few days ago I finally updated my Firefox version on Debian 9 to the developer version 56 (previous 54). The problem has been finding documentation on both the Debian and Mozilla websites on how to do it the right way. Found a decent tutorial, made a back up, installed and my machine is working much better. It is old and got little memory :)

Then FF said there was an update. I got stumped. From what I could find FF sync has a limit of X bytes. I ended up looking into making a web extension but that's going to take me some time. It seems I could save the open tabs using Tabs.tab from the API, maybe. Sorry, I digress.

In short, how can I have two FF versions on at the same time, different directories obviously, with their respective icons, while manually updating them? I guess it would be to delete all the files in the dir, unpacking, and call it a day? Still need help with the two versions though.

If you read all of this, thank you.


I'm not sure exactly what you're asking, but try just using the same profile for both installs. It contains all the information about your browsing session including tabs, bookmarks, extensions, cookies, etc.

https://support.mozilla.org/en-US/kb/profile-manager-create-...


His point is concurrent usage of one profile. FF does not support this.

Queue the FF is not tread safe jokes (and only that, not a flame).

OP: you can use Firefox Sync. There is a cloud version by Mozilla but also a WSGI app you could host. I know it is not a 100% solution, but what was recommended to me when I asked during FF alpha and beta testing with stable way back when.


Sorry if I explained myself incorrectly. My main point isn't concurrent usage of profiles.

I lack the necesary knowledge on how to correctly install two different FF versions and have them accesible with their respective icons. In my case, FF dev version and the one discussed in the article to help out with bug testing/development.

The issue is since I'm on Debian, auto update does not work. From my limited understanding, I would need to install FF 57 in its own directory. For this one I don't need a profile. But I don't know how to install it and have it accesible either by command line or the FF icon that comes with the browser since I already have my main FF I want to keep.

The second part is regarding the profile, sync with the other browser I recently installed.... I was about to go down a mental rabbit hole :)

The problem is, the lack of noob friendly documentation. I know I could figure it out if I research the issue through different topics on Linux administration. I was hoping that maybe someone knew of a tutorial which talked about the different steps on how to install Firefox on a nix system, explaining the different concepts as if I were a five year old. The best I've managed to find is a tutorial laying out the steps. But it didn't explain the concepts behind those steps.

That tutorial mentioned how to replace debians FF from the repo. But whenever a new version comes out I would need to delete it and unpack the new one, linking it to the old icon.

I hope it's clear now. Thanks for the reply.


I use Ubuntu, which should be pretty similar to Debian as far as this is concerned:

I literally just downloaded the zip file (or maybe a .deb?) and unzipped it in a directory in my desktop. All I had to do after that was make sure I didn't have both browsers open, and everything worked properly. They both used the same profile, and the tabs I opened in nightly were available in stable and vice versa.

I did nothing more than unzip the browser, and everything works perfectly.


I'll give that a try. How did you link the icon? Or was that automatic too?


Maybe this is what you're looking for? (I just installed Nightly on my Ubuntu 17.10 install and wanted separate icons myself).

https://www.reddit.com/r/firefox/comments/6tt7wj/nightly_cre...


Seems to be about right. Thanks!


I didn't link the icon, I just ran the script (I use gnome-do and ran it from there).


Thanks. I'll read this in a bit. When I was looking into different options I did find a mention about the profile, but the doc/tutorial seemed to be old and couldn't find it in the dir when I looked for it.


> update & save the open tabs without making bookmarks

Not sure if you are looking something like this, but OneTab extension it's pretty useful to me. https://addons.mozilla.org/en-US/firefox/addon/onetab/


Maybe. I'll have to take a deeper look into this. It seems I could upload the tab list and retrieve them later. Thanks.


I won't be happy if it were to fail but certainly I won't despair either. I don't want amazing performance improvements at the cost of huge functionality losses. This update is simple a regression: almost every add-on I have been using will stop working and has a stupidly limited or no replacement at all. There will really be no reason for me to still use Firefox over any other browser now.

The only option seems to switch to the ESR version, at least for a while it will work.


This sums up my feelings re. Firefox 57: https://www.change.org/p/mozilla-save-mozilla-firefox-s-best....

Sorry Mozilla - for the reasons expressed above, 56 will be the last version of your browser I'm able to effectively use given the demands of my browser workflow. If my use-case no longer fits your business case, then we will part ways.

Currently, it looks like my best options are either Waterfox (with its pledge of continued XUL extension support), or Vivaldi.


Vivaldi has less powerful extensions than Firefox 57.

I can't imagine Waterfox or Pale Moon being able to keep XUL extensions alive for much longer. It was a major maintenance burden for Mozilla, so it's pretty much an impossible task for the comparatively tiny developer teams behind those.

Having said that, Firefox 52 ESR is still supported until June 26, 2018. So, you could use that. And the Waterfox + Pale Moon devs could base themselves on that until its EOL.


Vivaldi provides arguably the most customizable UI of any browser out of the box (see: ability to move tabs to any side of the screen), plus their culture seems to me to be all about customization and providing options. I like that.

Firefox ESR is pointless - it's just delaying the inevitable. I tried Midori, but it runs poorly on KDE. Palemoon is pretty bad - I tried it and it doesn't properly support existing XUL extensions that work fine in Firefox, so it's a no-go in my book. Waterfox, on the other hand, is fast, works exactly like Firefox, and is at least willing to give continued XUL support a shot.

So, it's Vivaldi or Waterfox as best I can figure at this point.


57 nightly feels extremely snappy. Scrolling is no longer janky. The theme is good and sensible (at least on Windows). I am seriously going to try to ditch chrome for this. Yay! The old Firefox is back!


I just updated FF Nightly on Android and the new Photon UI is very big, very white and very obtrusive feeling in comparison to before I updated. I've got a pretty small android phone, I'm sure it looks better on the phablets of the world, but I ask for a dark UI option and small screen footprint please.

Edit: Found some images http://www.androidpolice.com/2017/08/08/mozillas-new-photon-... I just noticed, it recolors my system stsusbar too. So, it's white on white while FF is open. Everywhere else it's black.


Recoloring the statusbar is something most apps do nowadays, as part of material design[0]. They introduced the API[1] for it in Android Lollipop.

[0] https://material.io/guidelines/style/color.html#color-color-...

[1] https://developer.android.com/reference/android/view/Window....


The new look screams "generic". You can say what you want about the old design, but you could always tell that someone was using FF Mobile because of the distinct black rounded background on the buttons. The new design looks like literally every other mobile browser in existence.


On the other hand, the most comment complaint I've seen in the wild about FF for Android is "the UI looks weird / doesn't fit in with the rest of Android". You can't please everyone.


This design is indistinguishable from any other mobile browser. The old design was a bit dated but it was distinct at least.


Comment on and/or CC yourself to bug 1195138 [0] which is this.

[0] https://bugzilla.mozilla.org/show_bug.cgi?id=1195138


I think it uses the same number of pixels though, so the only issue would be a dark UI option (which I agree would be nice, especially in the evening)


I've been doing more browsing in private mode because of the dark bars

Do hope there's an option to change it soon


To counter all the negativity in the comments: I've been exclusively nightly-only for a year now, and I've only seen improvements. Most addons I use have replacements/WebExt versions, and the UI and performance are much better.


Most comments here seem to be the result of an "How you can complain about Firefox 57" article posted somewhere else.


Mozilla can contribute to Firefox 57's success by finishing all the missing WebExtensions APIs. So many extensions are going to be impossible to recreate.


Help us! Write patches to dependencies of this bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1215059


Thanks for that link, now I can clearly see that from the power-user perspective WebExtensions is a dead-end. If you look at the RESOLVED WONTFIX issues linked in that bug, you will find a lot of them are rejected based on political reasons. For example: https://bugzilla.mozilla.org/show_bug.cgi?id=1246706

Listen, I don't care whether features that I need don't fit the WebExtensions security model or whether they are confusing to a clueless user. If you can't support these features as part of WebExtensions, please design a separate set of APIs. Fine with me if you market extensions targeting these lower-level APIs as "dangerous, use at your own risk".


Lovely blog post. Well written, thorough and clear. Getting people involved in a large project is not an easy feat, and I think this may open the eyes of people not knowing were to start.

Well done!


You can grab a copy of Firefox 57 Nightly in portable form so it won't affect any local stuff from PortableApps.com here: https://portableapps.com/apps/internet/firefox-portable-nigh...

It's a great way to test out the features in a lower risk fashion. You can do the same with Beta, Dev, and ESR channels as well.


I'm still surprised that they're spending time in a VR API while lacking many basic things like print preview functionality. Also In the last year every time I give firefox a try they dev tools are very buggy. Instead of improving the dev tools they waste time in Firefox for devs (which is funny because as web developer I want to develop in the same browser that my users, but instead Firefox for devs is a version ahead, so once I thought my website worked okay in Firefox but it really didn't because it didn't work well in the stable version of Firefox). I feel like the browser is getting faster though which is good. I also like Mozilla as an organization but I feel that they still lack the direction and focus to make Firefox better than Chrome. To be fair, I think Firefox for Android is better than Chrome for Android. And Firefox Focus it's neat too!


Um... Firefox has print preview and has had it for a long time.

It's not enabled on Mac, where there is a native print preview available from the print dialog: you can fiddle with the settings, then "Open PDF in Preview". Is it possible you're on Mac?


There's an XUL print preview plugin. :-)


Good to know, thanks! I'm just still astonish that after so many years they haven't implemented that functionality. I don't print stuff all the time but sometimes you want to save things as pdf and fiddle a little bit margins/etc.


It would be nice if I didn't have to abandon my privacy - reveal my personal extensions to some third party server - in order to use my own extensions without too much hassle.

Now, I have to patch omni.ja every single time I get update to Firefox from Arch Linux, to disable signature verification. I can automate it a bit, although it's not a robust solution, because I'm patching some JS code that can change at any time.

Beta is not an option, because I don't want it for regular use and it is not in Arch Linux repositories anyway, so I can't get it either.

I don't get it. It's still possible to disable signature verification with little effort (one line change) programmatically from outside of Firefox, yet power users have to jump through stupid hoops in order to continue using their browser the way they want.

I wish fight against malware/malicious extensions was not also inconveniencing power users.


Wow. What a difference.

Congratulations. This is definitely a huge step forward. The UI finally looks as good as the other guys, and it's fast, too!

If you had finally figured out that private mode's UI is too similar to the normal one, too, I'd even be amazed (let me help you: make the address bar and tabs dark instead of light).


So I downloaded Nightly on a lark. 4 addons disabled, with a link to this page:

https://support.mozilla.org/en-US/kb/firefox-add-technology-...

>You can still personalize Firefox with add-ons the same way you do now, except they won’t break in new Firefox releases.

This is a bald faced fucking lie. And I don't mean that in the "sorta misleading" sense, I mean that somebody wrote something that they knew was false and 'outta be slapped for it.

57 is getting pushed out the door with missing functionality, and that is an absolute fact. You absolutely "can not still personalize Firefox the same way you do now", because you're taking that away from everyone without having the replacements ready!

I can sorta understand the reason to kill XUL/XPCOM, but what I can't understand is this desire to drop it on the world come November before all the bits are in place yet.

Guys, Firefox isn't going to turn into a pumpkin if you let it cook for another few months.

On another note, all companies have a resource that they may not be aware of. It's called "good will". Mozilla is burning prodigious amounts of theirs:

    * Between the CEO drama,
    * the UI rewrite that was apparently received so badly it's getting reverted 4 years later
    * the forced addon signing,
    * the telemetry dark patterns, 
    * the forced addon breakage... 
    * the junk like Pocket, Hello, and VR APIs recieving nontrivial amounts of developer attention while stuff people actually use like print preview and SSL is either ignored or have years old debilitating bugs
Combined with the demonstrated "fuck you, we know better than you" attitude that permeated all of these decisions (I really think Mozilla and GNOME are going to wind up under the same umbrella at some point)...

And what to show for it? Pissed off users, and an also-ran niche browser with almost no tangible competitive advantage (and in fact, a mounting number of disadvantages)

I've got no faith in this company anymore. Eventually, there is going to be a "Mozilla foundation shutting down" topic at the top of this site, and the people making the decisions are going to be completely blindsided as to why.


Thank you, good summarization of the unfortunate drama around Mozilla.

What ever is going on in their management, they give a shit about the Firefox and Thunderbird community. The last 5 years they did so many stupid things. It's time to close Mozilla corp and turn it back to a real open source community project. It's time to fork, like Firefox forked of Mozilla Suite. I have no faith that Mozilla going anywhere, in a few months we will read a post mortem of how the managed to loose 30% world wide browser market share in 2 years.


This includes a call for help on Photon, the "new user interface."

"Since the UI is written in major parts in JavaScript and CSS and an HTML-like language called XUL, anyone with webdev skills can improve it!"

I thought 57 was dumping XUL completely. Wasn't that a major point of 57?


It's dumping the so-called "XUL/XPCOM extensions", so that's an extension API. The UI toolkit XUL, they'll continue to use.

There are some long-term investigations into using HTML/JS/CSS for the UI [1], but for the moment, the performance difference is still too much.

[1]: https://github.com/browserhtml/browserhtml


Why is HTML slower than XUL? I thought XUL was pretty much just a weird version of HTML, possibly with more native toolkit support. Is it because XUL is being used from C++ rather than JS?

(I checked the browser.html issue tracker but most of their perf bugs seemed to be resolved.)


I remember, for instance, that XUL was highly optimized to not create DOM nodes for huge/infinite lists/trees/... Last time I heard about someone attempting to implement these optimizations with comparable APIs in HTML5, the results were ok, but not nearly as good.

That was a few years ago, results may be obsolete, of course.


Ha, the xul tree and it's virtual view...

Someone (working now at Apple) wrote a super fast list component in HTML for Firefox OS so this can definitely be done.


Not sure which APIs you're referring to, but the Vivaldi browser UI is largely rendered in HTML5 and its performance feels native to me.


It may also be that XUL is more restrictive or less forgiving. By not having to handle as many edge cases they can skip a lot of code and speed things up.

This is just a guess on my part.


I suspect it's because the XUL used in Firefox and later is not "pure".

I seem to recall that one of the big "optimizations" that FF did compared to the mozilla suite (these days known as Seamonkey) was to translate a bunch of UI elements into native equivalents rather than drawing it all from XUL.

Thus replacing that with HTML may well be closer to reverting to the pre-FF XUL way.


The eventually goal is to fully dump XUL but that's going to take time to finish. The major point of 57 is legacy addons will no longer be supported.

Think about it this way: the transition away from XUL means the UI code (and more) is going to be very unstable for awhile. Thus addon authors would need to be continually rewriting their code for each update to keep up with the changes. It's better to bite the bullet now and switch to a more stable api.


Are the new APIs stable yet? I didn't think they have all the features required for some addons.

There are some people (like me) who use firefox because of addons that can't be supported on other browsers due to the lack of API support, so it's a bit disappointing if they're going to pull the rug from under those addons. If they do, they might see people leave for another browser.

For reference, the addon in question for me is Tabs Mix Plus.


All new APIs taken together, no, won't be stable for at least half a year still, very likely longer. They're still very much in the process of designing and implementing new APIs.

But the core APIs are the same as Chrome's. The implementation might not yet always be perfect, but the APIs themselves are stable. So, for the vast majority of add-on developers, the experience will already be much less burdensome. Or it will be in the future.

Yes, some add-ons will not happen or might only be possible long after the release of Firefox 57. But they're not going back on that plan. They haven't just forbidden these extensions for funsies. They've very much already used the opportunity to rip out tons of legacy code and refactor things that they could have never refactored before without breaking tons of add-ons each time. At this point, there's very few legacy extensions left that still function in Firefox Nightly, even if you tell it to load them.

Not to mention that this has helped them to speed Firefox up by a significant factor. All the performance improvements since Firefox 48 would not have been made without this deprecation on the horizon. And considering that a significant chunk of Firefox users use no extension or just the obligatory ad blocker (or any set of extensions that's going to be easily replaceable with WebExtensions), this is also very much the decision that has to be made when asking yourself what's best for users.

As for people possibly leaving Firefox when their favorite extensions stop working, yes, that's possible, but Firefox's market share has been growing again since Firefox 48, and there's also just the fact that Firefox is still going to be the browser with the most powerful extension API.

You're not getting more extensions by switching. You'd have to be fine with even less extensions in exchange for maybe a particular feature that you like in another browser.


This post is an excellent example of the kind of arrgoance from mozilla that upsets people.

You have basicly told us all "your problems dont exist, dont matter, or we dont care, deal with it"

I'm done with firefox post 56.


I have never said anything along those lines. Your problems do exist, I very much also am sad to see certain add-ons go. All I said is that in my opinion, and evidentally in Mozilla's opinion, your and my problem with add-ons going away, don't weigh up against the problems of 95+% of the user base.

I have no problem with you being sad about these add-ons going away. I do have a problem with you acting like what Mozilla does is completely unreasonable and objectively bad for users.


What makes you think he's a Mozilla dev?


Exactly. That's why I left Firefox also.

And due to jilted lover syndrome, I feel compelled to post in threads like this one saying so. Heh.


They are not transitioning away from XUL


Are they not ? I thought that was the long term goal.

https://github.com/browserhtml/browserhtml


Source? Last I heard from Moz, XUL was considered a maintenance burden that increasingly suffers from neglect. HTML now does a comparable job, is standardised, is widely used and needs to be implemented anyway.

Maintaining two interface languages doesn't make much sense in the long term.


The XUL/XPCOM extension API is a massive maintenance burden. The XUL UI toolkit is something different and while it's not either magically free of maintenance cost, it's much less of a maintenance burden than the XUL/XPCOM extension API.


XUL is the same XUL both in extensions and in browser ui.

XPCOM is another independent technology https://developer.mozilla.org/en-US/docs/Mozilla/Tech/XPCOM.

They are disabling the direct access from extensions to apis used by browser itself (that is access to XUL _and_ XPCOM), to be able to modify the browser ui without worrying about breaking extensions, with the goal to remove XUL eventually.


This is correct. Or, at least, people would like to remove XUL eventually: the tractability of that is still a topic of research.


The Browser Architecture team at Mozilla [0] is investigating stripping XUL(/XBL) out of Firefox, among other things, but it's more complex than you might expect.

[0] https://github.com/mozilla/firefox-browser-architecture


Surely Photon would use HTML, given it's new code?


Nope. It uses XUL. It is in some parts new code, but not completely. If you were using Firefox Nightly, you could actually see them slowly changing Australis over to Photon.


XUL is still there, and there are no immediate plans to get rid of it. Mozilla has just decided that you dont get to use it anymore, the deadline is purely an administrative decision.

Its great to see they are going to drive firefox off a cliff with yet another anti-user decision.

The situation with popular extensions has improved, but there is going to be a lot of pain, especially with XUL functionality that they refuse to implement in WebExtensions.

https://arewewebextensionsyet.com/


The question is not about "XUL functionality" specifically, XUL is just an implementation detail, and html could work in its place just as well.

The question is how much the extensions can do, in the old model they could do anything firefox code could do, in WebExtensions model, they can do only very limited set of things.

On one hand it is quite sad to see firefox extensions go, on the other hand the old model was a maintenance hell both for firefox developers and for addon developers, because smallest changes in firefox ui were breaking bunch of addons, and even changes in addons themselves were breaking other addons, which all had incompatibilities and had to do lots of subtle hacks to work together.

I have worked on firefox addons for several years, and had to quit firefox when they started ignoring addons and investing in old addon sdk. But saying Mozilla is intentionally killing firefox is not fair the whole thing was shaky and full of hacks and would crumble by itself too.

What they should do now, is splitting their ui from the engine, so that advanced users can hack on the ui and get all the benefits of old addon system without the mess. If they fail to do that brave browser with https://github.com/brave/muon will take that niche.


The UI toolkit "XUL" is something different from the extension API "XUL/XPCOM".

And no, it's certainly not just an administrative deadline. For now, you can still tell Firefox Nightly to install legacy extensions, the vast majority of them are just by now completely broken, because Mozilla is ripping out all of that legacy code and refactoring what's left. Here's a list of all of the things that they can get rid of: https://bugzilla.mozilla.org/show_bug.cgi?id=1347507


> XUL is still there, and there are no immediate plans to get rid of it

It's true it's still there, but I think the plan is using it less and less. No new UIs are written in it and many things have been rewritten from XUL to HTML5. Eventually it could be removed entirely.


longer term that is sensible sure, but the ecosystem is not ready for it.

Its amusing to me that mozilla expects everyone else to be moved off XUL, but not their own code because it has legitimate reasons....


They are moving everyone else off so they themselves can move off XUL.


It is my understanding that getting rid of XUL in favor of a less peculiar way to do UI is a necessary step to improve security and performance and get rid of a lot of legacy ballast that keeps Firefox from getting better. So there are some upsides to the transition. Time will tell whether it will be a success or whether it will "drive Firefox off a cliff". I don't think you can say a priori what the result will be.


I think you can say a priori, that a major remaining reason to use firefox over chrome is about to be eliminated.

Hilariously, dispite the pro-privacy stance mozilla pretends at, they want to start tracking urls visited

https://news.ycombinator.com/item?id=15071492

Why keep using firefox? Because I love Mozilla so much when they clearly have zero interest in my needs as a user?


So the devtools were rewritten from scratch? Are they fast now? If they are I'll jump in.


IME pretty fast.

You can even contribute to them without a Firefox clone, they're an entirely separate repo that contains a webpage that can be run in both Firefox or Chrome as a regular website and used to debug both. It's fun!



That's a good idea.


I would say the FF devtools are almost there. And I prefer using them. But there are small things here and there and keep bringing me back to Chrome.

Example of a network timeline. Chrome: https://i.imgur.com/16ZLENd.png Firefox: https://i.imgur.com/1j0NZcd.png

I get a better representation of the timeline waterfall in Chrome.

Like I said, the tools are almost there.


Honestly, I can't believe how tone-deaf the entire organization has been on this. I know that's not popular to hear, but I keep seeing people tell them, over and over again, that the loss of customizability and functionality -- and the repeated elimination and crafting of design and UX to bring it into a mirror-image of Chrome UX -- is not what their users want. Yet every person on the Firefox team that I've seen actually respond has insisted, in the face of this overwhelming feedback, that it's necessary and desirable. That's fine, but I think they're going to get a rude awakening afterwards. It'll be wonderful if I have to eat my words on this, but I don't think so. Firefox is really just becoming a mirror image of Chrome, and if they're not, then they're not doing a great job on understanding why people valued -- and, yes, that's purposefully past-tense -- Firefox, and they're doing a very poor job on distinguishing why people should choose to stay with Firefox over Chrome.


Sure, we CAN contribute. But what's in it for us?

Are we going to get a browser that caters to our own needs? No, evidently power users are no longer the target demographic.

Are we going to make a browser that we can recommend to our nontechie friends? No, I don't trust them to navigate all the opt-outs and dark patterns around your telemetry. I don't even trust myself to never misclick.

Is contributing going to win us goodwill from our collegues? No, you've alienated them too.

Is this about ideology, then? Are we building a browser for a better world? ..it would be much more convincing if you guys didn't fire your CEO over political speech.

And that's where we're at right now. Maybe the 97% or whatever non-addon-using demographic in your telemetry data will make up the shortfall in contributions.


Does anyone know how can I stop the menu from vanishing (very slowly) and reappearing when I hover the top of the window? I just want it to stay there.


Right, that's a fullscreen thing only. I was getting it because i3wm puts Firefox in fullscreen mode when I do <Mod>+f.

Just remembered that about:config exists and disabled the vanishing menu.


Right click on the toolbar, disable "Hide Toolbars".


How can we best contribute to the total removal and scrapping of WebExtensions in favor of a more robust XUL development environment? How can we contribute to Firefox not becoming another meaningless Chrome clone like Opera or Vivaldi? Because that's what I think would be the best success of Firefox, long-term.


I know you're sarcastic, but I'll try to answer anyway :)

> How can we best contribute to the total removal and scrapping of WebExtensions in favor of a more robust XUL development environment?

I don't know if that would be possible. We're moving a lot of XUL around these days and we'll be moving even more, hopefully to HTML soon.

You'd have to gain trust in the community as a reliable person willing to maintain and keep relentlessly updating your addons APIs as the UI of Firefox moves underneath it.

> How can we contribute to Firefox not becoming another meaningless Chrome clone like Opera or Vivaldi?

I assume by "another clone" you refer to the fact that we're switching to WebExtensions which is a system shared with other browsers. In that case, pick any of the bugs here: https://bugzilla.mozilla.org/show_bug.cgi?id=1215059 and help us extend the API surface to support more features :)

> Because that's what I think would be the best success of Firefox, long-term.

I agree! We already support more APIs in WebExtensions than any other engine as far as I know. And we only started.


I'll be blunt. When you break compatibility with Classic Theme Restorer, Status-4-Evar, and QuickNote, I'm switching over, 100%, to Pale Moon. Between Australis and WebExtensions, Firefox has been moving farther and farther away from a browser I want to use. It's really getting to a point where I'm happier using Lynx.


Mozilla has lost any good will they had with me, I'm actually dropping Firefox this year. The entire reason I switched to it was tab groups, which isn't getting ported over, so I'll be switching back to Chrome.



Due to breaking XUL extensions with no adequate replacements, mouse gestures in particular, I recently switched from Firefox to Vivaldi.

After thirteen loyal years with Firefox as my primary browser they finally forced me away. Bye, Mozilla.


https://addons.mozilla.org/en-US/firefox/addon/foxy-gestures...

Works great on my nightly installation, but have fun with Vivaldi.


And yet, no Yubikey support, that they were hoping will land in November in stable.



I had good luck with the add-on [1], but it looks like it will break in 57.

[1]: https://addons.mozilla.org/en-US/firefox/addon/u2f-support-a...


So, one stable release without something like a two-factor authentication method. Since the addon will break and no official support will be released on time, there's no other choice but to use Chrome/Chromium to use Yubikeys.


Or, you know, just use the ESR release for a few weeks.


So does that mean Mozilla don't use 2FA?


2FA is actually globally enforced for all internal Mozilla accounts, IIRC---it's just not done over U2F. I'm not sure whether the details are public so I can't elaborate further.


It'd be easier to test and keep current with what Mozilla wants tested if they had a delta update for the nightly rather than a monolithic binary. For example if they jumped on board with delivering a flatpak version.


If you download nightly from nightly.mozilla.org, we'll keep offering you delta binary updates twice per day! :)


For people that need legacy addons like I needed LastPass, go to about:config and toggle extensions.legacy.enabled to true


first AMD with Ryzen and now FF. Working with nightly in the past few days and don't miss a thing. Of course not all plugins are ported yet, but ublock works already. keep up the phenomal work.


Is OpenSearch support planned? Or is there an extension that does implement it?


Oh, nevermind, there's a different kind of OpenSearch support that seems very nice. I think I'll stay with Firefox this time.


Two words: Tab groups.



It has been a while since I last tried compiling Firefox, so I don't know if things have improved. In some of my previous setup and build attempts I ended up giving up out of frustration. Just updating the docs would probably be of huge help to many users.


I'll add a +1 to this very issue. I likewise am not sure if it's still the case, but a few years ago I was looking into contributing some ES2015 implementations to the FF JS engine but couldn't get Firefox to compile based on the docs. I get that software development involves some amount of grit to grind out a problem and solve it, but to have to do so right at the beginning of possibly contributing to an open source project was asking just a bit too much of my free time and I bailed and have never made the attempt again.


Building SpiderMonkey is not hard at all today: https://developer.mozilla.org/en-US/docs/Mozilla/Projects/Sp...

The only issue is that you have to clone whole mozilla-central even if you only work with small part of it. But once you pull it, the building and hacking on it is quite easy :)


Fwiw if you go to their irc server and go into the introduction channel you can get help on stuff like this. The artifact builds are also much quicker to download and compile too if you don't need to change the c++ core.


Firefox does try to make it easy to build. However they're quite opinionated and want you to use a "bootstrap" script which downloads some stuff into ~/.mozbuild regardless of you potentially already having it.

It is however still very close to a clone & bootstrap & build process, at least on Linux and Mac - i.e. one of the better projects. (For comparison, I recently tried to build Facebook's Infer, and that was even worse than building OpenOffice back in the day.)


Firefox now has artifact builds, so if you're not contributing to the C++ portions (the UI or whatever) it will just download build artifacts for you.

If you do want to tweak the native code it should just be ./mach boostrap + ./mach build, but it's entirely possible you'll have trouble with that.


More and more it feels that self-compiling is not a considered usage for large projects. Instead if you compile at all, you do so as part of a distro maintenance team. And thus should be on all the relevant mailing lists and irc channels, and keeping notes.

At least that is how it feels to me after having spent some years using a distro that is a fair bit more hands on than most.


This is definitely relatable. I've lost interest in many projects after wasting hours trying to get things to compile.

The worst offender I've encountered has been android. The initial checkout is like +100GB, and it has incredibly poor fault tolerance. Based on my experience, it fails regularly at random points, requiring you to re-fetch everything. Despite multi-day efforts, I've never managed to build a ROM for my Nexus 5X or Pixel.


What about development? You need to be able to compile to test feature implementation and bug fixes.


Nightly automated builds...


I've always had trouble compiling large projects. The only one I got working with no issues along the way was ReactOS, which (in a twist of fate) will not support Firefox 57.


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: