Everything is clear, you know what's a button and what's not. Information density is also high, which is a good thing on a computer screen.
But the main thing is that Windows 9x felt responsive. The Windows widgets felt solid and performant, while "modern" UWP apps feel clunky and prone to breakage. And don't even get me started on Electron.
Windows 9x did not feel ‘performant’. This is a false memory people seem to have.
Spinning rust hard drives were slow. It took ages to launch a program - loading screens typically had time to display a progress bar and a series of notes about what they were up to - ‘loading extensions’, ‘reticulating splines’, etc. Word would stall whenever it was autosaving. Carrying out an operation like spell checking or doing a find across a whole document or getting a word count took time.
Remember windows used to have an hourglass cursor? You used to have to watch that thing flip and empty multiple times when doing things like emptying the recycle bin.
Windows 9x was typically not running on a permanently networked computer. The computer wasn’t running a bunch of background network tasks like checking for updates or polling your email - generally it was just being slow because it could barely cope with running more than one program at once.
Thanks for reviving the reality of it. I saw the hourglass icon so many times, that I am able to picture all of its frames in my mind. Windows 9x might have been close to peak UI design (for me I'd say Windows 2000 is the sweet spot, and I always configured "Classic Style" on Windows XP and 7), but 95/98 was not responsive. And that's when it wasn't crashing.
Word 2.0 forever made me distrust Microsoft with an error: "File too large to save."
It was on a 20Mb hard drive, and the file was small enough to fit on a 1.44Mb floppy.
Turns out it was an obscure defect in Word: if you had quick save on, and the last page had a diagram or image, it would choke. Quick save or fast save - I forget what it was called. This would append deltas to your file, I think.
> Everything is clear, you know what's a button and what's not. Information density is also high, which is a good thing on a computer screen.
I would say information density was too high. All those always-on indicators: 3D scrollbars, buttons, etc. create a very busy picture. Today's interfaces are much cleaner which comes at a price of less information and hence, more ambiguity, but I for would rather pay that price than go back.
One problem I see is that while the UI itself has been simplified, incidental complexity has crept in other ways. Most importantly, the OSes themselves as software systems have clearly grown ponderous and unwieldy so that today they are more bugs and more of those bugs can be subtle and surprising. Also, there is less uniformity in UX across apps (and UI frameworks).
> I would say information density was too high. All those always-on indicators: 3D scrollbars, buttons, etc. create a very busy picture.
Have you been in nature recently? We've evolved to deal with very busy pictures and parse relevant information from them.
New UIs often don't have relevant information - like what is clickable or scrollable - and that's a problem.
It used to be that grandpa couldn't find a button with his poor eyesight, since everything was cramped and too tiny on the small screens we used to have.
Nowadays the clickable label is hidden behind a tiny hamburger menu that you can't tell apart from a mere stylistic flourish on the massive screens we keep mostly empty. Now neither of us can find the button.
That's progress? Every time I open an old application I breathe a sigh of relief because I can feel the cognitive load decreasing. Not having to put myself in the headspace of the designer to figure out what random geometric shape happens to be interactive is like taking a vacation.
"Today's interfaces are much cleaner which comes at a price of less information and hence, more ambiguity, but I for would rather pay that price than go back."
If you take today's interfaces to an extreme, you would get a white sheet. Very clean, but unusable. I wouldn't call interfaces "clean" where users increasingly have trouble figuring out what's clickable, how to scroll, move or resize a window.
I’ve been buying the same model of Brooks (the Ghost one) for _years_ because they fit well for me and while they release a refresh every other year or so they feel the same to me when I buy a new pair.
Sure hope they keep producing them because finding a different model that works for me would be kind of shit
> Sadly, I don't know anymore what kind of web service that was.
I don't know that either, but I remember there were websites specifically for that purpose, where you could look up a file extension and what program to open it with.
I imagine that the database for the online file association service for Windows XP has been lost to time, but for those who remember it: was it any good?
It seems like quite a good idea now -- if I remember correctly, Windows as of current seems to suggest a generic Bing search, which brings up all the spam "What extension is XXX?" sites.
That could have changed; I haven't really used Windows after 11's debut.
I wish this was more of a full tour than a random sampler of Windows versions. For instance I feel like Windows Me introduced the online search but it's been like... 26 years... since I've used it.
> Now there's a big gap. I don't have access to anything between Windows XP and Windows 10. So, Windows 10 (2015) is next
I'm guess these are just what the author already had set up. They're not really difficult to find or set up in a VM...
Can it really be called evolution? That sort of implies an improvement.
Interestingly enough, the default GTK file chooser also sucks. I notice
this nowadays because I broke something in my setup but I don't know
what, and the default file chooser does not remember anything I do.
Prior to that I found out that for opening files via the browser,
I need to have e. g. xdg-desktop-portal-gtk running. Well, my browser
never told me that; it just silently failed to download anything, I
could not choose any local file for file upload. I only found out
eventually, but when I found out, the fix was easy, but still, the
question is why such things break silently. This is simply incredibly
poor engineering and design, and that happens on linux too. That way
they'll never achieve linux desktop of the year. The decision makers
here are just horribly bad at designing anything. The whole GTK team
fell victim to this, now that it is a GNOMEy toolkit only.
When we ever get one-toolkit-that-fixes-everything (well ...), hopefully
they are really allowing only mega-smart people who can think objectively
and try to IMPROVE things rather than regress or take away functionality
willy-nilly style (as the GNOMEy devs do).
One thing that drives me nuts every time I save a new file or 'save as'. File chooser appears and I automatically start typing to enter/change the filename. But the filter input always gets focus and now I'm filtering the list of visible files instead of naming my file...
(Yes, I know I could try to submit a PR but I don't have the energy to figure out the Gnome governance process.)
at this point GNOME is a joke. I'm not sure KDE is always better but at least KDE is clearly trying. btw Wayland is also a project of the GNOME team and almost everyone is using the KDE extensions to it to make it suck less.
I daily drive GNOME on Ubuntu 22 on a daily basis and its UX is the favorite of all the UIs I've ever used, and I've used classic mac, OS X, all windows versions, xfce, kde and a trillion programs. And both android and ios devicess, both from early versions onto recent.
And GNOME really shines here. I'm on X11 though. Wayland lagged my mouse when I tried it years back, so I gave up on Wayland. Maybe they've fixed the lag spikes.
Most works GNOME did are somewhat invisible to users. GNOME is the main driving force behind immutable OS and containerised sandboxed apps, both are intended to make the OS maintain itself and simplify software management to single/zero click.
Valve adopted them afterwards and now everyone in the KDE team wants to join the ride.
I'm not sure you can blame the GNOME people for not trying. Personally, I feel like they are actively trying to make it worse, ie they are not passive.
Their approach to file system dialogs (open file/folder, save file),
was what got me to finally realize, that I must necessarily prefer KDE, because I must necessarily prefer anything that is not GNOME.
Now if I could only figure out how to stop KDE from opening file dialogs ALWAYS BEHIND ALL OTHER WINDOWS, that would be even better, but you can't have it all..
One day, when I die, and go to heaven or hell, when I arrive, my first question to the ones receiving me, will be "Finally, tell me - is there ANY possible way to navigate upwards to the parent folder, in GNOME?"
> One day, when I die, and go to heaven or hell, when I arrive, my first question to the ones receiving me, will be "Finally, tell me - is there ANY possible way to navigate upwards to the parent folder, in GNOME?"
You click the previous folder in the navigation bar.
evolution need not necessarily be an improvement, foe example mammals have a blind spot in their eyes because of the optic nerve, octopuses don't have a blind spot. but octopuses were before mammals
Biological evolution has indeed lead much more frequently to simplified, streamlined structures good for a single purpose, which had evolved from more complex and more versatile structures, than to more complex structures that had evolved from simple structures.
The latter kind of evolution events, while very rare, had a greater importance by being the origin of various kinds of very successful living beings.
Your example shows that because evolution proceeds through random search through the space of solutions, inside the neighborhoods of the starting point, followed by the choice of the best solution among the candidates, it frequently fails to find a global optimum, but it remains stuck on a local optimum.
However, octopuses were not before mammals. Both octopuses and mammals had appeared around the middle of the Mesozoic, but this is not really relevant for their eyes, which already existed in much older ancestors, hundreds of millions of years earlier.
Cephalopods and vertebrates with complex eyes already existed during the Ordovician. Chordates with complex eyes might have already existed quite early during the Cambrian, most likely before the separation between cephalopods and other mollusks, at a time when mollusks must have had only simple eyes that could detect light and perhaps the shape of shadows, but which could not form images.
When cephalopods separated from the other mollusks, they did this by evolving the ability to swim, instead of being forced to crawl on the bottom like most mollusks. (Swimming was achieved by filling their shell with a gas, which made it buoyant, while the other mollusks were held on the bottom by the weight of the shell.)
Chordates have also separated from their ancestors by evolving the ability to be fast swimmers (the elastic and incompressible dorsal chord reduced the energetic cost of anguilliform swimming in comparison with that for worm-like bodies that need to contract a muscular layer in order to prevent the shortening of the body when it is flexed).
This is likely to not be a coincidence, so the evolution of complex eyes in chordates and cephalopods is likely to be linked with the evolution of swimming in both groups, which made important the detection of objects located in various directions, while for a bottom crawler it could have been sufficient to sense when a shadow appeared due to something coming above it.
I think Windows 9x was peak Windows.
Everything is clear, you know what's a button and what's not. Information density is also high, which is a good thing on a computer screen.
But the main thing is that Windows 9x felt responsive. The Windows widgets felt solid and performant, while "modern" UWP apps feel clunky and prone to breakage. And don't even get me started on Electron.
Edit. See OP's previous article here, he managed to capture what I was trying to say in more details, with nice screenshots: https://movq.de/blog/postings/2026-06-16/0/POSTING-en.html
Windows 9x did not feel ‘performant’. This is a false memory people seem to have.
Spinning rust hard drives were slow. It took ages to launch a program - loading screens typically had time to display a progress bar and a series of notes about what they were up to - ‘loading extensions’, ‘reticulating splines’, etc. Word would stall whenever it was autosaving. Carrying out an operation like spell checking or doing a find across a whole document or getting a word count took time.
Remember windows used to have an hourglass cursor? You used to have to watch that thing flip and empty multiple times when doing things like emptying the recycle bin.
Windows 9x was typically not running on a permanently networked computer. The computer wasn’t running a bunch of background network tasks like checking for updates or polling your email - generally it was just being slow because it could barely cope with running more than one program at once.
Thanks for reviving the reality of it. I saw the hourglass icon so many times, that I am able to picture all of its frames in my mind. Windows 9x might have been close to peak UI design (for me I'd say Windows 2000 is the sweet spot, and I always configured "Classic Style" on Windows XP and 7), but 95/98 was not responsive. And that's when it wasn't crashing.
Word 2.0 forever made me distrust Microsoft with an error: "File too large to save."
It was on a 20Mb hard drive, and the file was small enough to fit on a 1.44Mb floppy.
Turns out it was an obscure defect in Word: if you had quick save on, and the last page had a diagram or image, it would choke. Quick save or fast save - I forget what it was called. This would append deltas to your file, I think.
Nah, that goes to either Windows 2000 or Windows 7.
Windows 2000 Professional was peak Windows for me. Skipped XP etc thanks to mac. Then 7 for a while.
> Everything is clear, you know what's a button and what's not. Information density is also high, which is a good thing on a computer screen.
I would say information density was too high. All those always-on indicators: 3D scrollbars, buttons, etc. create a very busy picture. Today's interfaces are much cleaner which comes at a price of less information and hence, more ambiguity, but I for would rather pay that price than go back.
One problem I see is that while the UI itself has been simplified, incidental complexity has crept in other ways. Most importantly, the OSes themselves as software systems have clearly grown ponderous and unwieldy so that today they are more bugs and more of those bugs can be subtle and surprising. Also, there is less uniformity in UX across apps (and UI frameworks).
> I would say information density was too high. All those always-on indicators: 3D scrollbars, buttons, etc. create a very busy picture.
Have you been in nature recently? We've evolved to deal with very busy pictures and parse relevant information from them.
New UIs often don't have relevant information - like what is clickable or scrollable - and that's a problem.
It used to be that grandpa couldn't find a button with his poor eyesight, since everything was cramped and too tiny on the small screens we used to have.
Nowadays the clickable label is hidden behind a tiny hamburger menu that you can't tell apart from a mere stylistic flourish on the massive screens we keep mostly empty. Now neither of us can find the button.
That's progress? Every time I open an old application I breathe a sigh of relief because I can feel the cognitive load decreasing. Not having to put myself in the headspace of the designer to figure out what random geometric shape happens to be interactive is like taking a vacation.
"Today's interfaces are much cleaner which comes at a price of less information and hence, more ambiguity, but I for would rather pay that price than go back."
If you take today's interfaces to an extreme, you would get a white sheet. Very clean, but unusable. I wouldn't call interfaces "clean" where users increasingly have trouble figuring out what's clickable, how to scroll, move or resize a window.
Products have to change in a visible way so people know they're new.
Same with running shoes, and they get worse too..
I’ve been buying the same model of Brooks (the Ghost one) for _years_ because they fit well for me and while they release a refresh every other year or so they feel the same to me when I buy a new pair.
Sure hope they keep producing them because finding a different model that works for me would be kind of shit
> Sadly, I don't know anymore what kind of web service that was.
I don't know that either, but I remember there were websites specifically for that purpose, where you could look up a file extension and what program to open it with.
I imagine that the database for the online file association service for Windows XP has been lost to time, but for those who remember it: was it any good?
It seems like quite a good idea now -- if I remember correctly, Windows as of current seems to suggest a generic Bing search, which brings up all the spam "What extension is XXX?" sites.
That could have changed; I haven't really used Windows after 11's debut.
If I recall correctly, it was pretty bad. It used to open Internet Explorer on this page: http://shell.windows.com/fileassoc/0409/xml/redir.asp?EXT={e... where {ext} was the extension.
Suggestions were vague and they only made sense with well-known filetypes.
There's a screenshot of how it looked here: https://protoweb.org/wp-content/uploads/2025/08/screen.jpg
I wish this was more of a full tour than a random sampler of Windows versions. For instance I feel like Windows Me introduced the online search but it's been like... 26 years... since I've used it.
> Now there's a big gap. I don't have access to anything between Windows XP and Windows 10. So, Windows 10 (2015) is next
I'm guess these are just what the author already had set up. They're not really difficult to find or set up in a VM...
I like your Windows postings, please don’t stop!
Something went horribly wrong on my computer and it takes about 2 minutes for that dialog to load.
raplayer.exe is stirring up some nostalgia...
Can it really be called evolution? That sort of implies an improvement.
Interestingly enough, the default GTK file chooser also sucks. I notice this nowadays because I broke something in my setup but I don't know what, and the default file chooser does not remember anything I do. Prior to that I found out that for opening files via the browser, I need to have e. g. xdg-desktop-portal-gtk running. Well, my browser never told me that; it just silently failed to download anything, I could not choose any local file for file upload. I only found out eventually, but when I found out, the fix was easy, but still, the question is why such things break silently. This is simply incredibly poor engineering and design, and that happens on linux too. That way they'll never achieve linux desktop of the year. The decision makers here are just horribly bad at designing anything. The whole GTK team fell victim to this, now that it is a GNOMEy toolkit only.
When we ever get one-toolkit-that-fixes-everything (well ...), hopefully they are really allowing only mega-smart people who can think objectively and try to IMPROVE things rather than regress or take away functionality willy-nilly style (as the GNOMEy devs do).
One thing that drives me nuts every time I save a new file or 'save as'. File chooser appears and I automatically start typing to enter/change the filename. But the filter input always gets focus and now I'm filtering the list of visible files instead of naming my file...
(Yes, I know I could try to submit a PR but I don't have the energy to figure out the Gnome governance process.)
at this point GNOME is a joke. I'm not sure KDE is always better but at least KDE is clearly trying. btw Wayland is also a project of the GNOME team and almost everyone is using the KDE extensions to it to make it suck less.
I daily drive GNOME on Ubuntu 22 on a daily basis and its UX is the favorite of all the UIs I've ever used, and I've used classic mac, OS X, all windows versions, xfce, kde and a trillion programs. And both android and ios devicess, both from early versions onto recent.
And GNOME really shines here. I'm on X11 though. Wayland lagged my mouse when I tried it years back, so I gave up on Wayland. Maybe they've fixed the lag spikes.
Most works GNOME did are somewhat invisible to users. GNOME is the main driving force behind immutable OS and containerised sandboxed apps, both are intended to make the OS maintain itself and simplify software management to single/zero click.
Valve adopted them afterwards and now everyone in the KDE team wants to join the ride.
I'm not sure you can blame the GNOME people for not trying. Personally, I feel like they are actively trying to make it worse, ie they are not passive. Their approach to file system dialogs (open file/folder, save file), was what got me to finally realize, that I must necessarily prefer KDE, because I must necessarily prefer anything that is not GNOME. Now if I could only figure out how to stop KDE from opening file dialogs ALWAYS BEHIND ALL OTHER WINDOWS, that would be even better, but you can't have it all..
One day, when I die, and go to heaven or hell, when I arrive, my first question to the ones receiving me, will be "Finally, tell me - is there ANY possible way to navigate upwards to the parent folder, in GNOME?"
> One day, when I die, and go to heaven or hell, when I arrive, my first question to the ones receiving me, will be "Finally, tell me - is there ANY possible way to navigate upwards to the parent folder, in GNOME?"
You click the previous folder in the navigation bar.
Same thing Windows did. Did anyone like it?
I never understood why file open/save dialog is a separate thing. I want to use my file manager to open and save files!
evolution need not necessarily be an improvement, foe example mammals have a blind spot in their eyes because of the optic nerve, octopuses don't have a blind spot. but octopuses were before mammals
Biological evolution has indeed lead much more frequently to simplified, streamlined structures good for a single purpose, which had evolved from more complex and more versatile structures, than to more complex structures that had evolved from simple structures.
The latter kind of evolution events, while very rare, had a greater importance by being the origin of various kinds of very successful living beings.
Your example shows that because evolution proceeds through random search through the space of solutions, inside the neighborhoods of the starting point, followed by the choice of the best solution among the candidates, it frequently fails to find a global optimum, but it remains stuck on a local optimum.
However, octopuses were not before mammals. Both octopuses and mammals had appeared around the middle of the Mesozoic, but this is not really relevant for their eyes, which already existed in much older ancestors, hundreds of millions of years earlier.
Cephalopods and vertebrates with complex eyes already existed during the Ordovician. Chordates with complex eyes might have already existed quite early during the Cambrian, most likely before the separation between cephalopods and other mollusks, at a time when mollusks must have had only simple eyes that could detect light and perhaps the shape of shadows, but which could not form images.
When cephalopods separated from the other mollusks, they did this by evolving the ability to swim, instead of being forced to crawl on the bottom like most mollusks. (Swimming was achieved by filling their shell with a gas, which made it buoyant, while the other mollusks were held on the bottom by the weight of the shell.)
Chordates have also separated from their ancestors by evolving the ability to be fast swimmers (the elastic and incompressible dorsal chord reduced the energetic cost of anguilliform swimming in comparison with that for worm-like bodies that need to contract a muscular layer in order to prevent the shortening of the body when it is flexed).
This is likely to not be a coincidence, so the evolution of complex eyes in chordates and cephalopods is likely to be linked with the evolution of swimming in both groups, which made important the detection of objects located in various directions, while for a bottom crawler it could have been sufficient to sense when a shadow appeared due to something coming above it.
I concur, I would say evolution is more like "adaption" than improvement.