This makes me curious - does Mozilla's new Tab Center experiment use Web Extensions? Or are they refusing to take their own medicine, and using inaccessible internal APIs? https://testpilot.firefox.com/experiments/tab-center (Edit: Deprecated, not inaccessible.)
There is a rough version [0] based on the sidebarAction API [1], this is probably going to improved when it actually lands in Firefox Nightly, which should hopefully be soon.
It uses very inaccessible internal APIs currently (including XUL), but there is an effort to use it to create the necessary APIs so it can be ported to a Web Extension.
Emacs can already embed a webkit widget, but much more work would be needed to make interaction with the widget "emacsy". At the moment all you can do with it is send JavaScript snippets to it to make it change internal state. It would be great to intertwine this more with Emacs so that e.g. completing-read could be used on the DOM.
It would be rather fun to download a webpage in "BrowserEmacs" and have a REPL available with the page turned into a S Expression for you to manipulate at your or your plug-ins leisure. Throw in the tools to make plug-ins have there own UI and you have a developer's browser.
Ubiquity https://en.wikipedia.org/wiki/Ubiquity_(Firefox) (now defunct) had natural-language commands with localization including subject/verb/object order. It supported subscribing to commands too, so you could keep up with changing APIs automatically. It even understood pronouns, so you could write "email this to Dave" and it would email your selected text to your gmail contact named Dave. The only thing it was missing was a permissions model to limit what untrusted command code could do.
I've been praying that Mozilla would rethink this, but that looks increasingly unlikely. I guess switching to a LTS release will tide me over for a time, but this really does feel like the end of an era.
This was my thinking as well. I'll probably move to the last ESR ('extended support release') before the cutoff period, just so I can keep using Tab Groups.
Forking brave browser [1], or one of other electron based browsers may be a good way forward, as it will allow easy customizability of the ui, and get all the hard to maintain native parts from chromium.
Mozillas browser.html project may be helpful too, but it seems to be a bit too experimental for now.
No need to fork Brave, we will collaborate on extensions APIs as we scale. We're supporting tested chromium extensions that we curate and fetch from our own S3 (essentially a cache backed by the Chrome Web Store).
For future extension APIs that go beyond the chromium ones, we won't support XUL, but I'd like to take advantage of our Muon (https://github.com/brave/muon - we forked Electron, because security) React.js-based front end.
In 2Q2017 I'll hope to have more to say about extensions and where we'll go. Interested extension developers are welcome now to join our https://community.brave.com/ discourse. Thanks.
Great to hear you are interested in colaborating on this issues. muon and react based ui is exactly why i was suggesting to use brave as a base for experimentation.
I was not talking about the type of extension that allow to do some common things based on small api created specifically for that.
I am more interested in extensions that get full access to existing react components, and can modify the browser ui in arbitrary ways, e.g. by adding splitview similar to what cloudy has, or adding proper tabs on mobile instead of carousel.
I was chief architect of Mozilla back in the 1998 October roadmap period and onward, when we created XUL. Indeed we wanted powerful extensions, and got them. Problem is they tied Mozilla down for too long, without anyone stepping up to find a way forward that didn't just seem to signal "XUL going away... <angry dev feedback>. Oh never mind; or maybe later".
Brave UX will evolve but slowly and without big changes that break too many users (Australis). Still, we need to be able to change our UX. So if we have powerful extension APIs into that UX, we can't promise never to break extensions that use those APIs in ways that become unsupportable.
All we can do is try to co-evolve nicely and cooperate better. I don't see a silver bullet here, but I do see how Mozilla has burned a lot of add-on developer good will -- and we won't make that mistake at Brave.
In principle, yes. But maintaining a browser is, as a understand it, a LOT of work. So it would take a sizable contingent of volunteers to take on a project like that and keep the forked version current.
The real concern is that browser is a hugely complex piece of software with lots of security, stability, and compatibility concerns. You could fork it, but now you have to constantly ensure you maintain all those things between your fork and the mainline. That will become increasingly difficult as they diverge more and more.
Has pale moon ever moved past their "windows first, foremost (and practically only)" attitude?
I seriously looked at it when australis happened before classic theme restorer was stable, but at the time they seemed to have no interest in anything but desktop windows. I suspect a single-platform browser (especially when the platform is windows) is a non-starter for most people.
I wonder how vetting of Certificate Authorities will happen. I suppose that if your OS has a sufficient number of trusted CAs, then any browser can rely on those.
But consider this, from Debian's ca-certificates package:
---
This package includes PEM files of CA certificates to allow SSL-based applications to check for the authenticity of SSL connections.
It includes, among others, certificate authorities used by the Debian infrastructure and those shipped with Mozilla's browsers.
Please note that Debian can neither confirm nor deny whether the certificate authorities whose certificates are included in this package have in any way been audited for trustworthiness or RFC 3647 compliance. Full responsibility to assess them belongs to the local system administrator.
---
So as long as Mozilla allows OSes, and presumably other browser forks, to benefit from their CA vetting and monitoring activities ...
There really is a lot that does, or should, go into browser development and maintenance.
The Classic Theme Restorer page says "CTR will stop working on Firefox 57 (release) when Mozilla drops support for XUL / XPCOM / legacy add-ons, but it will still work on Firefox 52 ESR until Firefox 59 ESR replaces it in 2018 (~Q2)."
It sounds like Tree-style Tabs will stop working then as well. That will be a sad day. Tree-style Tabs and some form of mouse gestures are pretty much all I need.
I won't need either of those, but this is a major part of why I like Firefox, each person can make it their own.
These in particular make Firefox my own (your Firefox is not my firefox, but your firefox is OK), and I will be sad if they disappear without replacement:
It's just incredible that they're going to do away with XUL. Imagine if Google somehow decided to remove all the Android apps in the Play Store and told everybody to rewrite their apps or die. And what's even worse, most addon writers don't even receive money for that.
This is going to kill the addon community but I guess they don't care at all.
most of those will end up with webextension versions. I don't know what they all do, so i can't say all.
I do know that ones like ghostery and lastpass have financial incentives, while others (like vimfx) are already planning on it. Others don't seem to rely on many (if at all) XUL/XPCOM features
The only plugin I really care about is ublock origin. If that stays, I'll keep ff otherwise I'll go back to chrome. I suppose there's no way of knowing whether a given plug in is possible in future without having access to the source.
Chrome it is then. I hate every single change to the Firefox UI since Firefox 2. The only reason I can still stomach using Firefox is Classic Theme Restorer, oldbar, etc which make it look and behave like it used to.
It's really sad to see how far Mozilla has taken Firefox from its roots. Extensions are really the only reason I still continue to use Firefox and now Mozilla has set the date that I'll be moving on to Chrome into stone for me.
If this WebExtensions move makes Tab Mix Plus stop working (I use multi-row tabs extensively) then I will no longer have any reason to continue to use Firefox. I put up with it's poor performance and instability merely for this one feature.
Nobody needs a clone of Chrome; we already have Chrome.
I would use the last version of Firefox that supported Tab Mix Plus until it no longer worked with the web in general. At that point I don't know what I'd do.
Looks like ESR 52 will continue to support XUL addons, and get regular security updates, well past the release of version 57 which is the estimated version for the cutoff. We can "vote with our feet" somewhat. If a large number of users switch from 56 back to 52 ESR instead of "upgrading" to 57, we will have some actual numbers to show.
> I put up with it's poor performance and instability merely for this one feature.
Ah, but performance and stability is exactly what Mozilla is improving with WebExtensions, Electrolysis and Servo/Quantum, so you won't need to put up with the poor performance and instability any more! What's more, it will still respect your privacy more than Chrome, be open source unlike Chrome, and allow you to configure it more than you can configure Chrome.
Tree Style Tab for me. I remember reading that it is not a matter of someone porting it to chrome but that the chrome UI does not support such tab flexibility. Sad times..
Do you want a fork? Do you want to drive more people to Pale Moon? Do you want to become even less relevant relative to Chrome? Do you want to make the web even less user-centric?
A change like this is how you do that. Good job, Mozilla.
I'm only sticking to Firefox because there's a specific set of extensions I want to keep using. If only one of my key extensions breaks, my last reason to use Firefox vanishes, and I will immediately switch to Chromium/Iridium and never come back.
Same. I guess Mozilla's stubborn decision makers -- who will never admit being wrong -- can comfort themselves by thinking of all the people who decide to quit using Firefox, when their cherished extensions break, as idiots. It's so sad.
I am not familiar with Web Extensions, but I am familiar with the fairly common event in which application vendors severely change functionality, with significant downstream effects. Such behavior seems more common amongst open-source products. Consider Python 3. Asterisk's scripting language. Perl 6.
Incompatible version changes drive away users and developers. The uncertainty that comes along with a planned update in which compatibility is not guaranteed makes us all shy away from developing in that application at all, not just on the old version.
Microsoft always seemed to do a good job at maintaining compatibility over its versions. Most applications that used to run on Windows still do, although there are a lot of exceptions. Python's change to the print statement broke everything, though the fix was easy; most of the time the fix is not easy - and even when it is, nothing works until someone gets around to the fix.
I appreciate the need to move forward, and sometimes you have to break stuff to make it better, but these incompatible changes just destroy momentum. It's a shame to see it happen.
It's not merely an incompatible change, it's an incompatible change that removes significant functionality. People seem to be coming around to Python 3 eventually, as it does offer a much improved experience. I can't see that happening with subsequent versions of Firefox unfortunately.
Just today I was unable to access a modem configuration site (at http://192.168.1.1) with Chrome, because without internet it couldn't phone Google to decide wether to use it as an address or a search term.
>>Moz://a is paid millions by Google so that they're the default search engine in Firefox
That has not been the case for a few years(2012? or 2013 I believe), Yahoo is currently the Default Search and pays the bulk of Mozilla budget, With the Verizon Deal Mozilla is likely worried that will change when the contract is up for renewal
Yes Verizon must honor the Contract, all contracts, Yahoo committed to, that is standard
That contract however I believe was for 5 years, I am talking about what happens after the 5 years is over, which I believe will be in 2019 or 2020. After which Verizon has no contractual obligation to continue paying Mozilla for anything and likely will cease the partnership
That's not the point. There's a poison pill in the contract. Mozilla can walk out of the existing contract, get all of the money in lump sum, and sign with another search partner (while still being paid by Verizon). There are other search engines waiting in line, Mozilla's not married to Verizon.
>>There are other search engines waiting in line, Mozilla's not married to Verizon.
Who?? I know of none
MS will never support Firefox, They want people to use Edge.
DuckDuckGo do not have the cash to do it.
Google has the market share and does not need FF anymore that is why they dropped them
So Who....
>>That's not the point.
Actually it is the point, by 2020 Mozilla's IMO mozilla revenue stream will dry up, they need to find a way to survive with out the Search money. This is why they are taking these risky moves which IMO are doomed to fail.
Search competition is different from browser competition that mainstream media likes to hype up. A few hundred million for 10-15% of the search market is nothing, Bing and Google would gladly offer a bit of money. There are also partners around with world (Yandex, Baidu).
I believe that, early on, Perl 6 was intended to be the next version of Perl in a more traditional sense, with an upgrade path and compatibility mode and so on, but with the time lag before the 1.0 release and increasingly radical design changes, at this point Perl6 and Perl5 are different languages, which share a name for historical reasons.
This is going to kill pentadactyl, which is currently the only reason I'm using Firefox.
For those who aren't familiar, pentadactyl is a Firefox extension that transforms the user interface to make it more minimal and vim-like. You get to use standard vim bindings to navigate the web and it even gives you a vim-like command line interface to the browser's functionality.
After discovering cvim for Chrome, I switched and never looked back. Vimium, on the other hand, is a bit different from pentadactyl/vimperator experience.
Ditto! Unfortunately, it appears that there will be no replacement any time soon, if ever. I am not 100% certain on the details, but apparently, the new WebExtension APIs make certain browser window manipulation impossible. So looks like we're stuck with ESR in short-to-mid term future.
Sadly, I mean a replacement for Firefox altogether. The only plugins I use are Vimperator and uMatrix. I'm posting this from Qutebrowser and I'm very impressed. I just hope I can get uMatrix functionality.
Horrible news. I use Vimperator, used to use Pentadactyl. The latter forked from the former, but they are still very closely related. Like Emacs and XEmacs.
They are addictive. They turn Firefox into a different kind of thing. I rely so much on custom commands like:
Vimperator is one of two things that keep me on Firefox.
The other is not having a single process per tab. I typically have 40-50 tabs open at any one time. On chrome that chews through memory. On Firefox it chews through significantly less.
Unfortunately neither pentadactyl or vimperator support multiprocess firefox IIUC. The being said adding support for multiprocess is possible, whereas porting to a webextension doesn't appear to be.
Firefox has three APIs: XUL, Jetpack, and WebExtensions. XUL has been on the way out for years, and never worked in mobile Firefox anyway. (Almost nobody runs mobile Firefox, so that turned out not to matter much.)
Jetpack was introduced in 2013, and by 2014, it was actually usable. It's not clear when and whether Jetpack add-ons will stop working. Mozilla's developer documentation is so confused that you can't tell. WebExtensions and Jetpack are essentially the same functionality with different names.[1]
The trouble with Firefox chasing Google Chrome compatibility is that, once they get there, Google will probably change something so some Google service is essential to an add-on. Like Google did with Android. Then compatible add-ons stop working on Firefox. Nobody runs add-ons on Chrome much, anyway; I have the same add-on on both platforms and its 50:1 Firefox over Chrome. But usage is declining, along with Firefox.
> The trouble with Firefox chasing Google Chrome compatibility is that, once they get there, Google will probably change something so some Google service is essential to an add-on.
I'd be more concerned that, once they get to Chrome compatibility, they've simply sacrificed one of the key things that differentiates them from Chrome. Firefox's die-hard adherents use extensions heavily, and many of the most popular, particularly those that are on Firefox and not elsewhere, depend on the old XUL layer or other APIs that have no replacement.
I don't think users are going to see it as a great improvement that they have access to Chrome extensions they didn't want in the first place. Instead they're going to focus on the fact that they're losing the functionality that's been keeping them using Firefox.
I don't think you'll need Google to act maliciously or really do anything at all to have this end poorly for Mozilla.
the fact that they have no replacement for removed [and much-used] functionality seems completely nuts to me. their market share will plummet. i understand that old stacks must be removed at some point for security/performance/ergonomic reasons but leaving so many extension devs and users out in the cold with no alternatives is unthinkable.
i really believe they are signing their own death certificate. neither google nor yahoo will pay much for search traffic of a browser no one uses. i say this as a long-time fan, nightly firefox user and bug submitter (since v1.5). everything since Australis has been an unmitigated trainwreck :/. When Classic Theme Restorer [1] stops working, i'm sorry but i'm out....probably to https://github.com/Eloston/ungoogled-chromium for the time being.
That sub-9 digit number includes me though, hard to understate the importance of this fact.
(that said after reading tfa, if there's really no tab groups after next year i'll be really upset, especially after the whole "it's ok that we're removing it because now it's just an add-on" thing)
Most or all stats will show mobile FF browsing usage at about IE6 levels, and I don't think you'll find anyone caring much for the later market any longer.
We [web developers] owe a lot to FF, Firebug, etc, but the writings on the wall for mobile and desktop. Chrome is going to have to screw up really bad, and likewise for the base installs of IE/Edge and Safari. These days, I'm more likely to suggest Brave over FF, if someone really requires a Chrome alternative.
> These days, I'm more likely to suggest Brave over FF, if someone really requires a Chrome alternative.
Brave is just a reskinned Chrome.
> We [web developers] owe a lot to FF, Firebug, etc, but the writings on the wall for mobile and desktop.
Where I work developers have mostly switched to Firefox over the last few years. Firefox is just a better browser (faster, less bloated) under the hood. Yes, Firefox will have a difficult time sine they don't have their own proprietary walled-garden ecosystem as a distribution channel, but the technical product is solid.
Do you mean chromium? And it's much more than a skin, lol.
And it's not the experience for many to have a faster browsing experience, but I'm not going to argue how anyone quantities that -- really don't care what others use.
On desktop, I find the tooling much worse and sluggish. But the thread was about mobile, and there are a vanishing number of users of mobile FF.
If you [web developers] ignore mobile Firefox, you do all of us a disservice. Having a choice of browsers on mobile is important to me; Firefox is one way on Android you can avoid being fully locked in to Google's ecosystem and a crucial piece of the puzzle for anybody who wants a Google free Android device.
I understand business realities, and why Firefox mobile would rate a lower priority than many browsers. But I think competition in this space is incredibly important and I urge you not to ignore us [Firefox mobile users] completely.
That does not help if you're using Chrome as your main browser, there's no cross-synchronization. Which I do because it's just way faster, and especially rogue tabs can't shoot down the entire browser including its UI.
e10s might be the thing that could save FF, but I fear that it's way too late.
There actually are some cross-sync addons out there, IIRC. Or at least there were; I used to use one when I was still in the process of migrating from all-Chrome to all-Firefox.
Because Firefox is slow and pages are sometimes broken because nobody uses Firefox mobile. I also used Firefox Mobile for the sake of using uBlock Origin. But since I discovered Adguard for Android I can keep using Chrome (beta) which is way faster and it also supports web compression for http content (also speeds up).
Also use Mobile Firefox because Chrome is 'sticky' on android, 'oh you signed in by clicking the dialog when you where tired, let me sign you into everything else and re-enable sending your web history to us as well', I disable it then accidentally turn it back on and they don't have a 'I want to turn this feature of permanently' option its always 'suspend' not 'disable forever'.
So I switched to Firefox on Android, it's not 'quite' as smooth but Ublock Origin works great so on balance I like it better.
FF is slow, pages rarely load the first time and I constantly have to manually reload, the interface for making config changes are requires sooooo many clicks. I have no real experience if chrome is better because I never use it, but FF mobile has some real warts
most sites that are slow are also slow on chrome. so doesnt even bother.
the only solution is installing noScript. the only version available is in beta. but it works fine.
you will be suprised how few sites I have to whitelist to see the content I want. and usually besides faster, they get much more usable without Javascript!
Exhibit A: You can't tell the GMail app to use anything but Chrome (at least on iOS).
I'm sure there are more such examples. Even if a default was changeable the initial setting (as we saw in the Microsoft vs. Europe over IE) is very powerful.
and for whatever reason (i'll give benefit of doubt its a bug), every time i open a link it asks me which browser to use, no matter how many times i set the default to firefox.
i dunno, whenever i open a link it had radio button that says "use this to open links every time?" and i click yes then select firefox. but ya doesnt work so apparently you can't.
It's not "chasing chrome compatibility", it's sharing the same base.
Firefox wants to add its own APIs to webextensions. Chrome will too. They may not always match with each other with the newest APIs. It cuts both ways here, it's wrong to characterize it as something where only Chrome will be adding features and Firefox will play catch-up. Firefox is "copying Chrome" only as far as copying the base API.
Sad, but hardly unexpected. The truth is, the mozilla leadership has never had much vision. In the early days of Seamonkey (and then Firefox) there was a slavish devotion to chasing Internet Explorer, and far too much rationalization of feature selection based on "IE has it" or "but IE doesn't support that." Later, with the ascension of Chrome, Firefox began to try and mimic Chrome the same way.
Other than MathML support, I'm having a hard time thinking of anything truly unique about Firefox, as compared to Chrome these days.
> The truth is, the mozilla leadership has never had much vision
I'm not sure I agree there. I feel like they've had at least a few pretty strong points of vision. The failure has been more on the execution of producing and maintaining useful software in support of that vision.
Heh. Funny that I just updated to Firefox Developer Edition 53 and now Tree Style Tab (one of the addons that will go when XUL stuff does) has severe theming issues and weird flickers.
It'd be nice if a mod could change the title to "I cannot continue working on my Firefox add-ons"
Aw snap, Beyond Australis and its slim chrome is one of the best way (along with tree style tabs) to regain some vertical space with sub optimal resolution (namely, the 1368x768 of my x220t).
If Tree Style Tab doesn't work anymore. What are the reasons to use Firefox? Also what are the alternatives. I tried Chrome but the Tree Style Tab-like plugin is clumsy.
Should I just not update firefox and wait until devs realize tabs were meant to be on the side? Honnestly as a power user I can't browse the web without tree style tab anymore.
It is far far far far far way from replacing TreeStyle, as it has no Tree.... It just moves the tabs from horizontal to vertical, which has similar usability problems
It is better than Sidewise on chrome though, however TabCenter is XUL as well I believe, there are no APIS currently for it to be moved to Extensions,
To be fair, "showing tabs on the side" is still leaps and bounds better than "yet another browser that squeezes them into the top, end to end, like a god damn maniac". :)
But yes, the "tree" part of Tree Style Tabs is essential, too.
Firefox has literally been kicked by its userbase and Chrome to get into its current semi competitive state. One understands the need to evolve but change just for the sake of change and activity can just be spinning your wheels. We don't need another Chrome clone. Firefox and Chrome's objectives are completely different.
We definitely need more browser competition but growing complexity rules out one person or small team efforts like before all over the open source ecosystem.
Untill open source users can get their act together and find a way to fund development they will find projects increasingly less accountable to them than their corporate sponsors.
Once this happens I will leave Firefox after being one of its strongest supporters for years. What makes Firefox worth it is the level to which I can customize it and this will knock out 5 of my must-have add-ons in one shot. When you do that you have nothing but a Chrome clone left and why use a Chrome clone when you can use Chrome. I hate Chrome because I feel locked in on UI design (old-school user here). Once FF takes away my options guaranteed I'll abandon it.
If the only reason you hate Chrome is being locked into the UI, you should switch to Vivaldi. Sure, it's a Chrome clone, but the UI is super configurable (and as far as I'm aware, it doesn't do any opt-out user tracking).
Is Vivaldi free software (as in speech, not as in beer)? Last time I gave it a serious go, it didn't seem to be (and it was buggy anyway, but that was when it was first made publicly available, so I'm sure that's improved, right?).
> When you do that you have nothing but a Chrome clone left and why use a Chrome clone when you can use Chrome.
Why use Chrome when you can use a Chrome clone that better respects your privacy, is open source, and gives you more freedom to configure it? Because it's more secure or faster? It won't be for long: that's exactly what Mozilla is working on. At the expense, yes, of some configurability.
Security is debatable. Sand boxing was just introduced. Customizable maybe, but between the australis, xul deprecation, and the limited theming API I don't think Mozilla really cares.
Also for a open source browser. Are you aware the is chromium? Or hell any number of electron based web browsers (which is also chromium).
Edit and loss of configurability is not the issue. Loss of Functionality is.
I completely rely on Firegestures, the premiere mouse gesture addon.
Mouse gestures are in my muscle memory. I can't live without them, and every single Webextensions Chrome mouse gestures addon either:
a) Sucks. Doesn't work at all on internal pages including new pages (which is not a blocker, but sucks), and more importantly doesn't work _well_ ANYWHERE. Inconsistent behavior, poor performance. Just unacceptable.
b) Is actually literally malware. Once Chrome extensions get popular, a shady Chinese or Russian consortium buys them up, adds user tracking, and silently updates the extension so your every move is followed.
Chrome has been around for almost 10 years now, so I'm absolutely convinced that Webextensions are unable to support _GOOD_ mouse gestures.
Now that Firefox will shortly become hot stinking Chinatown garbage, my only hope will be Vivaldi, a browser using Blink (Chrome's backend). Its mouse gestures are really close to being usable-- it only lacks configurable mouse button chording.
I'm with you. I was hooked on mouse gestures after using Opera 12. They came built into the browser, and you could tell. Gestures worked everywhere-- on a new tab, on a loading page, even on the browser window outside a tab.
Now I'm using Firefox plus a dozen extensions to poorly replicate a base Opera 12 installation. The e10s update already broke some FireGesture in some minor ways (e.g. wheel gestures scroll the page content now). Other extensions stopped working completely.
I use Tab Groups constantly, I switched from Chrome to firefox almost entirely for Tab Groups. Runs terribly on my machine, I'll be back on Chrome the second Tab Groups are gone, unfortunately.
I've been worried since web extension only announcement we were heading for something like this.
It's sad, but I am seriously worried that this could be the end for firefox. Add to that the recent refocus/re-prioritization on firefox within mozilla, it could be a first death knell for the organization entirely.
I know in my own case, and it seems from comments here and other anecdotal experience, that a significant chunk of remaining firefox users depend on at least one extension that has no migration path, as there simply aren't equivalent APIs implemented or even planned. I really don't understand how they think they are going to maintain a userbase when they remove the last remaining feature that differentiates firefox from the other common browsers. Mozilla starts out in a tough competitive position, as they don't have an OS or platform to bundle a browser with. Alienating the users with the most reason to seek out an alternative seems to just be digging themselves deeper into the hole.
I don't expect much from the leadership who brought us FirefoxOS, and indeed the WebExtension migration is being mismanaged - BUT: the hard truth is that the Chrome extension ecosystem is already on top and growing, while the Firefox one has been waning for a while.
If you build a browser extension today, you build it for Chrome; if you have a Mac, you also get a Safari version almost for free; and then eventually, if you have time, if you can be bothered, you port it to FF. This could not go on, something had to be done. And they did it: JetPack/Webextensions, despite the uber-confused documentation, is an almost-complete compatibility layer that makes porting extensions from Chrome much easier. If they'd left it at that, nobody would have complained.
Unfortunately, the next step is to gut the old infrastructure. They might have their technical reasons for that, but it seems a suicidal move from a commercial perspective. They are throwing overboard almost 20 years of competitive differentiation and community engagement. It's like Microsoft told people to rewrite their .Net apps in ObjC.
217 comments
[ 3.6 ms ] story [ 311 ms ] thread[0] https://github.com/bwinton/TabCenter/blob/master/main.js
[1] https://developer.mozilla.org/en-US/Add-ons/SDK
[0] https://github.com/bwinton/TabCenter/tree/webext [1] https://bugzilla.mozilla.org/show_bug.cgi?id=1208596
Is the real concern about forking the userbase and ecosystem, and the fork won't have the critical user base to become self-sustaining?
Edit: added paragraph for clarity.
Now it looks like it will once again break (this time permanently).
I'm in the same boat with other folks. Why use Firefox if it's just a poor man's Chrome clone?
Mozillas browser.html project may be helpful too, but it seems to be a bit too experimental for now.
[1] https://brave.com
For an extension to get support, we need to rank it as in-demand, and then try adding it to Brave to see what missing chrome.* APIs it might need. https://twitter.com/bravesampson is leading this charge, using https://github.com/brave/browser-laptop/projects/1 to keep track and engage with developers.
For future extension APIs that go beyond the chromium ones, we won't support XUL, but I'd like to take advantage of our Muon (https://github.com/brave/muon - we forked Electron, because security) React.js-based front end.
In 2Q2017 I'll hope to have more to say about extensions and where we'll go. Interested extension developers are welcome now to join our https://community.brave.com/ discourse. Thanks.
I was not talking about the type of extension that allow to do some common things based on small api created specifically for that.
I am more interested in extensions that get full access to existing react components, and can modify the browser ui in arbitrary ways, e.g. by adding splitview similar to what cloudy has, or adding proper tabs on mobile instead of carousel.
Brave UX will evolve but slowly and without big changes that break too many users (Australis). Still, we need to be able to change our UX. So if we have powerful extension APIs into that UX, we can't promise never to break extensions that use those APIs in ways that become unsupportable.
All we can do is try to co-evolve nicely and cooperate better. I don't see a silver bullet here, but I do see how Mozilla has burned a lot of add-on developer good will -- and we won't make that mistake at Brave.
I seriously looked at it when australis happened before classic theme restorer was stable, but at the time they seemed to have no interest in anything but desktop windows. I suspect a single-platform browser (especially when the platform is windows) is a non-starter for most people.
But consider this, from Debian's ca-certificates package:
---
This package includes PEM files of CA certificates to allow SSL-based applications to check for the authenticity of SSL connections.
It includes, among others, certificate authorities used by the Debian infrastructure and those shipped with Mozilla's browsers.
Please note that Debian can neither confirm nor deny whether the certificate authorities whose certificates are included in this package have in any way been audited for trustworthiness or RFC 3647 compliance. Full responsibility to assess them belongs to the local system administrator.
---
So as long as Mozilla allows OSes, and presumably other browser forks, to benefit from their CA vetting and monitoring activities ...
There really is a lot that does, or should, go into browser development and maintenance.
[1] https://addons.mozilla.org/en-US/firefox/addon/classicthemer...
[2] https://addons.mozilla.org/en-US/firefox/addon/tree-style-ta...
[1] https://wiki.mozilla.org/WebExtensions/Future
[1] https://webextensions.uservoice.com/
These in particular make Firefox my own (your Firefox is not my firefox, but your firefox is OK), and I will be sad if they disappear without replacement:
Disable Style Button https://addons.mozilla.org/en-US/firefox/addon/disable-style...
Disconnect https://addons.mozilla.org/en-US/firefox/addon/disconnect/
Flash Control https://addons.mozilla.org/en-US/firefox/addon/flash-control...
Ghostery https://addons.mozilla.org/en-US/firefox/addon/ghostery/
It's All Text https://addons.mozilla.org/en-US/firefox/addon/its-all-text/
LastPass https://addons.mozilla.org/en-US/firefox/addon/lastpass-pass...
One Tab https://addons.mozilla.org/en-US/firefox/addon/onetab/
Pinboard https://addons.mozilla.org/en-US/firefox/addon/pinboard-exte...
Privacy Badger https://addons.mozilla.org/en-US/firefox/addon/privacy-badge...
Suspend Tab https://addons.mozilla.org/en-US/firefox/addon/suspend-tab/
uBlock Origin https://addons.mozilla.org/en-US/firefox/addon/ublock-origin...
Uppity https://addons.mozilla.org/en-US/firefox/addon/uppity/
VimFX https://addons.mozilla.org/en-US/firefox/addon/vimfx/
I have a few others which aren't much important to me, but these make FF my own.
This is going to kill the addon community but I guess they don't care at all.
I do know that ones like ghostery and lastpass have financial incentives, while others (like vimfx) are already planning on it. Others don't seem to rely on many (if at all) XUL/XPCOM features
Nobody needs a clone of Chrome; we already have Chrome.
https://en.wikipedia.org/wiki/Uniform_Resource_Locator
But if I copy a specific part of the url I expect nothing to be added when I paste it
Ah, but performance and stability is exactly what Mozilla is improving with WebExtensions, Electrolysis and Servo/Quantum, so you won't need to put up with the poor performance and instability any more! What's more, it will still respect your privacy more than Chrome, be open source unlike Chrome, and allow you to configure it more than you can configure Chrome.
Firefox's primary distinguishing feature is the powerful plugin system. With that gone, it's just another browser.
A change like this is how you do that. Good job, Mozilla.
Incompatible version changes drive away users and developers. The uncertainty that comes along with a planned update in which compatibility is not guaranteed makes us all shy away from developing in that application at all, not just on the old version.
Microsoft always seemed to do a good job at maintaining compatibility over its versions. Most applications that used to run on Windows still do, although there are a lot of exceptions. Python's change to the print statement broke everything, though the fix was easy; most of the time the fix is not easy - and even when it is, nothing works until someone gets around to the fix.
I appreciate the need to move forward, and sometimes you have to break stuff to make it better, but these incompatible changes just destroy momentum. It's a shame to see it happen.
Chrome....
Mozilla is turning Firefox into just another Chrome Browser...
Firefox 57 will like see me stop using FireFox it will be no different than chrome so I might as well just use chrome
>>and sometimes you have to break stuff to make it better
It is yet to be seen if it will be better, so far everything they are doing in firefox to make it more like chrome I am 1000000% opposed to.
Firefox probably won't phone google.
I use Firefox for a few Productivity Extensions that will likely stop working in 57, as such there is no reason for me to continue to use FireFox.
2. Firefox just became a little bit more like Chrome, at least as far as some of its users here are concerned
Google has been slowly consuming Firefox, so don't be surprised if any phoning home is happening.
That has not been the case for a few years(2012? or 2013 I believe), Yahoo is currently the Default Search and pays the bulk of Mozilla budget, With the Verizon Deal Mozilla is likely worried that will change when the contract is up for renewal
> Verizon must pay even if Mozilla walks from Yahoo search deal
That contract however I believe was for 5 years, I am talking about what happens after the 5 years is over, which I believe will be in 2019 or 2020. After which Verizon has no contractual obligation to continue paying Mozilla for anything and likely will cease the partnership
Who?? I know of none
MS will never support Firefox, They want people to use Edge.
DuckDuckGo do not have the cash to do it.
Google has the market share and does not need FF anymore that is why they dropped them
So Who....
>>That's not the point.
Actually it is the point, by 2020 Mozilla's IMO mozilla revenue stream will dry up, they need to find a way to survive with out the Search money. This is why they are taking these risky moves which IMO are doomed to fail.
For those who aren't familiar, pentadactyl is a Firefox extension that transforms the user interface to make it more minimal and vim-like. You get to use standard vim bindings to navigate the web and it even gives you a vim-like command line interface to the browser's functionality.
For those who don't mind, or prefer, a less comprehensive vim experience in the browser, I'm very happy with VimFx https://addons.mozilla.org/en-US/firefox/addon/vimfx/
It has a ton of vim commands, but I only use hjkl, g and gg. I never can remember whether search with / comes via VimFx or plain FF.
Firefox says it's compatible with multiprocess, but I don't know if it will survive through the switch to WebExtensions. It works fine now.
I switched from pentadactyl a while back and never looked back.
There are ways to run both versions side-by-side, but I will let you figure that out.
They are addictive. They turn Firefox into a different kind of thing. I rely so much on custom commands like:
command -nargs=1 proxy set! network.proxy.type=<args>
command -nargs=+ -complete=url site open google site:<args>
map s :site<space>
I'm really afraid of Firefox breaking Vimperator long term. Alternatives are not that nice, and niche browsers tend to be brittle and insecure.
The other is not having a single process per tab. I typically have 40-50 tabs open at any one time. On chrome that chews through memory. On Firefox it chews through significantly less.
Jetpack was introduced in 2013, and by 2014, it was actually usable. It's not clear when and whether Jetpack add-ons will stop working. Mozilla's developer documentation is so confused that you can't tell. WebExtensions and Jetpack are essentially the same functionality with different names.[1]
The trouble with Firefox chasing Google Chrome compatibility is that, once they get there, Google will probably change something so some Google service is essential to an add-on. Like Google did with Android. Then compatible add-ons stop working on Firefox. Nobody runs add-ons on Chrome much, anyway; I have the same add-on on both platforms and its 50:1 Firefox over Chrome. But usage is declining, along with Firefox.
[1] https://developer.mozilla.org/en-US/Add-ons/WebExtensions/Co...
I'd be more concerned that, once they get to Chrome compatibility, they've simply sacrificed one of the key things that differentiates them from Chrome. Firefox's die-hard adherents use extensions heavily, and many of the most popular, particularly those that are on Firefox and not elsewhere, depend on the old XUL layer or other APIs that have no replacement.
I don't think users are going to see it as a great improvement that they have access to Chrome extensions they didn't want in the first place. Instead they're going to focus on the fact that they're losing the functionality that's been keeping them using Firefox.
I don't think you'll need Google to act maliciously or really do anything at all to have this end poorly for Mozilla.
i really believe they are signing their own death certificate. neither google nor yahoo will pay much for search traffic of a browser no one uses. i say this as a long-time fan, nightly firefox user and bug submitter (since v1.5). everything since Australis has been an unmitigated trainwreck :/. When Classic Theme Restorer [1] stops working, i'm sorry but i'm out....probably to https://github.com/Eloston/ungoogled-chromium for the time being.
my hope is that someone builds out a nice cross-platform browser around servo, or at least i can hack on https://github.com/browserhtml/browserhtml
http://electron.atom.io/ is an option, too
[1] https://addons.mozilla.org/en-US/firefox/addon/classicthemer...
Well. The number of installs of Firefox for Android is a nine-digit number.
Depending on your perspective that can be thought of as a big or small number.
But it certainly is not "nobody".
Per https://www.netmarketshare.com/browser-market-share.aspx?qpr... it doesn't look like the usage numbers are 9 digits.
(that said after reading tfa, if there's really no tab groups after next year i'll be really upset, especially after the whole "it's ok that we're removing it because now it's just an add-on" thing)
We [web developers] owe a lot to FF, Firebug, etc, but the writings on the wall for mobile and desktop. Chrome is going to have to screw up really bad, and likewise for the base installs of IE/Edge and Safari. These days, I'm more likely to suggest Brave over FF, if someone really requires a Chrome alternative.
Brave is just a reskinned Chrome.
> We [web developers] owe a lot to FF, Firebug, etc, but the writings on the wall for mobile and desktop.
Where I work developers have mostly switched to Firefox over the last few years. Firefox is just a better browser (faster, less bloated) under the hood. Yes, Firefox will have a difficult time sine they don't have their own proprietary walled-garden ecosystem as a distribution channel, but the technical product is solid.
And it's not the experience for many to have a faster browsing experience, but I'm not going to argue how anyone quantities that -- really don't care what others use.
On desktop, I find the tooling much worse and sluggish. But the thread was about mobile, and there are a vanishing number of users of mobile FF.
No, see https://news.ycombinator.com/item?id=13534564.
I understand business realities, and why Firefox mobile would rate a lower priority than many browsers. But I think competition in this space is incredibly important and I urge you not to ignore us [Firefox mobile users] completely.
FWIW, I love mobile Firefox and I really don't understand why folks use Chrome instead.
If there's one thing that sucks it's typing secure passwords into a phone.
e10s might be the thing that could save FF, but I fear that it's way too late.
So I switched to Firefox on Android, it's not 'quite' as smooth but Ublock Origin works great so on balance I like it better.
FF is slow, pages rarely load the first time and I constantly have to manually reload, the interface for making config changes are requires sooooo many clicks. I have no real experience if chrome is better because I never use it, but FF mobile has some real warts
the only solution is installing noScript. the only version available is in beta. but it works fine.
you will be suprised how few sites I have to whitelist to see the content I want. and usually besides faster, they get much more usable without Javascript!
I'm sure there are more such examples. Even if a default was changeable the initial setting (as we saw in the Microsoft vs. Europe over IE) is very powerful.
How is that a good idea?
and most extensions are making do with xul. I use noscript and stylish and httpsEverywhere. life is good.
I'm here :(
I use it to play YouTube in the background.
Firefox wants to add its own APIs to webextensions. Chrome will too. They may not always match with each other with the newest APIs. It cuts both ways here, it's wrong to characterize it as something where only Chrome will be adding features and Firefox will play catch-up. Firefox is "copying Chrome" only as far as copying the base API.
See https://wiki.mozilla.org/Add-ons/Contribute/Triage#Bug_Queri... for the current list of APIs that are planned or being decided on.
Other than MathML support, I'm having a hard time thinking of anything truly unique about Firefox, as compared to Chrome these days.
I'm not sure I agree there. I feel like they've had at least a few pretty strong points of vision. The failure has been more on the execution of producing and maintaining useful software in support of that vision.
It'd be nice if a mod could change the title to "I cannot continue working on my Firefox add-ons"
Should I just not update firefox and wait until devs realize tabs were meant to be on the side? Honnestly as a power user I can't browse the web without tree style tab anymore.
Between auto-update, security patches and new web standards support, how long will this approach be viable though?
It is better than Sidewise on chrome though, however TabCenter is XUL as well I believe, there are no APIS currently for it to be moved to Extensions,
Furthermore, "showing tabs on the side" is not "similar to Tree Style Tabs"
But yes, the "tree" part of Tree Style Tabs is essential, too.
We definitely need more browser competition but growing complexity rules out one person or small team efforts like before all over the open source ecosystem.
Untill open source users can get their act together and find a way to fund development they will find projects increasingly less accountable to them than their corporate sponsors.
Why use Chrome when you can use a Chrome clone that better respects your privacy, is open source, and gives you more freedom to configure it? Because it's more secure or faster? It won't be for long: that's exactly what Mozilla is working on. At the expense, yes, of some configurability.
Also for a open source browser. Are you aware the is chromium? Or hell any number of electron based web browsers (which is also chromium).
Edit and loss of configurability is not the issue. Loss of Functionality is.
If every browser looked exactly like Chrome, I'd still choose the one that contained the least amount of Google.
Mouse gestures are in my muscle memory. I can't live without them, and every single Webextensions Chrome mouse gestures addon either:
a) Sucks. Doesn't work at all on internal pages including new pages (which is not a blocker, but sucks), and more importantly doesn't work _well_ ANYWHERE. Inconsistent behavior, poor performance. Just unacceptable.
b) Is actually literally malware. Once Chrome extensions get popular, a shady Chinese or Russian consortium buys them up, adds user tracking, and silently updates the extension so your every move is followed.
Chrome has been around for almost 10 years now, so I'm absolutely convinced that Webextensions are unable to support _GOOD_ mouse gestures.
Now that Firefox will shortly become hot stinking Chinatown garbage, my only hope will be Vivaldi, a browser using Blink (Chrome's backend). Its mouse gestures are really close to being usable-- it only lacks configurable mouse button chording.
https://vivaldi.com/
Now I'm using Firefox plus a dozen extensions to poorly replicate a base Opera 12 installation. The e10s update already broke some FireGesture in some minor ways (e.g. wheel gestures scroll the page content now). Other extensions stopped working completely.
Time to give Vivaldi another shot...
Unmodified Firefox is just an inferior clone of Chrome. Firefox is pointless without its UI-altering extensions.
Quite amazing.
I've been worried since web extension only announcement we were heading for something like this.
I know in my own case, and it seems from comments here and other anecdotal experience, that a significant chunk of remaining firefox users depend on at least one extension that has no migration path, as there simply aren't equivalent APIs implemented or even planned. I really don't understand how they think they are going to maintain a userbase when they remove the last remaining feature that differentiates firefox from the other common browsers. Mozilla starts out in a tough competitive position, as they don't have an OS or platform to bundle a browser with. Alienating the users with the most reason to seek out an alternative seems to just be digging themselves deeper into the hole.
If you build a browser extension today, you build it for Chrome; if you have a Mac, you also get a Safari version almost for free; and then eventually, if you have time, if you can be bothered, you port it to FF. This could not go on, something had to be done. And they did it: JetPack/Webextensions, despite the uber-confused documentation, is an almost-complete compatibility layer that makes porting extensions from Chrome much easier. If they'd left it at that, nobody would have complained.
Unfortunately, the next step is to gut the old infrastructure. They might have their technical reasons for that, but it seems a suicidal move from a commercial perspective. They are throwing overboard almost 20 years of competitive differentiation and community engagement. It's like Microsoft told people to rewrite their .Net apps in ObjC.