I don’t understand how heat maps are used as a measuring tool, it seems pretty useless on its own. If something is little clicked, does it mean people don’t need the feature or people don’t like how it’s implemented? Or how do you know if people would really like something that’s not there to begin with?
It reminds about the Feed icon debacle: it’s been neglected for years and fell out of active use, which lead Mozilla to say “oh look, people don’t need the Feed icon, let’s move it away from the toolbar”. And then after a couple of versions they said “oh look, even less people use the Feed functionality, let’s remove it altogether”. Every time I see a click heatmap as a means to drive UI decisions I can’t shake the feeling that it’s only used to rationalize arbitrary product choices already made.
(P.S. I’ve been using Firefox since it was called Netscape and never understood why so many people left for Chrome, so no, I’m not just a random hater.)
Yeah, reminds me of some old Spiderman game where you could “charge” your jump to jump higher. They removed the visible charge meter in a sequel but kept the functionality, then removed the functionality in the sequel after that because nobody was using it (because newcomers didn’t know it was there, because there was no visible indication of it!).
It’s particularly annoying that the really cool things, which might actually have a positive impact for everyone – if not now, at least in a later release – are buried at the end of the announcement. Meanwhile, some of the things gathered through metrics would be hilarious were it not for the pretentious marketing language:
There are many ways to get to your preferences and settings, and we found that the two most popular ways were: 1) the hamburger menu – button on the far right with three equal horizontal lines – and 2) the right-click menu.
Okay, first off, this is why you should proofread/fact-check even the PR and marketing boilerplate: there’s no way to get to your preferences and settings through the right-click menu. Not in a default state at least, maybe you can customize the menu to include these items but somehow I doubt that’s what’s happening here…
Anyway, assuming “get to your preferences and settings” should’ve actually been “do things with the browser”: the “meatball” menu icon has no indication that it’s a menu, and a fourth way – the old-style menu bar – is hidden by default on two of the three desktop platforms Firefox supports, and isn’t even available on mobile. If you leave out the menubar through sheer common sense, you can skip the metrics altogether, a fair dice throw gets you 66% accuracy.
People love or intuitively believe what they need is in the right click menu.
I bet they’ll get the answer to this dilemma if they:
Look at the frequency of use for the “Copy” item in the right-click menu, and
For a second-order feature, if they break down right-click menu use by input device type and screen size
And I bet the answer has nothing to do with love or intuition ;-).
I have also divined in the data that the frequency of use for the right-click menu will further increase. The advanced machine learning algorithms I have employed to make this prediction consist of the realisation that one menu is gone, and (at least the screenshots show) that the Copy item is now only available in the right-click menu.
Out of those 17 billion clicks, there were three major areas within the browser they visited:
A fourth is mentioned in addition to the three in the list and, as one would expect, these four (out of… five?) areas are: the three areas with the most clickable widgets, plus the one you have to click in order to get to a new website (i.e. the navigation bar).
They use their UX experts & measurements to rationalize decisions done to make Firefox more attractive to (new) users as they claim, but … when do we actually see the results?
The market share has kept falling for years, whatever they claim to be doing, it is exceedingly obvious that they are unable to deliver.
Looking back, the only thing I remember Mozilla doing in the last 10 years is
a constant erosion of trust
making people’s lives miserable
running promising projects into the ground at full speed
I would be less bitter about it if Mozilla peeps wouldn’t be so obnoxiously arrogant about it.
Isn’t this article pretty off-topic, considering how many stories are removed for being “business analysis”?
This is pretty much “company losing users posts this quarter’s effort to attract new users by pissing off existing ones”.
The whole UI development strategy seems to be upside down: Firefox has been hermorrhaging users for years, at a rate that the UI “improvements” have, at best, not influenced much, to the point where a good chunk of the browser “market” consists of former Firefox users.
Instead of trying to get the old users back, Firefox is trying to appeal to a hypothetical “new user” who is technically illiterate to the point of being confused by too many buttons, but somehow cares about tracking through 3rd-party cookies and has hundreds of tabs open.
The result is a cheap Chrome knock-off that’s not appealing to anyone who is already using Chrome, alienates a good part of their remaining user base who specifically want a browser that’s not like Chrome, and pushes the few remaining Firefox users who don’t specifically care about a particular browser further towards Chrome (tl;dr if I’m gonna use a Chrome-like thing, I might as well use the real deal). It’s not getting anyone back, and it keeps pushing people away at the same time.
You’re not gonna win current users away from X by doing what X does, unless you do it much cheaper (not an option), or 10x better (hard to see how you could do more of chrome better than chrome).
The fundamental problem with Firefox is: It’s just slow. Slower than Chrome for almost everything. Slower at games (seriously, its canvas performance is really bad), slower at interacting with big apps like Google Docs, less smooth scrolling, even more latency between you hit a key on the keyboard and the letter shows up in the URL bar. This stuff can’t be solved with UI design changes.
Well, but there are reasons why it’s slow - and at least one good one.
Most notably, because Firefox makes an intentionally different implementation trade-off than Chrome. Mozilla prioritizes lower memory usage in FF, while Google prioritizes lower latency/greater speed.
(I don’t have a citation on me at the moment, but I can dig one up later if anyone doesn’t believe me)
That’s partially why you see so many Linux users complaining about Chrome’s memory usage.
These people are getting exactly what they asked for, and in an age where low CPU usage is king (slow mobile processors, limited battery life, more junk shoved into web applications, and plentiful RAM for people who exercise discipline and only do one thing at once), Chrome’s tradeoff appears to be the better one. (yes, obviously that’s not the only reason that people use Chrome, but I do see people noticing it and citing it as a reason)
I rarely use Google Docs; basically just when someone sends me some Office or Spreadsheet that I really need to read. It’s easiest to just import that in Google Docs; I never use this kind of software myself and this happens so infrequently that I can’t be bothered to install LibreOffice (my internet isn’t too fast, and downloading all updates for it takes a while and not worth it for the one time a year I need it). But every time it’s a frustrating experience as it’s just so darn slow. Actually, maybe it would be faster to just install LibreOffice.
I haven’t used Slack in almost two years, but before this it was sometimes so slow in Firefox it was ridiculous. Latency when typing could be in the hundreds or thousands of ms. It felt like typing over a slow ssh connection with packet loss.
CPU vs. memory is a real trade-off with a lot of various possible ways to do this and it’s a hard problem. But it doesn’t change that the end result is that for me, as a user, Firefox is sometimes so slow to the point of being unusable. If I had a job where they used Slack then this would be a problem as I wouldn’t be able to use Firefox (unless it’s fixed now, I don’t know if it is) and I don’t really fancy having multiple windows.
That being said, I still feel Firefox gives a better experience overall. In most regular use it’s more than fast enough; it’s just a few exceptions where it’s so slow.
That being said, I still feel Firefox gives a better experience overall. In most regular use it’s more than fast enough; it’s just a few exceptions where it’s so slow.
I agree. I absolutely prefer Firefox to Chrome - it’s generally a better browser with a much better add-on ecosystem (Tree Style Tabs, Container Tabs, non-crippled uBlock Origin) and isn’t designed to allow Google to advertise to you. My experience with it is significantly better than with Chome.
It’s because I like Firefox so much that I’m so furious about this poor design tradeoff.
(also, while it contributes, I don’t blame all of my slowdowns on Firefox’s design - there are many cases where it’s crippled by Google introducing some new web “standard” that sites started using before Firefox could catch up (most famously, the ShaddowDOM v0 scandal with YouTube))
I remember more things from Mozilla. One is also a negative (integration of a proprietary application, Pocket, into the browser; it may be included in your “constant erosion of trust” point), but the others are more positive.
Mozilla is the organization that let Rust emerge. I’m not a Rust programmer myself but I think it’s clear that the language is having a huge impact on the programming ecosystem, and I think that overall this impact is very positive (due to some new features of its own, popularizing some great features from other languages, and a rather impressive approach to building a vibrant community). Yes, Mozilla is also the organization that let go of all their Rust people, and I think it was a completely stupid idea (Rust is making it big, and they could be at the center of it), but somehow they managed to wait until the project was mature enough to make this stupid decision, and the project is doing okay. (Compare to many exciting technologies that were completely destroyed by being shut out too early.) So I think that the balance is very positive: they grew an extremely positive technology, and then they messed up in a not-as-harmful-as-it-could-be way.
Also, I suspect that Mozilla is doing a lot of good work participating to the web standards ecosystem. This is mostly a guess as I’m not part of this community myself, so it could have changed in the last decade and I wouldn’t know. But this stuff matters a lot to everyone, we need to have technical people from several browsers actively participating, it’s a lot of work, and (despite the erosion of trust you mentioned) I still trust the Mozilla standard engineers to defend the web better than Google (surveillance incentives) or Apple (locking-down-stuff incentives). (Defend, in the sense that I suspect I like their values and their view of the web, and I guess that sometimes this makes a difference during standardization discussion.) Unfortunately this part of Mozilla’s work gets weaker as their market share shrinks.
Agreed. I consider Rust a positive thing in general (though some of the behavioral community issues there seem to clearly originate from the Mozilla org), but it’s a one-off – an unexpected, pleasant surprise that Rust didn’t end in the premature death-spiral that Mozilla projects usually end up in.
Negative things I remember most are Persona, FirefoxOS and the VPN scam they are currently running.
I consider Rust a positive thing in general (though some of the behavioral community issues there seem to clearly originate from the Mozilla org), but it’s a one-off
Hard disagree there. Pernosco is a revolution in debugging technology (a much, much bigger revolution than what Rust is to programming languages) and wouldn’t exist without Mozilla spending engineering resources on RR. I don’t know much about TTS/STT but the Deepspeech work Mozilla has done also worked quite nicely and seemed to make quite an impact in the field. I think I also recall them having some involvement in building a formally-proven crypto stack? Not sure about this one though.
Mozilla has built quite a lot of very popular and impressive projects.
Negative things I remember most are Persona, FirefoxOS and the VPM scam they are currently running.
None of these make me as angry as the Mister Robot extension debacle they caused a few years ago.
To clarify, I didn’t mean it’s a one-off that it was popular, but that it’s a one-off that it didn’t get mismanaged into the ground. Totally agree otherwise.
They are trying to use the reputation of their brand to sell a service to a group of “customers” that has no actual need for it and barely an understanding what it does or for which purposes it would be useful.
What they do is pretty much the definition of selling snake oil.
I am a Firefox user and I’m interested in their VPN. I have a need for it, too - to prevent my ISP from selling information about me. I know how it works and what it’s useful for. I can’t see how they’re possibly “selling snake oil” unless they’re advertising something that doesn’t work or that they won’t actually deliver…
…which was my original question, which you sidestepped. Your words seem more like an opinion disguised as fact than actual fact.
It’s a tool like a lot of other things. Sure, you can abuse it in many ways, but unless we know how the results are used we can’t tell if it’s a good or bad scenario. A good usage for a heatmap could be for example looking at where people like to click on a menu item and how far should the “expand” button go.
As an event counter, they’re not great - they can get that info in better/cheaper ways.
This is tricky and also do for survey. I often am in a situation where it asks me “What do you have the hardest time with it” or “What prevents you from using language X on your current project?” and when the answer essentially boils down to “I am doing scripting and not systems programming” or something similar, I don’t intend to tell them that they should make a scripting language out of a systems language or vice versa.
And I know these are often taken wrongly, by reading the results and interpretation. There rarely is a “I like it how it is” option or a “Doesn’t need changes” or even “Please don’t change this!”.
I am sure this is true about other topics too, but programming language surveys seem to be a trend so that’s where I often see it.
I feel like they’re easily gamed, too. I feel like this happened with Twitter and the “Moments” tab. When they introduced it, it was in the top bar to the right of the “Notifications” tab. Some time after introduction, they swapped the “Notifications” and “Moments” tab, and the meme on Twitter was how the swap broke people’s muscle memory.
I’m sure a heat map would’ve shown that after the swap, the Moments feature suddenly became a lot more popular. What that heat map wouldn’t show was user intent.
from what I understand, the idea behind heat maps is not to decide about which feature to kill, but to measure what should be visible by default. The more stuff you add to the screen, the more cluttered and noisy the browser becomes. Heat maps help Mozilla decide if a feature should be moved from the main visible UX to some overflowing menu.
Most things they moved around can be re-arranged by using the customise toolbar feature. In that sense, you do have enough bits to make your browser experience yours to some degree.
The killing of feed icon was not decided with heat maps alone. From what I remember, that feature was seldom used (something they can get from telemetry and heat maps) but also was some legacy bit rot that added friction to maintenance and whatever they wanted to do. Sometimes features that are loved by few are simply in the way of features that will benefit more people, it is sad but it is true for codebases that are as old as Firefox.
Anyway, feed reading is one WebExtension away from any user, and those add-ons usually do a much better job than the original feature ever did.
I’m wondering how this whole heatmaps/metrics thing works for people who have customized their UI.
I’d assume that the data gained from e. g. this is useless at best and pollution at worst to Mozilla’s assumption of a perfectly spherical Firefox user.
@soc, I expect the browser to know it’s own UI and mark heat maps with context so that clicking on a tab is flagged the same way regardless if tabs are on top or the side. Also, IIRC the majority of Firefox users do not customise their UI. We live in a bubble of devs and power users who do, but that is a small fraction of the user base. Seeing what the larger base is doing is still beneficial.
worst to Mozilla’s assumption of a perfectly spherical Firefox user.
I’m pretty sure they can get meaningful results without assuming everyone is the same ideal user. Heat maps are just a useful way to visualise something, specially when you’re doing a blog post.
I don’t find it that tangible. If I was into speed, I’d be using Safari here which is quite fast. There are lots of different reasons to choose a browser. A lot of people switched to Chrome because of the constant advertising in Google webapps and also because Google has a tendency of breaking compatibility or reducing compatibility and performance with every other browser, thus making Google stuff work better on Chrome.
It’s a reference to Marie Kondo who found popularity with her methods of tidying up. She wrote a book and had (one? more?) popular reality TV show(s). She advocates for going through each of your possessions, asking if it “sparks joy” and to get rid of it if it doesn’t.
While I got the reference, I’m a bit apprehensive of a new Firefox re-design. I feel like I just got used to the last one. Admittedly that one was sorely needed. This time around I feel like it’s coming too soon and will waste my time unnecessarily. Hopefully it will still be as customizable as before though, so I can put it all back the way I like it.
A long time ago I worked for one of them dead tree paper magazines. I’m glad some things never change: twenty years later I still have to wonder just how the hell people come up with these things:
Today’s Firefox gets you where you want to go online
Of course it does, it’s a fscking browser!
The rest of the announcement is just a live demo of how metrics-driven development can backfire and it’s not even funny anymore and I don’t want to think about things that aren’t funny because I’ve had my coffee so I’ll shut up :(.
No, honestly it needed a refresh. Mozilla doesn’t seem to change their Firefox design anywhere near as much as google does with chrome. Which is nice but it does eventually get stale. When I first saw it on my personal laptop which I run beta on I loved it. I could not wait for it to get to stable for my other machines.
It seems like the tabs are blending with each other.
Why do you need clear definition? the second you hover over the middle (between the icon and the x) which is where I bet most people click even if there is clear definition, so honestly what is the difference? Seems like a quibble.
Not OP but I also hate this. I “need” clear definition so that I can read tab titles properly (otherwise the titles just blend into each other) and, more importantly, so that I “know” where to point my mouse at in order to switch to a new tab. The way you “need” clear definition on buttons, otherwise they just look like labels. I know there’s a hover animation but you have to get there in order to hover.
My initial reaction is that the tab titles are separated well enough thanks to the favicon between them. The screenshots honestly look pretty readable to me. I haven’t tried it for myself yet though, and this is all subjective anyways.
If the favicon is colourful and/or obvious enough, it’s not a problem. But in this day and age 8 out of 10 favicons are grayscale/monocolour anonymous icons, half of which are basically just a letter in a circle, so the favicon doesn’t look too different from the text. My eye vision is not the best, given that I haven’t been in the 18-25 target demographic for a while, so it’s not exactly the best mechanism for separation…
Yes, but… check out the new Add-On “Firefox Color”. It’s a point-and-click themer.
Admittedly, I don’t understand 2/3 of the choices, and the Firefox UI team seems to focus on completely different parts of the UI than I do, but I got it to make the current tab obviously different from the other tabs while keeping the other tabs high-contrast.
Yeah, on macos at least, the non-selected tabs sure could use some kind of demarcation. The selected tab has it, but on the non-selected tabs (especially with more than a few open tabs), it looks pretty messy.
I like the new design just fine. With the new look for tabs, I was almost tempted not to disable them in userChrome.css (I use Tree Style Tabs). But, I wonder how much of the emphasis on visual updates in Firefox comes from not having the engineering staff to do anything lower-level/high-impact anymore?
Not 100% sure if it was released in this release or not, but it looks like there’s a new 2-tone focus indicator. Love those because they show up on light and dark backgrounds. The old dotted line indicator could be hard to see.
Eh…the only change I’ve noticed is that, when switching vdesktops, firefox hangs for a moment on a stale frame before recovering. Sometimes it even requires explicit interaction before flushing the stale frame.
I just finished work and now it is dinner with the family so it was all written in a hurry but send me improvements, screenshots etc or comment on it and I’ll try to update it.
I use a sidebar tab extension in combination with some user chrome that hides the normal tabs. This works until it breaks.
I don’t know if bringing back the old extension API is the way to fix it, though. Wasn’t it removed because security? Extending the newer API to allow hiding the tab bar and putting a proper one (without weird UI) on the side would feel neater.
I don’t know if bringing back the old extension API is the way to fix it, though. Wasn’t it removed because security?
That is what I try to convey in my request to Mozilla that I linked above. I’ll take a look at it to see if I can make it clearer.
Edit: I see my suggestion might not be visible yet so I’ll paste it here:
Your idea is currently awaiting moderator review. Once approved, it will be visible to all users.
Extend the extension API so we can bring back the old fantastic extensions that made us love Firefox
by
Erik Itland
| 11 hours ago | in Customizing, extensions and themes
Hidden
Firefox used to be infinitely extendable. FTP handling was just another extension, and there were extensions for tab management, for changing the look and feel, for downloading entire web sites - or parts of them etc etc,
During the last few years a clean up have been done and while I recognize the reasons for it, it is now about time to fix the extension API, not by reintroducing the exact same old troubled API, but to look at what we have lost and then create the necessary improvements to the extension API so that enthusiasm once again can fill Firefox users, new and old.
Very specifically we could need, as a first step, to make available a way to move the tab strip around so that extensions like Tree Style Tabs once again can make users of others browsers envy us Firefox users instead of annoying us with a crowded but unused tab bar at the top of the screen.
I honestly don’t like the floating tab redesign. It looks and behaves more like a “button” to me than a “tab”. Now I feel like Mozilla broke the tabbed paradigm. Maybe that’s okay. Why do browsers have to have “tabs” and why not “buttons”? I’m not jiving with it though.
Ok I am going to be a bit conspiratorial here because I simply don’t believe you can have the stated goal of Mozilla and exhibit the behavior of Mozilla. That level of incompetence is outside of my “suspension of disbelief” so I must assume there will eventually be a “tell-all autobiography” that reveals the bribes and threats used to motivate them to self destruct.
People are incredibly loyal to Mozilla Firefox because there is no alternative but that loyalty is clearly misplaced. There is no hope being offered and now that they’ve fired their engineers I don’t believe there’s any way back.
I have no faith in the web. It’s going to get worse and then maybe, hopefully, replaced. In the meantime I’ll take my leave of it. Nyxt with noscript seems adequate for looking at documentation and for entertainment I’ll look elsewhere. Torrents still work and youtube-dl. In fact most of my (non-document-browsing) web usage has already been migrated off a browser.
I’ve never been a fan of Mozilla at all, and now I 100% believe Mozilla is only kept around to be Google’s strawman controlled opposition. Google could finish them off at almost any time- Mozilla’s revenue is mostly from them and Google has other means. My theory is they only keep them around to use to distract government regulators, so they avoid 90’s Microsoft’s fate.
I don’t understand how heat maps are used as a measuring tool, it seems pretty useless on its own. If something is little clicked, does it mean people don’t need the feature or people don’t like how it’s implemented? Or how do you know if people would really like something that’s not there to begin with?
It reminds about the Feed icon debacle: it’s been neglected for years and fell out of active use, which lead Mozilla to say “oh look, people don’t need the Feed icon, let’s move it away from the toolbar”. And then after a couple of versions they said “oh look, even less people use the Feed functionality, let’s remove it altogether”. Every time I see a click heatmap as a means to drive UI decisions I can’t shake the feeling that it’s only used to rationalize arbitrary product choices already made.
(P.S. I’ve been using Firefox since it was called Netscape and never understood why so many people left for Chrome, so no, I’m not just a random hater.)
Yeah, reminds me of some old Spiderman game where you could “charge” your jump to jump higher. They removed the visible charge meter in a sequel but kept the functionality, then removed the functionality in the sequel after that because nobody was using it (because newcomers didn’t know it was there, because there was no visible indication of it!).
It’s particularly annoying that the really cool things, which might actually have a positive impact for everyone – if not now, at least in a later release – are buried at the end of the announcement. Meanwhile, some of the things gathered through metrics would be hilarious were it not for the pretentious marketing language:
Okay, first off, this is why you should proofread/fact-check even the PR and marketing boilerplate: there’s no way to get to your preferences and settings through the right-click menu. Not in a default state at least, maybe you can customize the menu to include these items but somehow I doubt that’s what’s happening here…
Anyway, assuming “get to your preferences and settings” should’ve actually been “do things with the browser”: the “meatball” menu icon has no indication that it’s a menu, and a fourth way – the old-style menu bar – is hidden by default on two of the three desktop platforms Firefox supports, and isn’t even available on mobile. If you leave out the menubar through sheer common sense, you can skip the metrics altogether, a fair dice throw gets you 66% accuracy.
I bet they’ll get the answer to this dilemma if they:
And I bet the answer has nothing to do with love or intuition ;-).
I have also divined in the data that the frequency of use for the right-click menu will further increase. The advanced machine learning algorithms I have employed to make this prediction consist of the realisation that one menu is gone, and (at least the screenshots show) that the Copy item is now only available in the right-click menu.
A fourth is mentioned in addition to the three in the list and, as one would expect, these four (out of… five?) areas are: the three areas with the most clickable widgets, plus the one you have to click in order to get to a new website (i.e. the navigation bar).
They use their UX experts & measurements to rationalize decisions done to make Firefox more attractive to (new) users as they claim, but … when do we actually see the results?
The market share has kept falling for years, whatever they claim to be doing, it is exceedingly obvious that they are unable to deliver.
Looking back, the only thing I remember Mozilla doing in the last 10 years is
I would be less bitter about it if Mozilla peeps wouldn’t be so obnoxiously arrogant about it.
Isn’t this article pretty off-topic, considering how many stories are removed for being “business analysis”?
This is pretty much “company losing users posts this quarter’s effort to attract new users by pissing off existing ones”.
The whole UI development strategy seems to be upside down: Firefox has been hermorrhaging users for years, at a rate that the UI “improvements” have, at best, not influenced much, to the point where a good chunk of the browser “market” consists of former Firefox users.
Instead of trying to get the old users back, Firefox is trying to appeal to a hypothetical “new user” who is technically illiterate to the point of being confused by too many buttons, but somehow cares about tracking through 3rd-party cookies and has hundreds of tabs open.
The result is a cheap Chrome knock-off that’s not appealing to anyone who is already using Chrome, alienates a good part of their remaining user base who specifically want a browser that’s not like Chrome, and pushes the few remaining Firefox users who don’t specifically care about a particular browser further towards Chrome (tl;dr if I’m gonna use a Chrome-like thing, I might as well use the real deal). It’s not getting anyone back, and it keeps pushing people away at the same time.
The fallacy of Firefox, and quite a few other projects and products, seems to be:
The fallacy is that a lot of people are using your software is exactly because it’s not X and does Z instead of Y.
It also assumes that the popularity is because of Y, which may be the case but may also not be the case.
You’re not gonna win current users away from X by doing what X does, unless you do it much cheaper (not an option), or 10x better (hard to see how you could do more of chrome better than chrome).
You might; however, stop users from switching to X by doing what X does, even if you don’t do it quite as well.
The fundamental problem with Firefox is: It’s just slow. Slower than Chrome for almost everything. Slower at games (seriously, its canvas performance is really bad), slower at interacting with big apps like Google Docs, less smooth scrolling, even more latency between you hit a key on the keyboard and the letter shows up in the URL bar. This stuff can’t be solved with UI design changes.
Well, but there are reasons why it’s slow - and at least one good one.
Most notably, because Firefox makes an intentionally different implementation trade-off than Chrome. Mozilla prioritizes lower memory usage in FF, while Google prioritizes lower latency/greater speed.
(I don’t have a citation on me at the moment, but I can dig one up later if anyone doesn’t believe me)
That’s partially why you see so many Linux users complaining about Chrome’s memory usage.
These people are getting exactly what they asked for, and in an age where low CPU usage is king (slow mobile processors, limited battery life, more junk shoved into web applications, and plentiful RAM for people who exercise discipline and only do one thing at once), Chrome’s tradeoff appears to be the better one. (yes, obviously that’s not the only reason that people use Chrome, but I do see people noticing it and citing it as a reason)
I rarely use Google Docs; basically just when someone sends me some Office or Spreadsheet that I really need to read. It’s easiest to just import that in Google Docs; I never use this kind of software myself and this happens so infrequently that I can’t be bothered to install LibreOffice (my internet isn’t too fast, and downloading all updates for it takes a while and not worth it for the one time a year I need it). But every time it’s a frustrating experience as it’s just so darn slow. Actually, maybe it would be faster to just install LibreOffice.
I haven’t used Slack in almost two years, but before this it was sometimes so slow in Firefox it was ridiculous. Latency when typing could be in the hundreds or thousands of ms. It felt like typing over a slow ssh connection with packet loss.
CPU vs. memory is a real trade-off with a lot of various possible ways to do this and it’s a hard problem. But it doesn’t change that the end result is that for me, as a user, Firefox is sometimes so slow to the point of being unusable. If I had a job where they used Slack then this would be a problem as I wouldn’t be able to use Firefox (unless it’s fixed now, I don’t know if it is) and I don’t really fancy having multiple windows.
That being said, I still feel Firefox gives a better experience overall. In most regular use it’s more than fast enough; it’s just a few exceptions where it’s so slow.
I agree. I absolutely prefer Firefox to Chrome - it’s generally a better browser with a much better add-on ecosystem (Tree Style Tabs, Container Tabs, non-crippled uBlock Origin) and isn’t designed to allow Google to advertise to you. My experience with it is significantly better than with Chome.
It’s because I like Firefox so much that I’m so furious about this poor design tradeoff.
(also, while it contributes, I don’t blame all of my slowdowns on Firefox’s design - there are many cases where it’s crippled by Google introducing some new web “standard” that sites started using before Firefox could catch up (most famously, the ShaddowDOM v0 scandal with YouTube))
I’m interested in your citations :)
Here’s one about Google explicitly trading off memory for CPU that I found on the spot: https://tech.slashdot.org/story/20/07/20/0355210/google-will-disable-microsofts-ram-saving-feature-for-chrome-in-windows-10
I remember more things from Mozilla. One is also a negative (integration of a proprietary application, Pocket, into the browser; it may be included in your “constant erosion of trust” point), but the others are more positive.
Mozilla is the organization that let Rust emerge. I’m not a Rust programmer myself but I think it’s clear that the language is having a huge impact on the programming ecosystem, and I think that overall this impact is very positive (due to some new features of its own, popularizing some great features from other languages, and a rather impressive approach to building a vibrant community). Yes, Mozilla is also the organization that let go of all their Rust people, and I think it was a completely stupid idea (Rust is making it big, and they could be at the center of it), but somehow they managed to wait until the project was mature enough to make this stupid decision, and the project is doing okay. (Compare to many exciting technologies that were completely destroyed by being shut out too early.) So I think that the balance is very positive: they grew an extremely positive technology, and then they messed up in a not-as-harmful-as-it-could-be way.
Also, I suspect that Mozilla is doing a lot of good work participating to the web standards ecosystem. This is mostly a guess as I’m not part of this community myself, so it could have changed in the last decade and I wouldn’t know. But this stuff matters a lot to everyone, we need to have technical people from several browsers actively participating, it’s a lot of work, and (despite the erosion of trust you mentioned) I still trust the Mozilla standard engineers to defend the web better than Google (surveillance incentives) or Apple (locking-down-stuff incentives). (Defend, in the sense that I suspect I like their values and their view of the web, and I guess that sometimes this makes a difference during standardization discussion.) Unfortunately this part of Mozilla’s work gets weaker as their market share shrinks.
Agreed. I consider Rust a positive thing in general (though some of the behavioral community issues there seem to clearly originate from the Mozilla org), but it’s a one-off – an unexpected, pleasant surprise that Rust didn’t end in the premature death-spiral that Mozilla projects usually end up in.
Negative things I remember most are Persona, FirefoxOS and the VPN scam they are currently running.
Hard disagree there. Pernosco is a revolution in debugging technology (a much, much bigger revolution than what Rust is to programming languages) and wouldn’t exist without Mozilla spending engineering resources on RR. I don’t know much about TTS/STT but the Deepspeech work Mozilla has done also worked quite nicely and seemed to make quite an impact in the field. I think I also recall them having some involvement in building a formally-proven crypto stack? Not sure about this one though.
Mozilla has built quite a lot of very popular and impressive projects.
None of these make me as angry as the Mister Robot extension debacle they caused a few years ago.
To clarify, I didn’t mean it’s a one-off that it was popular, but that it’s a one-off that it didn’t get mismanaged into the ground. Totally agree otherwise.
Where have you found evidence that Mozilla is not delivering what they promise - a VPN in exchange for money?
They are trying to use the reputation of their brand to sell a service to a group of “customers” that has no actual need for it and barely an understanding what it does or for which purposes it would be useful.
What they do is pretty much the definition of selling snake oil.
I am a Firefox user and I’m interested in their VPN. I have a need for it, too - to prevent my ISP from selling information about me. I know how it works and what it’s useful for. I can’t see how they’re possibly “selling snake oil” unless they’re advertising something that doesn’t work or that they won’t actually deliver…
…which was my original question, which you sidestepped. Your words seem more like an opinion disguised as fact than actual fact.
It’s a tool like a lot of other things. Sure, you can abuse it in many ways, but unless we know how the results are used we can’t tell if it’s a good or bad scenario. A good usage for a heatmap could be for example looking at where people like to click on a menu item and how far should the “expand” button go.
As an event counter, they’re not great - they can get that info in better/cheaper ways.
This is tricky and also do for survey. I often am in a situation where it asks me “What do you have the hardest time with it” or “What prevents you from using language X on your current project?” and when the answer essentially boils down to “I am doing scripting and not systems programming” or something similar, I don’t intend to tell them that they should make a scripting language out of a systems language or vice versa.
And I know these are often taken wrongly, by reading the results and interpretation. There rarely is a “I like it how it is” option or a “Doesn’t need changes” or even “Please don’t change this!”.
I am sure this is true about other topics too, but programming language surveys seem to be a trend so that’s where I often see it.
I feel like they’re easily gamed, too. I feel like this happened with Twitter and the “Moments” tab. When they introduced it, it was in the top bar to the right of the “Notifications” tab. Some time after introduction, they swapped the “Notifications” and “Moments” tab, and the meme on Twitter was how the swap broke people’s muscle memory.
I’m sure a heat map would’ve shown that after the swap, the Moments feature suddenly became a lot more popular. What that heat map wouldn’t show was user intent.
from what I understand, the idea behind heat maps is not to decide about which feature to kill, but to measure what should be visible by default. The more stuff you add to the screen, the more cluttered and noisy the browser becomes. Heat maps help Mozilla decide if a feature should be moved from the main visible UX to some overflowing menu.
Most things they moved around can be re-arranged by using the customise toolbar feature. In that sense, you do have enough bits to make your browser experience yours to some degree.
The killing of feed icon was not decided with heat maps alone. From what I remember, that feature was seldom used (something they can get from telemetry and heat maps) but also was some legacy bit rot that added friction to maintenance and whatever they wanted to do. Sometimes features that are loved by few are simply in the way of features that will benefit more people, it is sad but it is true for codebases that are as old as Firefox.
Anyway, feed reading is one WebExtension away from any user, and those add-ons usually do a much better job than the original feature ever did.
I’m wondering how this whole heatmaps/metrics thing works for people who have customized their UI.
I’d assume that the data gained from e. g. this is useless at best and pollution at worst to Mozilla’s assumption of a perfectly spherical Firefox user.
@soc, I expect the browser to know it’s own UI and mark heat maps with context so that clicking on a tab is flagged the same way regardless if tabs are on top or the side. Also, IIRC the majority of Firefox users do not customise their UI. We live in a bubble of devs and power users who do, but that is a small fraction of the user base. Seeing what the larger base is doing is still beneficial.
I’m pretty sure they can get meaningful results without assuming everyone is the same ideal user. Heat maps are just a useful way to visualise something, specially when you’re doing a blog post.
The speed difference is tangible.
I don’t find it that tangible. If I was into speed, I’d be using Safari here which is quite fast. There are lots of different reasons to choose a browser. A lot of people switched to Chrome because of the constant advertising in Google webapps and also because Google has a tendency of breaking compatibility or reducing compatibility and performance with every other browser, thus making Google stuff work better on Chrome.
Invisible like the demarcations between tabs, apparently.
“sparking joy” is not very high on my list of things I want a web browser to do.
It’s a reference to Marie Kondo who found popularity with her methods of tidying up. She wrote a book and had (one? more?) popular reality TV show(s). She advocates for going through each of your possessions, asking if it “sparks joy” and to get rid of it if it doesn’t.
While I got the reference, I’m a bit apprehensive of a new Firefox re-design. I feel like I just got used to the last one. Admittedly that one was sorely needed. This time around I feel like it’s coming too soon and will waste my time unnecessarily. Hopefully it will still be as customizable as before though, so I can put it all back the way I like it.
A long time ago I worked for one of them dead tree paper magazines. I’m glad some things never change: twenty years later I still have to wonder just how the hell people come up with these things:
Of course it does, it’s a fscking browser!
The rest of the announcement is just a live demo of how metrics-driven development can backfire and it’s not even funny anymore and I don’t want to think about things that aren’t funny because I’ve had my coffee so I’ll shut up :(.
Definitely feels zippier than before, though I’m unable to discount confirmation bias here.
Maybe I won’t switch to Chrome after all.
Is anyone else having issues with the new tab style? It seems like the tabs are blending with each other.
If all your tabs are in containers (due to Temporary Containers especially), they’re all very clearly demarcated by the container color lines :D
No, honestly it needed a refresh. Mozilla doesn’t seem to change their Firefox design anywhere near as much as google does with chrome. Which is nice but it does eventually get stale. When I first saw it on my personal laptop which I run beta on I loved it. I could not wait for it to get to stable for my other machines.
Why do you need clear definition? the second you hover over the middle (between the icon and the x) which is where I bet most people click even if there is clear definition, so honestly what is the difference? Seems like a quibble.
Not OP but I also hate this. I “need” clear definition so that I can read tab titles properly (otherwise the titles just blend into each other) and, more importantly, so that I “know” where to point my mouse at in order to switch to a new tab. The way you “need” clear definition on buttons, otherwise they just look like labels. I know there’s a hover animation but you have to get there in order to hover.
My initial reaction is that the tab titles are separated well enough thanks to the favicon between them. The screenshots honestly look pretty readable to me. I haven’t tried it for myself yet though, and this is all subjective anyways.
If the favicon is colourful and/or obvious enough, it’s not a problem. But in this day and age 8 out of 10 favicons are grayscale/monocolour anonymous icons, half of which are basically just a letter in a circle, so the favicon doesn’t look too different from the text. My eye vision is not the best, given that I haven’t been in the 18-25 target demographic for a while, so it’s not exactly the best mechanism for separation…
Yes, but… check out the new Add-On “Firefox Color”. It’s a point-and-click themer.
Admittedly, I don’t understand 2/3 of the choices, and the Firefox UI team seems to focus on completely different parts of the UI than I do, but I got it to make the current tab obviously different from the other tabs while keeping the other tabs high-contrast.
I’m still tweaking it, but am fairly happy with the Acme-esque theme I’ve made. This is a neat little add on.
This helped.
Yeah, on macos at least, the non-selected tabs sure could use some kind of demarcation. The selected tab has it, but on the non-selected tabs (especially with more than a few open tabs), it looks pretty messy.
The tabs don’t feel less like tabs, more like random floating buttons, disconnected from the content.
Really digging the new UI.
Took a bit to get used to the new UI, but I like it.
I like the new design just fine. With the new look for tabs, I was almost tempted not to disable them in userChrome.css (I use Tree Style Tabs). But, I wonder how much of the emphasis on visual updates in Firefox comes from not having the engineering staff to do anything lower-level/high-impact anymore?
Not 100% sure if it was released in this release or not, but it looks like there’s a new 2-tone focus indicator. Love those because they show up on light and dark backgrounds. The old dotted line indicator could be hard to see.
Eh…the only change I’ve noticed is that, when switching vdesktops, firefox hangs for a moment on a stale frame before recovering. Sometimes it even requires explicit interaction before flushing the stale frame.
Slightly OT:
I experience this quite often with Firefox stable and nightly on Android 11 and 10.
Looks like that it hasn’t been shipped on official Fedora 34 repo
Most, if not all, Linux distros compile their own builds. Depending on the distro, it can sometimes take a full week :(
Free QA for Mozilla! And it makes sure distributions are actually providing only free software that can compile from source…
Feedback is also back after months (years?) so I have posted my feedback here: https://mozilla.crowdicity.com/post/720079
I just finished work and now it is dinner with the family so it was all written in a hurry but send me improvements, screenshots etc or comment on it and I’ll try to update it.
I use a sidebar tab extension in combination with some user chrome that hides the normal tabs. This works until it breaks.
I don’t know if bringing back the old extension API is the way to fix it, though. Wasn’t it removed because security? Extending the newer API to allow hiding the tab bar and putting a proper one (without weird UI) on the side would feel neater.
That is what I try to convey in my request to Mozilla that I linked above. I’ll take a look at it to see if I can make it clearer.
Edit: I see my suggestion might not be visible yet so I’ll paste it here:
Your idea is currently awaiting moderator review. Once approved, it will be visible to all users.
Extend the extension API so we can bring back the old fantastic extensions that made us love Firefox by Erik Itland | 11 hours ago | in Customizing, extensions and themes Hidden
Firefox used to be infinitely extendable. FTP handling was just another extension, and there were extensions for tab management, for changing the look and feel, for downloading entire web sites - or parts of them etc etc,
During the last few years a clean up have been done and while I recognize the reasons for it, it is now about time to fix the extension API, not by reintroducing the exact same old troubled API, but to look at what we have lost and then create the necessary improvements to the extension API so that enthusiasm once again can fill Firefox users, new and old.
Very specifically we could need, as a first step, to make available a way to move the tab strip around so that extensions like Tree Style Tabs once again can make users of others browsers envy us Firefox users instead of annoying us with a crowded but unused tab bar at the top of the screen.
I honestly don’t like the floating tab redesign. It looks and behaves more like a “button” to me than a “tab”. Now I feel like Mozilla broke the tabbed paradigm. Maybe that’s okay. Why do browsers have to have “tabs” and why not “buttons”? I’m not jiving with it though.
The menu items got huge padding; I miss a more compact design. I’m not on a touchscreen after all.
Ok I am going to be a bit conspiratorial here because I simply don’t believe you can have the stated goal of Mozilla and exhibit the behavior of Mozilla. That level of incompetence is outside of my “suspension of disbelief” so I must assume there will eventually be a “tell-all autobiography” that reveals the bribes and threats used to motivate them to self destruct.
People are incredibly loyal to Mozilla Firefox because there is no alternative but that loyalty is clearly misplaced. There is no hope being offered and now that they’ve fired their engineers I don’t believe there’s any way back.
I have no faith in the web. It’s going to get worse and then maybe, hopefully, replaced. In the meantime I’ll take my leave of it. Nyxt with noscript seems adequate for looking at documentation and for entertainment I’ll look elsewhere. Torrents still work and youtube-dl. In fact most of my (non-document-browsing) web usage has already been migrated off a browser.
I’ve never been a fan of Mozilla at all, and now I 100% believe Mozilla is only kept around to be Google’s strawman controlled opposition. Google could finish them off at almost any time- Mozilla’s revenue is mostly from them and Google has other means. My theory is they only keep them around to use to distract government regulators, so they avoid 90’s Microsoft’s fate.
Anyone else having soo much whitespace when opening multiple tabs ? picture