Firefox UI/UX History
github.com
github.com
Thankfully however I don't care too much because I can still completely hide the tab-bar with my userChrome.css, and use "Tab Center Reborn" to have my tabs on the side instead.
Thanks for the extension suggestion I may give it a try.
I think I read somewhere that the reason was that user studies showed that people did not understand that the tabs could be dragged around, moved to other windows, and generally were "detachable". Floating buttons should make that more obvious.
I'd be interested in research on whether they achieved that goal.
Firefox’s Redesigned Preferences Feel More like the Web
That expresses exactly what's wrong --- the native UI controls are predictable and accessible and styled with the rest of the OS, and as the cascade of dialogs above it (I'm not sure how that's even possible --- getting to "Offline Data" From "Exceptions - Saved Passwords"?) shows, you can actually see that you're controlling the browser and not merely a page inside it (related article: https://news.ycombinator.com/item?id=30697329).
Another important change is the introduction of a skeleton screen to make the start feel fast
I've never heard that term before, but whenever I see things like that, I'm reminded of fancy progress bars and such --- the real problem is that you need one in the first place, and instead of fixing the underlying problem of why it's taking so long, you try to hide it...
This is such a weird thing to me. Did anyone ever want this? Were users asking for this? Did user studies show that this was somehow desirable or preferable? Did they do a study of Chrome users and find that they were more likely to stick with Firefox if they made this change? Does it significantly reduce code maintenance burden somehow?
It seems like the kind of thing where users only "want" it because it's familiar, and it's only familiar because it's already been forced on them everywhere. Basically it's circular reasoning. It's like the meme of hiding a button and then claiming that users never use that feature, so you can justify removing it.
In this day and age, noone cares what actual users want. Instead, corporations tell users what they "want".
How do you design and implement an interface that feels native across platforms?
First there is the easy approach: The common denominator elements. Essentially what you get when creating a Java Application. The UI will probably look kind of native, but also outdated. Think Windows 89-Style Settings. You're also limited in how you can present options because you have so few form elements to play with.
The other approach to native is to use modern UI elements. For one thing these are not easy to get. While on macOS Cocoa bindings are at least available, on Windows it is pretty much impossible to get anything but the Windows-89 style form elements. Even if all the bindings were available, which also has to take various Linux Desktop Environments into account, then you had to design for and maintain multiple versions of the same settings menu.
So I hope this explains why non-native controls are a sensible approach. You create your own design system that is (mostly) consistent across platforms. Designers only have to design one surface, they can create new patterns if necessary and engineers only have to maintain one implementation. Of course they have to be accessible and they are. I say "mostly consistent" by the way, because Firefox still respects some UI conventions from operating systems. For example it uses system fonts, system font-scales, and changes naming conventions and button orders appropriately.
This reduces work but of course it still is work to update and maintain that surface. about:preferences in Firefox has quite become quite messy over the years and is neither loved by designers nor engineers, but it's time and resource intensive to update it.
Is it really more efficient to maintain a design framework chasing latest web trends instead of having separate native interfaces?
Developing a cross-platform app that uses native interface elements is extremely time intensive and sometimes laughably complicated[1].
With design system I meant that designers create a spec how the UI should look like and behave (eg. the primary button in Firefox) and then engineers implement that in a reusable way. Designers can then use their own specs to create new surfaces and engineers can use their reusable components to make the specs real.
This is a lot more efficient than either maintaining three code paths for the whole UI or building and maintaining a complex abstraction framework for native surfaces. That's what XUL was.
That's the reason why there are barely any UI-heavy cross-platform apps and a lot of web-apps.
This is the core of misunderstanding. Once you define your goal as "the app should look the same across platforms", you inevitably set yourself up for these shenanigans. In order to keep consistent look and feel across platforms you either implement (and maintain!) a complex abstraction layer over platforms or reduce the design to "common denominator elements". Web (both remote and electron) seems to be an attractive solution, because it offers the more pleasing abstraction option and a lot of heavy lifting is handled by browser and framework developers.
You base your argument on consistent look being the ultimate goal and desktops being somewhat moving targets. Well, the web is a target moving even faster and I see no reason to regard consistent look as the ultimate goal. It could be easily argued that platform native look is an even better option.
Essentially, it boils down to two fuzzy metrics: fraction of users regularly moving between platforms and preferring consistent look (a windows-y app on mac will look off, webby app will look off on both platforms) and long-term cost of maintaining the consistent look versus maintaining separate fully native interfaces. Somehow I tend to believe that in the long run native interfaces are the better option.
Web only makes it easier to build cross platform MVP and forces one to release features in tandem across platforms.
"What about users that want to customize things?"
Give them an alternate path. The browser loads slightly slower (the same as now), and they still get their ability to customize. The rest of us can use the default, faster path.
That is a bold statement and largely not true.
Most DAWs are only developed and maintained for Windows and Mac. Almost all of them are using their own set of widgets/gfx because they want control on their style and they want it to be the very same on both Windows and Mac. Their software is very specific and want their tool to look and feel like hardware, not a regular software.
It is mostly about control. Linux doesn't have anything to do with that and any software maker can decide to only support GTK or QT, libwine or whatever toolkit it wants to use and stick with it. Linux users are used to run QT apps on Gnome or GTK apps on KDE, they are not the one that would complain and given the market share, the software companies are happy to give the middle fingers if they do so anyway.
Bias towards fashion trends is one of the biggest problems in the software industry.
> Another important change is the introduction of a skeleton screen to make the start feel fast
Yeah. Sometimes you do have to make users wait— e.g. parsing a huge cache to put in a dialog or making shaky network calls— and most designers mental model for addressing that frustration is the slow elevator problem†. Skeletons seemed like a good way to address it. We don't parse screens instantly: we first interpret structure/visual hierarchy so we know where to scan for titles, ordinals, controls, or whatever else we need. Skeletons theoretically let us do that while content loads. More recent research doesn't corroborate claims of silver-bullet efficacy:
https://www.researchgate.net/publication/326858669_The_effec...
† The related anecdote goes something like this: Tenants of an NYC office tower complained of slow elevator service during peak use. The property manager consulted with experts to evaluate the algorithmic efficiency and mechanical components, but it resulted in little improvement. The property manager then turned the exterior elevator doors int mirrors. Most riders were sufficiently emotionally engaged by a life-sized version of themselves long enough to make the wait seem much shorter.
I thereby will claim that cross-platform UI toolkits (which includes "the web") actually affords me--as a user--tremendous value, and I am always pretty happy when I see a product start to figure out their house brand. I do appreciate the other side of this, and I even have sometimes argued it myself to engineering teams I work with as--if nothing else--using the local style is certainly better than being bad at style, and I frankly would rather avoid having to work with a designer. But is it really so shocking to appreciate mine? I always feel like people jus let casually dismiss the idea that users might actually prefer software to look the same everywhere, and yet I feel like it is only particularly snobby tech people (which does include me: I hate poorly done UI) that seem to want the fully unified toolkit, and I think the success of bespoke UI in games really demonstrates this.
Most users don’t use the same app on many devices and operating systems. That you consider the OS a constantly changing fungible component in your daily tool chain is incredibly unusual. Outside of some dev and IT tools, optimizing for that use case just doesn’t make sense and even most IDEs don’t try. Most device UI kits are more carefully designed to suit their users interaction needs than any app could hope to achieve, especially considering accessibility.
1) Create Window 2) Create UI (based on native controls for OS) 3) Load drop downs, top level down. 4) If you design your application this way, you won't have to build a skeleton. Mozilla is not using native UI/UX components to build their UI, so they have to go through several extra steps. I'm sure they do this in the name of customizability, but that, IMO, should be optional.
Wow, I think you've identified the probable cause of something I've felt was "wrong" about Firefox for a long time but couldn't pin down. Firefox just "feels" bad on a Mac even in comparison to Chrome, not to mention Safari.
The new web-based pop-ups can take a long while to render if a heavy page is being loaded. So if I try to quit Firefox, the "are you sure you want to quit" popup renders empty until the current page is done.
Sure, this happens only on high resource usage, but the previous implementation di not suffer from this.
What an absolute abomination that was. An insane waste of space and a violently assaulting visual distraction. Especially when you're trying to open a document that the current one is referring to, hiding what you're trying to see with useless space.
"Windows is called Windows, not Window."
I’m sometimes tempted to try to start a Firefox fork that returns its UI to a “smaller part of a larger desktop” look and feel, but then I remember how impossible it would be to maintain that over time. Wish Gecko was still embeddable so one could just write a new UI and not have to keep patches maintained.
The one thing I will say in its favor is that Firefox still maintains a decent amount of customizability, at least for the seriously dedicated. I still have a separate search bar, a menu bar, and have removed the "hamburger" menu button entirely. For me, the UI has changed very little over the years, but the pile of hacks in my usercss file has grown over time...
It's still "consistent", just on a different dimension (app-wise instead of OS-wise).
I know HN favors the former, but I don't mind the latter too much. Especially when products are able to strike a nice balance like Outlook on Windows & Mac.
Do they actually? I'm guessing most people in the world have one smartphone, followed by those who have a laptop to match it.
I would be careful to say that most people have more than one device though, which means a vast number see no consistency at all.
As for the 'curvy' tabs, I can't say I was a fan of it, even in Chrome, it feels like a waste of space and draws attention to it. The straight tabs, even the 'button' tab is a lot better in being out of the way.
Getting rid of the icons in this case means you don't get a hint that this menu entry is an alias for that button (or vice versa).
Could you give an example of it being a struggle?
Without the icon, you have to consciously read the lines in the menu to make sure you are clicking on the right one.
You still read the text with the printer icon, but it's a faster index-match when the icon is present.
Its UX "history" made it plummet from the top browser at one point to a forgotten has-been. These tweaks were not successes or celebrations, they were the death by a thousand cuts.
My guess is the rise and marketing of Chrome and its offspring had more to do with Firefox's decline than anything Mozilla has done.
I still think there is plenty of room for a privacy respecting browser in the market, but Mozilla hasn't even been trying to fill that niche in years and still claiming to do so just makes them look like untrustworthy liar.
It's kinda like saying "I'm moving to North Korea because freedom of speech in the US sucks!"
Every thread about Firefox is filled with complains, be it removal of essential feature, shifting around the UI for no reason or addition of stuff nobody asks for. Not every user might care about every of those issues, but do you think that amount of negative feedback is good for attracting new users or keeping existing ones? I don't.
Also the level Firefox has sunken too is mislabeling the Screenshot cloud upload button "Save". That's plain old malware dark pattern strategy to steal your data. They took almost a year to fix that and it's downright puzzling how that ever made it anywhere near a release in the first place.
Chrome might be crap, but I am not going anywhere near Firefox anytime soon either. They have shown time and time again that they really aren't on my side, yet love to claim so. The sooner we get rid of Firefox, the sooner there is a chance a real alternative might arise.
IMHO, the _moment_ we get rid of Firefox, the _moment_ we lost the free web. Building a web browser is just too damn hard, and never again would free software stand a chance. All will be Chrome-based, and Google will decide what "standards" to adhere to while it sits in a committee with itself.
That's what people said when I started to write an open-source DAW 22 years ago. "DAWs are just too damn hard, there'll never be an open source DAW".
Except that now there are several open source DAWs, and several proprietary ones, all created since that time.
I am very, very skeptical of claims like this. There is a reason why creating a new browser that lots of people will use is a challenge, but it's not because it's "too hard".
Browsers need to follow ever changing standards, do all that in a super performant way (remember the days people said they're leaving Firefox because Chrome "feels snappier"? Good luck beating that), keep it secure even though it's running remote code, and until they get it ALL 100% working, no one is really going to make it their daily driver. I already hear people saying that they're not using Firefox because some websites don't render well.
If it isn't "too hard", why do you think that over the last decade essentially no one managed to do it, while we do have several open source DAWs?
You don't have to beat Chrome at it's own game. I think the best course of action would be a drastic course change and building a browser that focuses heavily on the creation and publish of content, not just on the consumption. Focus on the Web as document storage instead of as App runtime. That's an area that is still in serious need of work and isn't really covered with Chrome. Also somebody really needs to reinvent bookmarks, they haven't fundamentally changed in 25 years and are in dire need of an upgrade.
Brave (IPFS and Crypto integration) and Project Gemini (focus on text content) are going a little into that direction, but there is still a lot more that could be done.
> Firefox because Chrome "feels snappier"?
It was less because "feels snappier" and more because "complete browser freezes when using multiple tabs". It has gotten better since then, but when Chrome started Firefox was in dire need of some rework.
Don't get me wrong, I appreciate the hacker mentality (go Gemini!), but I thought we were discussing the need for an open web. Sure we can even add NNTP support to Firefox, but that ain't gonna get my grandma switching over.
What exactly is "open" about just being a Chrome-clone? If you want to keep the Web open, you really have to work on the "how-to-get-content-on-the-web" side of things, just another browser for pure consumption really doesn't add all that much. Stuff like containerizing Facebook isn't really doing anything worthwhile, giving people something to get away from Facebook would. And forcing HTTPS on everybody really didn't help either, that just killed what remained of the old Web.
It's not like Firefox didn't try, at one point they added Firefox Hello and that looked promising and than it got removed again. And at one point they had RSS support, but instead of improving and building up on that, that got removed again. And instead of improving the Firefox bookmarks, Tab Groups or the Save button they gave us Pocket, which is no different than any other cloud service.
> but that ain't gonna get my grandma switching over.
You don't improve the Web by giving grandma a Chrome-clone, she'll just be Facebook'ing around all the same as she did in Chrome. And when grandma is paying your bills by visiting Google Search, why even bother, cut out the middleman and just use Chrome. It makes no difference.
Now I don't necessarily want NNTP support in Firefox, as at that point we'll just be full circle and back at where Mozilla started 20 years ago, but I want something that allows communication and publishing of content without having to rely on Facebook and friends.
Most people are consumers of data. That's why browsers are called Browsers and not Authors. Grandma isn't building a website for herself not because it's hard, but because she doesn't care. The risk with with Google being the only browser is that they define how we browse. They can decide that next year HTML is gone and Flutter is in. They can decide FLOC is mandatory to view a website. They can decide to only show AMP content. Then, my friend, then it's gets harder to put your own content on the internet. And this is why Firefox is important.
NAT is still a major issue when you try to self host anything. And most of the stuff you can hobble together in a shell one liner just ends up being broken or limited in one way or another, e.g. that `python3 -m http.server` fails with seeking in video files, just gives "Broke pipe" error. Also provides no way to encrypt or authenticate. And without any way to easily mirror the content it will be unreliable and slow anyway.
Trying to do almost any common task is a nightmare when you want to do it cloud-free. An open web run by users themselves is still an unsolved problem. There some projects working on it (libp2p, IPFS, etc.), but none of that is to the point where it works properly and often missing important features.
> They can decide that next year HTML is gone and Flutter is out.
That already happened years ago. Tons of popular Internet apps only exist as mobile apps with little or no Web interface. Worse yet, most of those apps are driven by user created content. Which is exactly why making publishing a first class citizen on the Web is so important, without that people are just leaving the Web and going to places that allow them to publish and those places will be controlled by your favorite mega-corp.
In 1999, and maybe even 2011, a DAW that couldn't do elastic time was still considered to be a major player. In 2011, a DAW that couldn't do clip launching was still OK. In 2011, the idea of cross-routable modulation (like Bitwig now does) was a wild and crazy idea (mostly). In 2011, the idea of a modular environment within the DAW was floating around (Logic had done it for a while) but was hardly mainstream. These things are now either already or becoming more or less mandatory to be considered a major DAW.
Then there is the matter of control surfaces, which keep evolving. In 2011, the idea of a controller that was essentially a programmable grid of illuminated pads was mostly the domain of experimental performance (Monome). And every month or two, a new controller appears that claims to support Mackie Control Protocol, but has in fact bastardized it in some small but significant way. There's also the touch-based controllers, frequently using OSC, and their constantly changing suggestion of what should be possible.
And then there's plugin APIs, which also keep evolving, and we are under enormous expectation to ensure that every single one of many thousands of little code blobs written using one of a half-dozen plugin APIs and by developers with a range of experience that goes from nothing to world expert should all just work.
Over the years, we've also seen constant changes on Windows in terms of the OS-level audio APIs, each change bringing with it new possibilities and new problems. There's no POSIX for this. And, of course, a slow but steady evolution in audio interface hardware, which doesn't often impinge on the DAW itself, but sometimes does.
As for performant, browsers don't even have RT-style constraints, and there's plenty of observations about "snappiness" in this domain too. It's just a much smaller niche, and developer culture (as testified to by posts here on HN) is now very dominated by web-centric thinking and experience. As a result those sorts of things aren't really part of the culture in the way that "wow, Chrome is so fast" etc. has become. Look at the praise Reaper receives because of its (apparent) ability to handle more plugins with less CPU cycles.
As to why new browsers are rare ... I think it's because (1) browsers are not really fun at most levels, and (2) adoption is hard when there's a default on just about every platform. The fun part is important though. Many developers enjoy "messing around with audio" and it provides a kind of gratification that is rare (video stuff would be similar). It's not that cooking up a standards compliant and crazy fast and beautiful CSS implementation isn't without its joys, but there's so little point doing that as a standalone project. Contrast that with the steady drip, drip, drip of developers who want to try their hand at a synth, or an audio file editor, or FX processor and eventually think "oooh, how about a DAW". There's not really any evolutionary pathway for browser development: you're either writing a whole browser, or you're not. Audio gives you a way in, and then the sky's the limit.
It's almost the same as old school Linux, where commercial software said "Hey I added this one click button for the main use case" and FOSS said "You don't need that, just have a bash one liner instead".
Chromium is openish. Other developers could fork it if they wanted to.
Firefox isn't doing anything about the fact that self hosting is way too much hassle for anyone but a few hobbyists. They're not really doing much in the IoT space.
They don't seem to be addressing the fact that the web is basically just Facebook and Youtube and Amazon in any way, except by adding ever-more tracking prevention tech that's not really relevant when all data goes through the same 5 sites anyway.
They actively get in the way of P2P tech by locking everything down so much it's impossible to implement a lot of things.
The lack of filesystem APIs just promotes even more vendor lockinful web services.
Mozzila does a lot of good things, but I'd rather they just switch to the Chromium engine, restore the removed features, and go back to what they were doing 5-10 years ago in the FlyWeb and WebThings era.
Every Firefox update on the other hand will always find one way or another to interrupt your flow at the worst possible option, whether it's with useless UI updates, post-update notifications about bullshit "features" such as Colorways or a VPN, etc. In contrast, Chrome's minimalistic UI has barely changed in its entire lifetime.
The question for Firefox users is whether Firefox is on their side. If neither vendor is on your side, why not use Chrome? Has the UI on Chrome ever changed significantly?
> IMHO, the _moment_ we get rid of Firefox, the _moment_ we lost the free web.
I'd date this to the moment that the company that owns Chrome became the source of the vast majority of Firefox revenue, and probably all of its profits.
Mozilla are doing stupid things sometimes (and I've argued for that many times, in this thread, and in the Mozilla community), but they're just not even remotely as bad as Google. Firefox has containers, uBlock, tracking protection. It's literally the source for Tor Browser.
The question isn't a dichotomy of "who is on my side? none? so it doesn't matter". Firefox is still miles ahead than Chrome in privacy AND in keeping the web open. True, this doesn't make it perfect. But it sure does make it better, for all of us.
The thing about Firefox is that it's an UNKNOWN evil. Mozilla always feel like it's on the cusp of bankruptcy and constantly searching for new dark patterns to sneak in. When Wikipedia needs money they beg for it, but don't purposely sabotage the user experience to get funding.
Mozilla does that with every new release. I always feel like they've added some shady new malware/adware with every new patch, and then use some stupid UI tweak to try to hide it. It's only a matter of time before they sneak Norton in there. I trust the Firefox team even less than Facebook at this point. Firefox just isn't trustworthy, whereas Chrome is a known compromise.
You trust the goons working for Facebook less than the goons working for Facebook? /s
It hit me really hard when during the whole FLoC controversy Mozilla published its own collaboration with Facebook on the future of browser based user tracking. No amount of technological hand waving could have fixed that first gigantic WTF and a description filled with privacy budgets, trusted third party servers, etc. didn't even have a snowballs chance in hell.
You provide valuable behavioural data to Google, which uses it to create very targeted demographics which are used for targeted advertising and analytics that are sold to advertisers.
Seen in this way, it’s quite darker than that, in my opinion.
I mean, our own government harvests even more data and does jack shit with it. At least Google provides a world-class office suite.
Was forced to used the Microsoft stack at a new job and it made me miss the Google ecosystem so much.
At the end of the day privacy isn't that valuable to people. Nobody cared about it in the 90s when the internet was developing, it was barely a blip in the 2000s, and somehow it exploded in the 2010s but plenty of people still use Facebook and TikTok and such. So?
Usability > privacy for most people, a lesson Firefox refuses to acknowledge, I guess.
It's outdated and insecure, and I don't care. There simply isn't any other browser worth using for me.
(Though that may change now that there's a way to sacrifice all the new security features on current Firefox to run TMP again.)
Firefox Sync eventually arrived, but it was a PITA to set up because I didn't need a Firefox account for anything else. And then they added a bunch of third party services that I still don't know what they do (Pocket), ads on the main screen, ads on startup, full-page release nags for pointless features, constantly changing UI for no good reason at all...
It was like Chrome took Phoenix's philosophy and Firefox tried to become Netscape Communicator again. Zero of the Firefox features added in the last decade helped me as a user, but instead constantly bugged me. I can't remember a single annoying thing that Chrome added in the same timeframe. It still feels leaner, quicker, and less intrusive.
Funny how the sides have switched...
It made the browser compete with itself, and pushed people into alternatives. After all, if you're going to learn a new UI, why not try another browser altogether?
Google added things like WebUSB, Bluetooth, all kinds of web app APIs Mozilla rejected because of tracking risk, etc.
Mozzila killed their coolest features like FlyWeb.
They just haven't kept up with Chrome, and their vision is way too "privacy at all costs" rather than allowing users to decide. They don't seem to share in the modern idea of web apps having full native parity.
What an awful idea that is.
Those are literally the last things keeping me with Firefox.
They don't seem to have fully made up their mind on who the target audience is, it seems like their goal is to convince the mainstream to prioritize privacy over convenience.
A better strategy would ve to either become just another Chrome-alike, or focus on an uncompromising privacy focused experience for the people who really want that. But that might require they not do some of the slightly shady stuff they always seem to want to do, and properly support uBO long term.
https://www.masswerk.at/nowgobang/images/netscape-navigator-...
Edit: The "Classic" themes of early Firefox and SeaMonkey were still very much reminiscent of Netscape 4, but the toolbar and the location bar merged. (In the early days Firefox and Netscape 6/7 still coexisted in parallel, packaging the same engine, and the "Classic" theme represented the Netscape legacy UI.)
There’s another state that may be related to uBlock Origin (or other extensions) immediately afterwards, too: You can click bookmarks etc. but nothing loads (only after a significant delay). Again, awesome UX.
I guess I’m slowly becoming conservative because I’ve come to hate change. Most of the time, it doesn’t improve anything. And then, with increasing odds, changes actually make things worse. Not just limited to Firefox, of course. Also Windows, macOS, iOS, Android, whatever. Hell, even cars are getting worse every year.
Some people in the browser space haven't lost their minds and track things like how often the layout of a page changes during loading, how often it wiggles around, etc - which penalizes skeletons and other bad tricks, as it should. But this particular sensibility hasn't caught on elsewhere.
Not everyone runs good hardware, in fact the majority doesn't. When you click to open something and it doesn't load immediately many people will click again and again which leads to opening 2+ windows and also adding to the loading time of the first one.
When 70%+ of the users are not skilled in tech using laptops on battery saving mode this happens a lot.
Nowadays, the innovation has plateaued and computing became an ad delivery mechanism and a way for employees of bloated corporations to justify their salaries "fixing" things that didn't need to be fixed.
Precisely. Much better. I know something is happening. And if done well, I can interact with the shell, like use the search box, while the content is still loading.
I also already know something is happening. That’s what the busy cursor is for.
It seems as if only Firefox on Windows has these skeleton screens and so couldn't try for myself. I assumed that much of the shell like search bar, close button, and the menu would be ready.
I get the feeling that many people have, that FF is following suit with chrome in terms of UX (and other things), rather than innovating[1]. There is some truth to that, but I do feel like FF tries out some nice things in their UI.
I'm currently running on FF on MacOS, and I find it incredibly hard to come up with any legitimate criticism of the UI, but it does help many apps that MacOS forces a top menu.
Yes, there are little things, like tabs are now buttons, and container tabs should be easier to use without first reading a guide.
[1] And maybe many vocal HN'ers objection is directed at the mere fact that innovation is attempted.
No wonder they peaked user share early on and have rapidly hemorrhaged since then.
Ummm yes? Funny, I just wrote another comment complaining how hitting refresh on a major airline's flight status page is liable to give one of those terrible "something went wrong" messages (apparently the search results are liable to expire, but until they do, refreshing will actually update the info, so it is useful to periodically press to see if the flight got delayed).
Combining stop and refresh is also dangerous since that's a race condition. Suppose the page is taking forever, I decide I've had enough of that and want to press stop. But in the second it takes me to actually click on it, the page decided to finally load! Now my click, instead of stopping the annoying spinner animation, clears out the information that FINALLY arrived and brought the annoying spinner back.
(I actually think there is a pretty good case to just remove the stop button though, it isn't vital and just hitting back or close does the same thing anyway. But combining it with the refresh button is about the worst possible thing you could do to either button. It takes one useful button and one meh button and combines them into one awful button.)
Speaking of clicks, that's actually the last straw that made me abandon firefox entirely: in version 75, they broke clicking on the url bar. Instead of doing the sane thing - putting the cursor where I clicked - it instead selects all. Before, double click would select all, meaning if you wanted that, it was still easily available. Now double click instead selects... a word. So they completely removed the very useful single click functionality and turned double click into a completely useless behavior. Absolutely unbearable.
I could tolerate or undo most the other UI changes they did, but this click thing couldn't be user style'd away, couldn't be configured away. They just shoved this massive breakage down. So I uninstalled it .
> takes space even when you're not loading
I'd rather have a predictable UI of marginal utility that doesn't change out from under me than a UI that randomly changes from bad to worse in its own time. At least you can get used to consistency, even if it is consistently mediocre.
> What makes v1 "the best ever"?
Don't put things in quotes that weren't actually said. I actually said "v1 was not great but ok, and it got overall worse".
There's actually specific changes I think were good, coupled with other changes that were bad.
What browser did you move to after you uninstalled Firefox because of the click in the URL thing? Seems like Chrome does the same. To me it actually makes sense - I think most of the times when a user reaches the URL they want to put a new one, not to edit the existing one. This behavior makes that easier.
"the best ever" implies it is the best.... ever. Inclusive of all competition. "the best firefox has ever had" limits the comparison to just firefox.
> What browser did you move to after you uninstalled Firefox because of the click in the URL thing?
I made my own ui skin for chromium. I'm not terribly happy about using chromium but I couldn't figure out how to use the firefox engine for it - another strategic mistake by the incompetent buffoons at Mozilla not making this easy, and of course doing the whole browser myself is a big job (heck, even the UI skin is a big job, there's a LOT of things still on my todo list, but it minimally works without driving me absolutely nuts which is more than I can say about Firefox's recent versions).
> To me it actually makes sense - I think most of the times when a user reaches the URL they want to put a new one, not to edit the existing one. This behavior makes that easier.
Most the time I want a new one, I'll open a new tab and thus have a blank url. If I'm clicking on an existing url, it is because I want to edit it.
But if I do want to replace it, the old double click behavior to select all made this utterly trivial too: single click to edit, double click to replace. Hell, they even used to have an about:config thing to make single click select all, but in their infinite wisdom, they removed even that hidden config option to turn it back off.
With the new behavior, editing a url becomes quite frustrating. The work around is to click, wait until the double click time period elapses, then click again. Don't do it too soon though, or it will do the completely useless "select word" thing instead. If you click a third time, it will now select all again! Fourth click? Back to select word. You must wait for a while before you can just click to edit.
Oh, and do you use middle click on X? That single click select all... does NOT assert the PRIMARY selection. But double and triple click do! So if you single click it, it looks selected... but it isn't. Double click out of old habit? Now it is wiped out. Ridiculous.
So this change marginally benefits some users (who probably don't really care) while being devastatingly crippling to the people who do. Maybe some utilitarian function can say +1 usability point times 10 million users outweighs -1000 usability points to 1000 users.
But I'm gonna go ahead and disagree. Mozilla is a niche product at this point, maybe they should start caring about the 1000 users they have left.
That was easier than this patch? https://bugzilla.mozilla.org/show_bug.cgi?id=1621570#c92
Many of these changes were done to unify the UI across platforms. Yet using Firefox on iPadOS is an even worse experience than on desktop. Tapping on the URL bar has the same select-all behavior, and it always takes me several taps in order to place the cursor where I want it, in order to change part of the URL. It's absolutely infuriating.
Not to mention that it ignores arrow keys on the Smart Keyboard Folio to clear the selection, so it forces you to use awkward and imprecise taps. /facepalm
And I constantly run into major UI bugs, like tabs getting confused and showing blank or wrong titles; sometimes tapping marks the tab as selected instead of switching to it; sometimes tapping on links inexplicably opens them in a new tab, etc.
How these issues continue to plague a browser as popular as FF is beyond me. It almost feels like it's never tested by developers and designers and updates are just pushed indiscriminately.
I'm considering just giving up and using Safari, since that's what I'm using anyway behind the awful FF UI.
As someone who's been using it since the phoenix/firebird days, I disagree strongly. There have been hiccups along the way (particularly the tabs appearance in proton), but in general the UI has gotten significantly better with each redesign. Looking back at the v1-3 UI all I can see is how much vertical screen real estate is wasted on things I would very rarely click.
v1 was a perfect baseline. Then you could reconfigure to your hearts content, or reskin with a few clicks.
That being said, I never did that. I like v1, and the only thing I would change about it is to take out the tabs and move them to the side.
The argument is similar to that with any major UI change - I have one coworker who still complains about the Microsoft Office Ribbon UI. For casual and newer users, the old office UI was tedious and cumbersome hunt down whatever you want from the dozens of icons and buttons. The newer layout may require several clicks for power users to do something they used to do in one. I found it mildly annoying when it first changed, but I committed to it and now I find it much more usable. Same situation with Firefox.
But then again I must have vertical tabs so ideally just kill the tab bar.
I have zero use for an actual title bar - it's a waste of 24 vertical pixels
Soon, we'll be stuck dealing with whatever UI Mozilla feels like implementing for the current week. Thanks for your efforts for now at least.
i value productivity above all else, so over the years, I've found that a familiar interface that i have muscle memory for is the most important aspect of browser ui/ux, imo
Could you please share your userChrome.css file or does it contain any sensitive data?
Ironically, it's now present in all other major browsers. Safari went above and beyond to make it a headline feature.
I look at the URL for the hostname, look at tabs to find the one I want and sometimes click on it (more often I use ctrl + pagedown/pageup), click on a certain extension GUI regularly because it lacks a keyboard UI, and that's about all but the edge cases. Everything else is done on the keyboard and sometimes the mouse scrollwheel (when presssing the spacebar doesn't work).
My guess is that browser GUI design has limited impact here; website GUI is a far bigger factor for me.
Changes which the user did not ask for and did not want are NON-CONSENSUAL.
Non-consensual changes foisted on the user are par for the course in proprietary software, but they should absolutely not be a thing in FOSS.
If the developer can repeatedly force non-consensual changes on users without any recourse, that means the project is not healthy and needs to be forked.
If it cannot be forked, then the idea of it being a FOSS project can be safely interred and a tombstone placed above it.
(Still using Firefox 3.x on occasion and ensuring my sites work with it.)
Modern interface philosophy means that if you don't use something every day you have no fucking idea what it does anymore. For example, every time I use the desktop Gmail interface I have to wait for tool tips just to figure out which inscrutable icon does what.
Damn do I hate Proton.
Of all the names they could've used, they chose one that's one Levenshtein distance from the previous. Looking forward to the next redesign dubbed Protan.
v4 was a ripoff of Opera's design at the time.
Australis was a Chrome UI clone.
The Photon design is my second favorite after v3.
I really hope Proton gets dropped. Worst one they've done since Australis.
Up until around Photon, I considered every UI update an improvement. Photon itself was a bit of a mixed bag, but mostly contained improvements. That marked the turning point for me; Photon is where I actively started disliking the more prominent features. Unlike Photon, Proton was just a huge step back, with no real improvements to compensate.
Trying to get details about a TLS certificate, something only power users do and even then is done very rarely, is now a full page ordeal, with a small column down the middle of the page for content. Apparently the popup was too clumsy or something? I seriously don't understand why any UI designer ever cared. Everything is now bring dumbed down, information that doesn't satisfy the top 80% just gets completely removed and the space saved is dedicated to padding and margins making the entire UI more bloated.
Then there were dumb plans to make Firefox interesting (I think? What else were they doing?) like the idea of "colourways" being available for a limited time only, which was simply preposterous. Earlier they took over Pocket, a service everyone has now seen the icon for and barely anyone ever uses. When it comes to decisions about the browser, everyone who dares reduce telemetry gets ignored . Mozilla brands itself as a privacy aware company, but when you try to use that privacy it turns its back on you.
I suppose I should be thankful Mozilla isn't trying to ruin Firefox in desktop as much as it's trying to push users to abandon Firefox on mobile. My tab bar has gone through three changes I don't really give a damn about, collections have come and gone, and I've given up on ever using the addon API again. The text encoding settings and menu seems to have inexplicably disappeared, leading to some very annoying broken web pages that can't be fixed anymore.
You can't even enter an IPv6 address on Android. The browser engine can browse to IPv6 addresses just fine, the URL bar just doesn't accept them. The regex to determine if an input is a URL or not doesn't test for IPv6 and the single attempt by someone else to fix it was apparently too complex and therefore removed later. That person abandoned their efforts and the whole process doesn't exactly make me want to contribute either.
In my frustration I've actually considered (and quickly dismissed) the idea of writing a tool to spam Mozilla with telemetry to get the features that I use more hits. Telemetry warfare, of sorts.
The funny thing is, extensions are completely functional on mobile. All you have to do is:
1. Create a collection at addons.mozilla.org containing exactly the addons you want 2. Install firefox for android, developer edition 3. Sign in to sync 4. go to "about firefox" and tap the logo 5 times 5. select "custom Add-on collection"
Yeah agree about the 'colorways' thing. I've read through their blog post a few times, and I still cannot understand why it was necessary or whether it accomplished its goals. https://blog.mozilla.org/en/products/firefox/introducing-new...
Nothing about this—the feature, and especially how they publicized it—made sense to me at first. Themes? OK, old feature, nothing to get excited about and I didn't even know they'd removed it in the first place, but sure. Limited time? Uh, what? Big announcement and interrupt my browsing session with a big ad? For a feature that wouldn't have been novel 25 years ago? None of it made sense.
... until they did the Disney tie-in colorway, at which point it all made sense. The original push must have been to generate "engagement" stats to show to potential advertisers, using a media push and user experience similar to what one would see for a paid advertising colorway.
https://user-images.githubusercontent.com/25581533/160292608... https://user-images.githubusercontent.com/25581533/160292618...
Each language has different behavior and does not display information properly.
1. Early (v1 ~ v3)
v1~v3 are classic UI that we remember when we were in the early 2000s.
Features
- Clear as the icons do only one thing
- Unique color for each icon
Limits
- Not fused with OS UI
- UI height
- Contrary to the modern interface philosophy as there is no abstraction
- Inconsistent icon size and texture
2. Classic (v4, 2016.10)
Thus, v4.0 was released after a large-scale UI reconstruction project was launched!!
It is the longest-lived UI and loved by many people.
Commonly called a Classic theme.
Features
- Orange app button at the top left like the symbol
- Calmer tone
- Win7 Aero Glass support
- Stop / Reload / Go with one button
- Tab moved to the top
Limits
- Unfamiliar interface with large-scale changes
3. Australis (v29, 2014.04)
Australis, which had a lot of likes and dislikes compared to Proton UI.
It was a change that put a lot of effort into simplicity.
Features
- Curved Tab
- Drag & Drop customizing UI
- Change settings UI pop-up to contents(tab) format
- Animation
Limits
- Panel UI that looks like a tablet
- Remove status bar
4. Photon (v57, 2017)
Photon was a generally well-received update to the UI that was used until June of this year(2021), when Proton appeared.
Features
- Components: List based panel and page actions, library menu
- Animation: Add animation to actions of buttons, tabs, panels, etc.
- Visual redesign: tabs, icons, density, etc.
- Performance: improved initialization, synchronization reflow, etc.
Limits
- It looks a bit complicated
- Only light weight themes are allowed.
5. Proton (v89, 2021.06)
This is the moment I started working on themes.
Features
- Neatly organized menu
- Icon that are pretty enough to remind of the edge
- Some of animations that I like & skeleton screens
- Stylish color scheme
- Moderate rounding
- Meticulous implementation
Limits
- Excessively wide padding
- Remove icons from menu
- Feels like a button rather than a tab with a connected look
- Confusion of tab indicators
- Remove page action menu in address bar
- Delete bookmarks / library animations and illustrations
- The icon size of the new tab contents
- Changed the search bar of the new tab to be performed on the URL bar.
Missed still! On large screens I prefer a permanent status bar with useful info persisting, than a URL flashing over the page. Chrome did it first, then Firefox copied like a mirror. The status bar was oddly banished to UI hell, not even a config switch lifeline. Replaced with google's flickering lower left overlay.
I customized my firefox so all the "bars" are in the window-title.
...which just shows that there's no one-size-fits-all, and that being able to customise the UI is a very important feature.
My personal biggest pet peeve: I hate, Hate, HATE that they changed the default behavior of clicks into the address or search bar, from simply "put the cursor here" to "select all". Oh my goodness, do I hate this, even after all these months. It breaks a fundamental behavior of text fields since, I don't know, the beginning of time?
Even today, I don't know how many times I keep inadvertenly deleting the contents of my search bar when all I want to do is add another term at the end. My brain is probably just too old to learn something new, I guess.
Also annoying: with the new add-on system, you cannot even write a plugin to restore the previous sane behavior.
My solution: touch to select all, touch again to reveal sliders, drag the (only visible) left slider to the right to reveal some unselected text, touch the unselected area to get a cursor, touch+hold to drag cursor to end of URL, edit URL, profit! I've seen easier puzzles in puzzle games!
However, the scrollbars since v98 are a big issue, they no longer follow the system UI (I'm using KDE Plasma on Fedora) and sometimes they take the same color as the web page background, making them invisible. Unfortunately there doesn't seem to be an alternative other than the n-th Chromium-based browser.
I’ve never been able to leave hubdreds of tabs open in Firefox for any length of time. Eventually it will consume so much RAM or crash. Any Chrome based browser seems to handle hundreds of tabs open for days at a time without any problems.
The actual implementation is exemplified to read the Firefox source code.
- https://searchfox.org/mozilla-central/search?q=reload-to-sto...
On other hand Chrome didn't change, they kept a simple, effective and clean UI/UX from the beginning
Firefox followed Gnome's design philosophy mess, hence why users left the ship