> Me: "I don't like smartphone UIs. Everything is flat, nothing indicates where you can touch or not. I have to randomly try everything on the screen."
Response by non-tech person: "Well, yeah, of course you have to try everything? How else would this work?"
I think this goes deeper than many tech people realize.
From what I understood from talking with "nontechnical"(*) friends, relatives, etc, for a good potion of them, computers had always been "unpredictable magic". They got by through memorizing some very strict and rigid interaction sequences - "click this icon, then click that menu, then click that button, etc" and prayed nothing unexpected would happen. They were too scared and/or uninterested in computers to even try and find any rules or consistency in it.
I feel as if those nontechnical people "won" now. Now all UIs feel as inconsistent and unpredictable even for "techies" as any computer interaction felt to those people back then.
(* repeated from another thread: "nontechnical" in the "not fluent with PC use" sense, which is actually quite arrogant - they can have very high technical skill in other areas obviously)
> They were too scared and/or uninterested in computers to even try and find any rules or consistency in it.
Yes, and this is still true today. I work for a company with a large very non-technical user base. People absolutely will not explore, click on things outside of what they have memorized, or even try basic troubleshooting steps out of fear of breaking things.
It's actually really frustrating when trying to train to certain software or concepts because you have to lay everything out very explicitly, step by step. You cannot leave anything to the imagination or just assume that because the UI is "intuitive" that they will figure it out, because they won't even try.
I've encountered some form of that attitude and behavior at nearly every job I've had.
My parents never had any problem understanding where they can click on Windows 3.11. They never could understand how to interact with DOS GUIs.
Nowadays they have decades of experience with computers. They still can't predict what part of a web-site they can interact with, but they have memorized all the actions they can make on the phone apps they use.
Let’s look at the very website we on. Would you prefer for every clickable element here to be a button? Or even underlined as a link? Do you ever get confused navigating this website?
Given it’s a website mostly for experienced computer people who, like you, don’t really need those visual cues, don’t you think adding them would be superfluous?
It's a website for experienced computer people, and I appreciate its minimalist aesthetic, but that doesn't mean there's not enormous room for improvement.
Now that you've said it, I think PG would have designed a better site if he didn't hide the underlines.
But also, it wouldn't get any worse if the same design was kept and the underlines were enabled. And it would improve the usability for first-time users, but the nature of HN is that those are always very few.
Steve Jobs was right. Then when he died (after removing Scott Forstall), Jony Ive got to do his hardware minimalism in software too. And everything Steve Jobs favored was suddenly derided as “skeumorphism”. It’s like what USSR did with Stalin under Khrustchev. I still remember when Chrome app just had a big white area where you’re supposed to enter the url and you had no idea unless you randomly happened to click there. And if the website background was white, too? Oh too bad LMAO. Minimalistic! Chrome had no… chrome.
I don't remember when I first encountered it (maybe before ios7), but their cardinal sin for me was my struggling for 10 minutes to make the calculator app go into scientific mode, seeing absolutely no setting for that, before eventually discovering from a friend that you have to physically rotate the phone to landscape orientation to get the additional buttons.
> Since that button down there is called "Start", it implies that you can probably do something with it, maybe start programs? Click and you'll see the Start Menu:
Over time it seems like a lot of designs stop feeling the need to lead the user in this way. There is an assumption that by now everyone knows what the menu in the bottom left corner does, and we are no longer in the phase of trying to teach the population to use a computer for the first time.
I feel like this is the wrong approach. Every day there are new young people using a computer for the first time. Wouldn’t it be nice if all these conventions that evolved over the past 50 years could be intuitively discovered, instead of needing explanations from someone who already understands them?
Of course, as the world becomes more digital, many skeuomorphic designs become more abstract to those same young users. The floppy disk, the traditional telephone, even the file folder.
I was a wee little kid when I first touch windows 95, and only ever used dos before.
The "Start" button made no sense. The computer was already started, and clicking randomly popped up menus and opened documents in their right programs, so it felt like the natural way to progress. The owner of the computer had to point me to the start menu.
Even now I still think it was a cursed UI. It was the place primarily to close and shutdown the computer (again, when you see that button the computer and OS will always be already started), get to the control panel or run commands. None of it felt like "start", and the current windows logo only design makes a lot more sense.
To your point, small kids get proficient very fast with smartphones and iPads. I'd call their interface a lot more "intuitive"
The new iOS beta buries the “share” button under the line button inside the search lozenge, which was previously for view settings (and downloads). The iOS design team is just coasting on good decisions make a decade ago. The only good things are stuff they haven’t gotten around to fucking up yet.
I also think about this, and worry about this. My go-to example is that the file-system in Windows used to clearly be a tree, and the file explorer was how you traversed this tree. With "Libraries" and other shenanigans with the location of the Desktop folder, it's not much of a tree anymore, and I wonder if this is related to kids-these-days not understanding the basics of file systems anymore.
I think smartphones obscuring the filesystem away, and desktop search making the skill less of a requirement really hurt the filesystem understanding for younger generations.
I remember Steve Jobs saying the last area of complexity they needed to be solved was the filesystem, which is why they made the iPhone the way they did, with apps owning the files, so users didn’t have to deal with it. We’ve seen the Files app introduced and those walls get broken down, so it was clearly the wrong approach, especially when various apps can all perform actions on the same type of files.
Jobs also said death would take care of the problem of people not knowing how to type. I often think he should have taken the similar approach to the filesystem. Required learning for the modern era, not something to hide away so skills never develop.
> There is an assumption that by now everyone knows what the menu in the bottom left corner does, and we are no longer in the phase of trying to teach the population to use a computer for the first time
Strong disagree, because:
> Every day there are new young people using a computer for the first time
I can assure you these people have no idea what the start button is or does... it doesn't help that it no longer even says "Start" for the last ~20 years.
We do agree. I said the first thing you quoted was the wrong approach, because of the second thing you quoted. Having it say “start”, like it used to, would help solve that.
I really liked when XP came out, and you could have both the clean 2000 look with advanced font rendering like cleartype. That was the perfect combo for desktop use.
I had my xp running a blackbox 4win and coLinux/cygwin,
then moved onto vmware/virtualbox with windows vm and linux desktop,
now I'm doing win11+wsl2 and loving it for cuda/ai/building projects.
I agree that we had much better patterns back then. The software industry in general worked towards sharing visual paradigms, making use of system designs of their host playforms, facilitated discovery etc etc. All that was good and the recent trends moving us away from that consistency and discoverability are a detrement being steamrolled over by agents…
But I don’t agree that it “looked nice”. I hated Windows 95 and 2000’s “style”. They looked like engineers had made them. They looked stiff and unfriendly, eith too much border and outline. Real life has no outlines. I was in my late teens when 2000 came out. My friends and I jumped on it and felt it was the Os we had been waiting for.
But even then I thought it looked like shit.
The affordances were great. I agree that details like button depress and consistent scrollbars are valuable.
But I genuinely prefer things a bit rounder, a bit flatter, less grey, or late Aqua-style flat-with-shiny-affordances.
I agree that backgrounds should be flat (or very subtly textured so they recede but arn’t “boring; again, late-00s Mac OS nailed this for me).
What I’d really like to see is something new that takes the consistency of NT/2000 and Mac OSX prior to Lion, mixed with the novel affordances of BeOS/Haiku (docking windows, small title handles), and puts it through Apple’s “zing” (but not too far - transparency is highly overrated).
You will never see a visually consistent operating system again, the material conditions that allowed for older Windows and Mac OS versions to be consistent no longer exist. The user base for desktop computers is no longer growing, so the investment into new desktop UI technologies is largely in technologies like Electron and SwiftUI that allow developers to cut costs by reusing their existing web and mobile interfaces for desktop software. The focus on cross-platform development means that developers are far more concerned about making their software look consistent across all platforms than they are about whether their software looks native on each platform it supports.
Computer UIs needed borders and outlines because there are no brain-intuitive visual cues: no depth parallax, no shading, nothing shifts as you move your head, and until relatively recently they had poor contrast and brightness variability compared to the real world.
It was also a compromise for interface device limitations. We didn't have 4000 DPI mice with scroll wheels and 26 configurable buttons; you were lucky to have a 1024x768 resolution; and 16 bit color was for people shelling out $$$. Obvious borders and some padding between elements were a necessity to click what you intended to click.
Andreas Kling has said that one of his inspirations for SerenityOS was the Windows 2000 UI (https://corecursive.com/serenity-os-with-andreas-kling/). I found his general goal for SerenityOS ("Roughly speaking, the goal is a marriage between the aesthetic of late-1990s productivity software and the power-user accessibility of late-2000s *nix.") to be strangely validating ('Wait... So it's not just me?!'). And so of course I decided to try out the KDE desktop, which I had always kinda dismissed as being a bit too much of a niche within a niche. And it's great. It really is wonderful to use an OS that is designed from the ground up for serious technical users. And the ubiquity of web apps nowadays makes Linux a far more practical choice than it was back in the day.
Maybe more importantly, Win2k was the first windows version actually WORKED in a predictable way after years of unstable post-Win3.1 (Win95 and onward) production releases.
> Hiding filename extensions was one of the capital sins in computer history
Amen. The first thing I do on any (Windows) OS installation is make sure file extensions are shown. I guess Microsoft did that for "simplicity", but it also made for easy "virus.jpg.vbs" files.
Design language, like any language is metaphorical.
The thing that makes these skeumorphic designs work so well is that it kinda forces a consistent metaphor, and consistency above all else is huge for UX.
The fact that it's based on things we've seen in real life is also helps, as it means we can reason about the UI with the same faculties we've spent our entire life training.
Why are designers not understanding this these days?
I think one reason is that flat UI is super easy. Skeuomorphic is extremely hard to get right, and if you don't get it right it looks super tacky. Most people who have the word "designer" in their job title don't have the artistic skills needed to pull it off. This is why most designers are opposed to skeuomorphic.
> flat UI is super easy. Skeuomorphic is extremely hard to get right
What I don't understand is why those are treated as the only two choices?
Just adding some shadows, dividers, 3D buttons and real scroll bars again would go a long way to making things more usable without going full on into skeuomorphism to represent elements of the physical world.
A good example of the wrong direction was macOS in the switch to Tahoe. Buttons in modal dialog boxes became flat instead of 3D. They no longer look like buttons, they just look like a web-UI card with a gray background. There is no visual indicator at all that it is a clickable button.
Why? Legitimately, I want to hear from the designer(s) that made that decision and what their reasoning was.
Flat UI is also really difficult, as it can easily look cheap, boring, and unfinished.
Even Apple’s initial move to a flat UI in iOS 7 suffered from this. It took a long time to get refined to the point of not feeling like a major step backward. I still preferred the look of iOS 6 to anything that came afterwards. The skeuomorphic designs were warm, inviting, and fun. They served as a nice juxtaposition to the rather austere hardware.
Consistency. Even if you've never held a telephone receiver, if it means 'call' in one place, it's very likely to mean the same thing in another.
We could be using random hieroglyphs to the same ends, but people seem to always make their own (barring a few exceptions, like the hamburger menu). It's probably a better idea to use something with some grounding in reality rather than make your own from nothing, since doing that is hard, even for actual designers.
It also takes a long time for those purely digital ideas to filter out into the mainstream and get a shared name. For many years I would hear people same, “click the thing in the upper left corner”, “click those 3 lines up there”, or something similar. The term hamburger menu is starting to filter out there to the point where I feel comfortable using it. Even the term, hamburger menu, is a skeuomorphic name for a digital control. No one knows what to call those 3 lines, but everyone knows what a hamburger is. I’ve even seen some sites use a literal hamburger icon.
It also took a long time to standardize on the hamburger menu. Many also tried to use an ellipsis for a long time. Some still do. Sometimes those dots are vertical… is that a thin hamburger or a vertical ellipsis? I heard one person trying to make the term “tots” happen for this style of menu… tatter tots to go with the hamburger.
Contrast that to a “gear” menu for settings. They see a cog or a gear and everyone knows what that is without training, even if they aren’t a mechanic.
It doesn't really matter if they haven't used the original physical artifacts. If it looks physical you can figure out how to use it based on your knowledge of other physical objects you have used in your life.
Of course if you display for example, a spin dial like old telephones that has a particularly quirky way to use, them this doesn't apply.
You still need some design experience and "taste" though.
I've seen some B2B apps built in Java which used the Windows look-and-feel and looked absolutely awful: Actions wildly scattered through buttons, menus and context menus, panels and tabs nested several layers deep until the UI started to look like a canyon formation - and no icons or color at all, because I suppose those would have been "unprofessional" - so everything was in dull gray.
(I think it's worth realizing how colorful the stock Windows dialogs and applications actually are through the use of icons, even despite all widgets being gray.)
I still believe the Windows 2000-era UI toolkit is one of the best, because at least it gives you straightforward pathways to build a good-looking and usable UI - but you still have to want to do it.
No different than "windows" or "desktops" or "files." When was the last time you actually saw a file folder? Or a document separated by sheets with "tabs."
95% of computer users have never seen an ethernet cable, but they're still the symbol for networking.
99% of car drivers have never seen brake calipers, but they're part of the icon when my car's parking brake is on.
> In Windows 95, those toolbar icons were still actual buttons. In Windows 2000, they are recognizable as a button when activated, but in their default state they're not and you have to hover over them:
This is something I've struggled with as toolkits change and old widget themes stop working. There are still some decent themes out there (e.g. Skulpture for Qt has been my default for many years), and with a little patching they can be dragged into working on the latest toolkit versions. Yet I can't seem to avoid this "you have to hover over to see that it's actually a button" behaviour. Very annoying!
The loss of the theme menu and 'Windows classic' from Windows 8 onwards is dearly missed. But Windows classic hasn't gone away. If you run a 32-bit executable on Windows 10 or 11 under Windows XP compatibility mode, and set 'reduced colour mode', Windows Classic comes back. I have also noticed that when Adobe Acrobat crashes (heh) it momentarily flashes Windows classic on the title bars.
It's all still there. Bring it back, Microsoft. And put HiDPI and all your other modern technologies like D3D12 and borderless full-screen on it. I want to write old-school Win32 applications that fly.
I agree with the author's wish for visual cues when something is clickable, scrollable, etc.... This, on the other hand:
> Imitating real objects is good, too -- I don't have a single one of Android's "sliders" anywhere in my house, for example, so why don't you make this a checkbox, because writing down a check mark on paper is something that I actually do:
feels like an idea from a time when many people were encountering UIs on screens for the very first time as adults. I think the slider would be recognized as a toggle in its usual context of a settings screen by most people who have seen a settings screen before, but not that specific design for a toggle.
> I think the slider would be recognized as a toggle in its usual context of a settings screen by most people who have seen a settings screen before, but not that specific design for a toggle.
I've been using computers since the early 90s; we got our first home computer when I was four; I've used many different operating systems.
As a professional Web developer, it took me an embarrassingly long time to figure out what those slider widgets are supposed to mean. It's still very easy to get them wrong/confused (both as a user, and as the designer/dev making the form), e.g. when the affirmative state involves a negative setting, like "Mute" (does "on" mean "muted"; or does "on" refer to the audio, which I can use this to mute?)
There is a specific UI guideline for something that can be enabled or not: it's called a tick box. The on-off switch thing is a distinctly iOS invention [1]. The funny thing is that OS X (now macOS) mostly held off using those 'switches' too, until fairly recently.
The switch widget is for settings where flipping the switch takes immediate effect, unlike a checkbox where you have to press some "OK", "Submit" or "Apply" button. In Windows 2000/Office UIs, this would've been shown as a button that could visibly switch between "pressed" and "not pressed" states, just like a push-down switch, but that was rarely used. The modern labeled switch design is a lot more intuitive than that.
I remember those times and there were a lot if windows that did not look like windows and buttons that did not look like buttons. Not in the apps provided by the OS itself, but in third-party software. It was the time of wild experiments, especially in software created by enthusiasts. We just tend to forget this stuff
It's not like the world was perfect before; it's just that it's gotten even worse. The lack of consistency has spread, to the point there barely is any.
I liked every version of Windows that I’ve used, back to 3.1 all the way up to 11. Nowadays I mostly use 10/11 at home and at work. A new Windows upgrade was always a magical treat for me, the old slow HDD speeds building up anticipation during the installation phase, making me excited to finally use it. Growing up with computers unfortunately demystifies many things, but such is the price of poignance.
> I liked the UIs of the entire era from 3.0 to 2000, really. I'm mostly using Windows 2000 as an example here because it runs so well in QEMU/KVM and that allows me to easily take screenshots.
I agree with these points, but I will still be disagreeable and take a more nuanced position: Windows NT 5.0 Interim Developer Release build 1796 (and some preceding builds) were peak Windows UI, because they had all the points in the article, plus a "show desktop" button in the lower right corner that you could blindly throw your mouse at-- a feature relegated to a Quick Launch action from the subsequent build, and only restored in Windows 7.
When Windows XP came out, computer magazines wrote articles on how to switch the more modern, colorful UI back to the old, grey, drab, boring UI of Windows 95/98/2000.
I was young at the time and this seemed absurd to me. Why would you willingly use a UI that looks like wearing an old grey tie for a dusty office job in a depressing concrete building?
> Me: "I don't like smartphone UIs. Everything is flat, nothing indicates where you can touch or not. I have to randomly try everything on the screen."
Response by non-tech person: "Well, yeah, of course you have to try everything? How else would this work?"
I think this goes deeper than many tech people realize.
From what I understood from talking with "nontechnical"(*) friends, relatives, etc, for a good potion of them, computers had always been "unpredictable magic". They got by through memorizing some very strict and rigid interaction sequences - "click this icon, then click that menu, then click that button, etc" and prayed nothing unexpected would happen. They were too scared and/or uninterested in computers to even try and find any rules or consistency in it.
I feel as if those nontechnical people "won" now. Now all UIs feel as inconsistent and unpredictable even for "techies" as any computer interaction felt to those people back then.
(* repeated from another thread: "nontechnical" in the "not fluent with PC use" sense, which is actually quite arrogant - they can have very high technical skill in other areas obviously)
> They were too scared and/or uninterested in computers to even try and find any rules or consistency in it.
Yes, and this is still true today. I work for a company with a large very non-technical user base. People absolutely will not explore, click on things outside of what they have memorized, or even try basic troubleshooting steps out of fear of breaking things.
It's actually really frustrating when trying to train to certain software or concepts because you have to lay everything out very explicitly, step by step. You cannot leave anything to the imagination or just assume that because the UI is "intuitive" that they will figure it out, because they won't even try.
I've encountered some form of that attitude and behavior at nearly every job I've had.
My parents never had any problem understanding where they can click on Windows 3.11. They never could understand how to interact with DOS GUIs.
Nowadays they have decades of experience with computers. They still can't predict what part of a web-site they can interact with, but they have memorized all the actions they can make on the phone apps they use.
Let’s look at the very website we on. Would you prefer for every clickable element here to be a button? Or even underlined as a link? Do you ever get confused navigating this website?
> Would you prefer for every clickable element here to be a button? Or even underlined as a link?
Yes.
> Do you ever get confused navigating this website?
No, because I've been browsing the web for a while and know that every website does things their own way.
Given it’s a website mostly for experienced computer people who, like you, don’t really need those visual cues, don’t you think adding them would be superfluous?
It's a website for experienced computer people, and I appreciate its minimalist aesthetic, but that doesn't mean there's not enormous room for improvement.
No.
Now that you've said it, I think PG would have designed a better site if he didn't hide the underlines.
But also, it wouldn't get any worse if the same design was kept and the underlines were enabled. And it would improve the usability for first-time users, but the nature of HN is that those are always very few.
I have recently discovered a config option "Underline all links" in Firefox, and it's really nice, actually.
It started with iOS 7 and Jony Ive
Steve Jobs was right. Then when he died (after removing Scott Forstall), Jony Ive got to do his hardware minimalism in software too. And everything Steve Jobs favored was suddenly derided as “skeumorphism”. It’s like what USSR did with Stalin under Khrustchev. I still remember when Chrome app just had a big white area where you’re supposed to enter the url and you had no idea unless you randomly happened to click there. And if the website background was white, too? Oh too bad LMAO. Minimalistic! Chrome had no… chrome.
I don't remember when I first encountered it (maybe before ios7), but their cardinal sin for me was my struggling for 10 minutes to make the calculator app go into scientific mode, seeing absolutely no setting for that, before eventually discovering from a friend that you have to physically rotate the phone to landscape orientation to get the additional buttons.
> Since that button down there is called "Start", it implies that you can probably do something with it, maybe start programs? Click and you'll see the Start Menu:
Over time it seems like a lot of designs stop feeling the need to lead the user in this way. There is an assumption that by now everyone knows what the menu in the bottom left corner does, and we are no longer in the phase of trying to teach the population to use a computer for the first time.
I feel like this is the wrong approach. Every day there are new young people using a computer for the first time. Wouldn’t it be nice if all these conventions that evolved over the past 50 years could be intuitively discovered, instead of needing explanations from someone who already understands them?
Of course, as the world becomes more digital, many skeuomorphic designs become more abstract to those same young users. The floppy disk, the traditional telephone, even the file folder.
I was a wee little kid when I first touch windows 95, and only ever used dos before.
The "Start" button made no sense. The computer was already started, and clicking randomly popped up menus and opened documents in their right programs, so it felt like the natural way to progress. The owner of the computer had to point me to the start menu.
Even now I still think it was a cursed UI. It was the place primarily to close and shutdown the computer (again, when you see that button the computer and OS will always be already started), get to the control panel or run commands. None of it felt like "start", and the current windows logo only design makes a lot more sense.
To your point, small kids get proficient very fast with smartphones and iPads. I'd call their interface a lot more "intuitive"
The new iOS beta buries the “share” button under the line button inside the search lozenge, which was previously for view settings (and downloads). The iOS design team is just coasting on good decisions make a decade ago. The only good things are stuff they haven’t gotten around to fucking up yet.
I also think about this, and worry about this. My go-to example is that the file-system in Windows used to clearly be a tree, and the file explorer was how you traversed this tree. With "Libraries" and other shenanigans with the location of the Desktop folder, it's not much of a tree anymore, and I wonder if this is related to kids-these-days not understanding the basics of file systems anymore.
https://www.theverge.com/22684730/students-file-folder-direc...
I think smartphones obscuring the filesystem away, and desktop search making the skill less of a requirement really hurt the filesystem understanding for younger generations.
I remember Steve Jobs saying the last area of complexity they needed to be solved was the filesystem, which is why they made the iPhone the way they did, with apps owning the files, so users didn’t have to deal with it. We’ve seen the Files app introduced and those walls get broken down, so it was clearly the wrong approach, especially when various apps can all perform actions on the same type of files.
Jobs also said death would take care of the problem of people not knowing how to type. I often think he should have taken the similar approach to the filesystem. Required learning for the modern era, not something to hide away so skills never develop.
> Jobs also said death would take care of the problem of people not knowing how to type. I often think he should have taken the similar approach
I was all set for this dark humor, and actually had double back to then understand the full sentence.
> There is an assumption that by now everyone knows what the menu in the bottom left corner does, and we are no longer in the phase of trying to teach the population to use a computer for the first time
Strong disagree, because:
> Every day there are new young people using a computer for the first time
I can assure you these people have no idea what the start button is or does... it doesn't help that it no longer even says "Start" for the last ~20 years.
We do agree. I said the first thing you quoted was the wrong approach, because of the second thing you quoted. Having it say “start”, like it used to, would help solve that.
I really liked when XP came out, and you could have both the clean 2000 look with advanced font rendering like cleartype. That was the perfect combo for desktop use.
I had my xp running a blackbox 4win and coLinux/cygwin, then moved onto vmware/virtualbox with windows vm and linux desktop, now I'm doing win11+wsl2 and loving it for cuda/ai/building projects.
I agree that we had much better patterns back then. The software industry in general worked towards sharing visual paradigms, making use of system designs of their host playforms, facilitated discovery etc etc. All that was good and the recent trends moving us away from that consistency and discoverability are a detrement being steamrolled over by agents…
But I don’t agree that it “looked nice”. I hated Windows 95 and 2000’s “style”. They looked like engineers had made them. They looked stiff and unfriendly, eith too much border and outline. Real life has no outlines. I was in my late teens when 2000 came out. My friends and I jumped on it and felt it was the Os we had been waiting for.
But even then I thought it looked like shit.
The affordances were great. I agree that details like button depress and consistent scrollbars are valuable.
But I genuinely prefer things a bit rounder, a bit flatter, less grey, or late Aqua-style flat-with-shiny-affordances.
I agree that backgrounds should be flat (or very subtly textured so they recede but arn’t “boring; again, late-00s Mac OS nailed this for me).
What I’d really like to see is something new that takes the consistency of NT/2000 and Mac OSX prior to Lion, mixed with the novel affordances of BeOS/Haiku (docking windows, small title handles), and puts it through Apple’s “zing” (but not too far - transparency is highly overrated).
You will never see a visually consistent operating system again, the material conditions that allowed for older Windows and Mac OS versions to be consistent no longer exist. The user base for desktop computers is no longer growing, so the investment into new desktop UI technologies is largely in technologies like Electron and SwiftUI that allow developers to cut costs by reusing their existing web and mobile interfaces for desktop software. The focus on cross-platform development means that developers are far more concerned about making their software look consistent across all platforms than they are about whether their software looks native on each platform it supports.
Computer UIs needed borders and outlines because there are no brain-intuitive visual cues: no depth parallax, no shading, nothing shifts as you move your head, and until relatively recently they had poor contrast and brightness variability compared to the real world.
It was also a compromise for interface device limitations. We didn't have 4000 DPI mice with scroll wheels and 26 configurable buttons; you were lucky to have a 1024x768 resolution; and 16 bit color was for people shelling out $$$. Obvious borders and some padding between elements were a necessity to click what you intended to click.
Andreas Kling has said that one of his inspirations for SerenityOS was the Windows 2000 UI (https://corecursive.com/serenity-os-with-andreas-kling/). I found his general goal for SerenityOS ("Roughly speaking, the goal is a marriage between the aesthetic of late-1990s productivity software and the power-user accessibility of late-2000s *nix.") to be strangely validating ('Wait... So it's not just me?!'). And so of course I decided to try out the KDE desktop, which I had always kinda dismissed as being a bit too much of a niche within a niche. And it's great. It really is wonderful to use an OS that is designed from the ground up for serious technical users. And the ubiquity of web apps nowadays makes Linux a far more practical choice than it was back in the day.
Was this peak Windows UI?
I would say so, but the Active Dekstop stuff wasn't the right move.
Fisher-price came next, with Windows XP. At least you could easily switch back to classic.
And then Windows 8, we won't even talk about that.
Maybe more importantly, Win2k was the first windows version actually WORKED in a predictable way after years of unstable post-Win3.1 (Win95 and onward) production releases.
> I would say so, but the Active Dekstop stuff wasn't the right move.
Even so, you could completely ignore it if you wanted to!
Windows Vista / 7 was peak UI for me.
I really liked the luna silver and olive green themes. They were not too bad to look at.
The title bar of windows in Windows XP was Fisher-Price. But I thought the rest was OK.
I think Windows XP looks very nice if you install the Royale theme. It's a shiny and glassy version of the default XP style.
Almost. The NT5 RCs, which became windows 2000, were better IMO - not massive differences but it hadn’t been slobbered upon by marketing yet.
> Hiding filename extensions was one of the capital sins in computer history
Amen. The first thing I do on any (Windows) OS installation is make sure file extensions are shown. I guess Microsoft did that for "simplicity", but it also made for easy "virus.jpg.vbs" files.
Design language, like any language is metaphorical.
The thing that makes these skeumorphic designs work so well is that it kinda forces a consistent metaphor, and consistency above all else is huge for UX.
The fact that it's based on things we've seen in real life is also helps, as it means we can reason about the UI with the same faculties we've spent our entire life training.
Why are designers not understanding this these days?
I think one reason is that flat UI is super easy. Skeuomorphic is extremely hard to get right, and if you don't get it right it looks super tacky. Most people who have the word "designer" in their job title don't have the artistic skills needed to pull it off. This is why most designers are opposed to skeuomorphic.
Somewhere in between is the right approach. The NeXTSTEP UI from the late 1980s is what we need to return to. It still looks beautiful today: https://guidebookgallery.org/screenshots/openstep42
That does look surprisingly good. And dated. For some reason my first thought was "why the hell is this reminding me of Carmen San Diego?!"
> flat UI is super easy. Skeuomorphic is extremely hard to get right
What I don't understand is why those are treated as the only two choices?
Just adding some shadows, dividers, 3D buttons and real scroll bars again would go a long way to making things more usable without going full on into skeuomorphism to represent elements of the physical world.
A good example of the wrong direction was macOS in the switch to Tahoe. Buttons in modal dialog boxes became flat instead of 3D. They no longer look like buttons, they just look like a web-UI card with a gray background. There is no visual indicator at all that it is a clickable button.
Why? Legitimately, I want to hear from the designer(s) that made that decision and what their reasoning was.
Flat UI is also really difficult, as it can easily look cheap, boring, and unfinished.
Even Apple’s initial move to a flat UI in iOS 7 suffered from this. It took a long time to get refined to the point of not feeling like a major step backward. I still preferred the look of iOS 6 to anything that came afterwards. The skeuomorphic designs were warm, inviting, and fun. They served as a nice juxtaposition to the rather austere hardware.
What’s the meaning of skeuomorphic design for a generation that has never worked with the original physical artifacts they’re based on, though?
Consistency. Even if you've never held a telephone receiver, if it means 'call' in one place, it's very likely to mean the same thing in another.
We could be using random hieroglyphs to the same ends, but people seem to always make their own (barring a few exceptions, like the hamburger menu). It's probably a better idea to use something with some grounding in reality rather than make your own from nothing, since doing that is hard, even for actual designers.
It also takes a long time for those purely digital ideas to filter out into the mainstream and get a shared name. For many years I would hear people same, “click the thing in the upper left corner”, “click those 3 lines up there”, or something similar. The term hamburger menu is starting to filter out there to the point where I feel comfortable using it. Even the term, hamburger menu, is a skeuomorphic name for a digital control. No one knows what to call those 3 lines, but everyone knows what a hamburger is. I’ve even seen some sites use a literal hamburger icon.
It also took a long time to standardize on the hamburger menu. Many also tried to use an ellipsis for a long time. Some still do. Sometimes those dots are vertical… is that a thin hamburger or a vertical ellipsis? I heard one person trying to make the term “tots” happen for this style of menu… tatter tots to go with the hamburger.
Contrast that to a “gear” menu for settings. They see a cog or a gear and everyone knows what that is without training, even if they aren’t a mechanic.
It doesn't really matter if they haven't used the original physical artifacts. If it looks physical you can figure out how to use it based on your knowledge of other physical objects you have used in your life.
Of course if you display for example, a spin dial like old telephones that has a particularly quirky way to use, them this doesn't apply.
Following the standard is easy, deviating from it is hard.
Every designer nowadays insists on deviating.
You still need some design experience and "taste" though.
I've seen some B2B apps built in Java which used the Windows look-and-feel and looked absolutely awful: Actions wildly scattered through buttons, menus and context menus, panels and tabs nested several layers deep until the UI started to look like a canyon formation - and no icons or color at all, because I suppose those would have been "unprofessional" - so everything was in dull gray.
(I think it's worth realizing how colorful the stock Windows dialogs and applications actually are through the use of icons, even despite all widgets being gray.)
I still believe the Windows 2000-era UI toolkit is one of the best, because at least it gives you straightforward pathways to build a good-looking and usable UI - but you still have to want to do it.
Like the floppy disk for “save”? Or the old school phone receiver for “call”?
For icons and some aspects of functionality why not as a starting place?
Where it falls down is when designers force too many of the paradigms of the RL original onto a platform that doesn't suit it [0].
[0] See for example the Mac OS X Address Book app https://www.betalogue.com/2012/01/15/abook6-dumb/.
There's nothing wrong with either of those.
No different than "windows" or "desktops" or "files." When was the last time you actually saw a file folder? Or a document separated by sheets with "tabs."
95% of computer users have never seen an ethernet cable, but they're still the symbol for networking.
99% of car drivers have never seen brake calipers, but they're part of the icon when my car's parking brake is on.
> When was the last time you actually saw a file folder? Or a document separated by sheets with "tabs."
Every day because normal functioning adults still need to keep paper records of things.
> In Windows 95, those toolbar icons were still actual buttons. In Windows 2000, they are recognizable as a button when activated, but in their default state they're not and you have to hover over them:
This is something I've struggled with as toolkits change and old widget themes stop working. There are still some decent themes out there (e.g. Skulpture for Qt has been my default for many years), and with a little patching they can be dragged into working on the latest toolkit versions. Yet I can't seem to avoid this "you have to hover over to see that it's actually a button" behaviour. Very annoying!
The loss of the theme menu and 'Windows classic' from Windows 8 onwards is dearly missed. But Windows classic hasn't gone away. If you run a 32-bit executable on Windows 10 or 11 under Windows XP compatibility mode, and set 'reduced colour mode', Windows Classic comes back. I have also noticed that when Adobe Acrobat crashes (heh) it momentarily flashes Windows classic on the title bars.
It's all still there. Bring it back, Microsoft. And put HiDPI and all your other modern technologies like D3D12 and borderless full-screen on it. I want to write old-school Win32 applications that fly.
For me: everything! I clicked so well with it, everything made sense and was responsive.
I agree with the author's wish for visual cues when something is clickable, scrollable, etc.... This, on the other hand:
> Imitating real objects is good, too -- I don't have a single one of Android's "sliders" anywhere in my house, for example, so why don't you make this a checkbox, because writing down a check mark on paper is something that I actually do:
feels like an idea from a time when many people were encountering UIs on screens for the very first time as adults. I think the slider would be recognized as a toggle in its usual context of a settings screen by most people who have seen a settings screen before, but not that specific design for a toggle.
> I think the slider would be recognized as a toggle in its usual context of a settings screen by most people who have seen a settings screen before, but not that specific design for a toggle.
I've been using computers since the early 90s; we got our first home computer when I was four; I've used many different operating systems.
As a professional Web developer, it took me an embarrassingly long time to figure out what those slider widgets are supposed to mean. It's still very easy to get them wrong/confused (both as a user, and as the designer/dev making the form), e.g. when the affirmative state involves a negative setting, like "Mute" (does "on" mean "muted"; or does "on" refer to the audio, which I can use this to mute?)
There is a specific UI guideline for something that can be enabled or not: it's called a tick box. The on-off switch thing is a distinctly iOS invention [1]. The funny thing is that OS X (now macOS) mostly held off using those 'switches' too, until fairly recently.
[1]: https://freeimage.host/i/CxYBW6G
The switch widget is for settings where flipping the switch takes immediate effect, unlike a checkbox where you have to press some "OK", "Submit" or "Apply" button. In Windows 2000/Office UIs, this would've been shown as a button that could visibly switch between "pressed" and "not pressed" states, just like a push-down switch, but that was rarely used. The modern labeled switch design is a lot more intuitive than that.
Whose specific UI guideline are you talking about? Android's recommends the switch component for setting toggles: https://m3.material.io/components/switch/overview
It was clear, clean and understandable.
Buttons looked like buttons.
Windows (which have frames), looked like windows.
And there was no distracting design elements.
I remember those times and there were a lot if windows that did not look like windows and buttons that did not look like buttons. Not in the apps provided by the OS itself, but in third-party software. It was the time of wild experiments, especially in software created by enthusiasts. We just tend to forget this stuff
It's not like the world was perfect before; it's just that it's gotten even worse. The lack of consistency has spread, to the point there barely is any.
And crazy themes I had the matrix one with the "dodge this" sound instead of a ping... good times
I liked every version of Windows that I’ve used, back to 3.1 all the way up to 11. Nowadays I mostly use 10/11 at home and at work. A new Windows upgrade was always a magical treat for me, the old slow HDD speeds building up anticipation during the installation phase, making me excited to finally use it. Growing up with computers unfortunately demystifies many things, but such is the price of poignance.
Maybe relevant: http://windows93.net/
I still use this background Color on my Mac or Linux.
The article praises the UI, but isn't Windows 2000 using the Windows 95/98 UI with a different kernel?
Hold on, don't give cred it to Windows 95. The beveled 3D look was the invention of NeXT. Windows 95 copied everything from them. https://guidebookgallery.org/screenshots/openstep42
Didn't the Amiga Intuition widgets have a beveled 3D look well before NeXT did?
The different kernel for 95/98 was NT 4... Windows 2000 was unified, same UI for Consumer and Server
> I liked the UIs of the entire era from 3.0 to 2000, really. I'm mostly using Windows 2000 as an example here because it runs so well in QEMU/KVM and that allows me to easily take screenshots.
I agree with these points, but I will still be disagreeable and take a more nuanced position: Windows NT 5.0 Interim Developer Release build 1796 (and some preceding builds) were peak Windows UI, because they had all the points in the article, plus a "show desktop" button in the lower right corner that you could blindly throw your mouse at-- a feature relegated to a Quick Launch action from the subsequent build, and only restored in Windows 7.
Check it out: https://dn721308.ca.archive.org/0/items/usa_1796.1_winNT50.w...
win2k is the goat. fast, clear, no bullshit. sadly rdp and cleartype dragged me away and i've regretted it ever since.
When Windows XP came out, computer magazines wrote articles on how to switch the more modern, colorful UI back to the old, grey, drab, boring UI of Windows 95/98/2000.
I was young at the time and this seemed absurd to me. Why would you willingly use a UI that looks like wearing an old grey tie for a dusty office job in a depressing concrete building?
I was youngish at the time and thought Windows XP looked like clown vomit.
There is a German saying "Grün und blau schmückt die Sau" ~"green and blue grace the pig" - and XP was green and blue and shiny like hard candy.
Windows XP was the first OS I ever used and still I switched to the classic UI because it looked just better in my opinion.