Live data from Hacker News

Designing Windows 95’s User Interface (2018)

socket3.wordpress.com

21–30 of 176 posts

Re: Designing Windows 95’s User Interface (2018)

#21
post #3

Earlier quoted context omitted.

I know, like can the UI/UX people please please PLEASE pay more attention to this stuff!

You'll find that the vast majority of bad UX decisions aren't coming from designers themselves, but from management. See also: A/B testing UIs to maximize arbitrary metrics

I wouldn't call those metrics arbitrary, they're just trying to measure something different. It's no longer measuring user productivity, it's measuring user engagement. It's no longer how quickly a user familiarizes and gets something done with the software, it's how often and how long the user uses the software.

Re: Designing Windows 95’s User Interface (2018)

#22

Some of the greatest UX of all time. Shame how far we have fallen in the current flat world.

A few people are trying to claim it back! My current favorite - https://github.com/dremin/RetroBar

This is unbelievable. Thanks so much for sharing.

Re: Designing Windows 95’s User Interface (2018)

#23

I think windows 95/98 era was the peak of UI discoverability. As a kid, I and my friends could learn and explore how a program worked just by expanding all the menus and opening the various dialogs. Dialogs which were filled with fields and their descriptions! Before then, it was the world of DOS and you had to know what you were doing ahead of time. And now, UIs are dynamic, initially hiding stuff from the user and…

Many dos programs with a gui had all the same menus and fields before Windows. They just ran one at a time.

Re: Designing Windows 95’s User Interface (2018)

#24
post #16

Earlier quoted context omitted.

You'll find that the vast majority of bad UX decisions aren't coming from designers themselves, but from management. See also: A/B testing UIs to maximize arbitrary metrics

Designers are the gatekeepers though, maybe they should stand up to these folks more often.

I think one of the main problems is that a surface-level argument can easily be made that user engagement represents the quality of the user experience (i.e. "users that spend a lot of time in the app clearly like it"), despite falling apart if you dig deeper than the surface. The allure of short-term profits makes such an argument attractive to upper leadership. It also makes it more difficult for a lower-level employee, like a designer, to convince them against it without direct evidence that optimizing for user productivity is better than optimizing for user engagement.

Combine the ease of making that argument with potential financial incentives to the designers to optimize for that, then you get relatively easy conformity to that goal. I'm sure there's plenty of designers that will happily accept it, so as a company, you can easily replace any designers that push back.

Re: Designing Windows 95’s User Interface (2018)

#25

I think windows 95/98 era was the peak of UI discoverability. As a kid, I and my friends could learn and explore how a program worked just by expanding all the menus and opening the various dialogs. Dialogs which were filled with fields and their descriptions! Before then, it was the world of DOS and you had to know what you were doing ahead of time. And now, UIs are dynamic, initially hiding stuff from the user and…

Now every step is full of dark pattern. Just like you wanna save a file in office. If you don't want store it in the cloud, there is always a step more. Also with the free One Drive and store it automaticly somewhere, I helped recently a friend (no computer guy), there are files just everywhere, some in the cloud some there and some somewhere else. Todays Windows is just hostile to non tech people, even to tech people.

Re: Designing Windows 95’s User Interface (2018)

#26

Back then you could test your UI on people that were a completely blank slate, having possible never used a computer before. You could focus on making the best possible UI there is. Now, users come already with certain habits and expectations developed by years of of computer use. This means, even if you really cared about the user experience of your customers you would only be able to make incremental improvements t…

> Back then you could test your UI on people that were a completely blank slate, having possible never used a computer before. You could focus on making the best possible UI there is.

This is a really great point. Like many others on this thread I miss "old" UI design, and like many posters can explain why. There's been a large transition to hiding UI, not marking interactable UI, and not spacially locating related UI elements.

Recently I watched one of my nieces use her phone. She was trying to do something, frustrated and telling me, but had not yet asked me to help. (I offer to help, but not to do something for them - to guide but let them achieve their goal.) I realised she was essentially tapping the screen looking for something that responded to taps. She didn't know what would react. Now, she knows lists can be scrolled, buttons etc can be clicked... but does she? Because there are no buttons, often, today and the button she needed, which I could see onscreen, was just text, no border indicating clickability.

I've heard anecdotally that the big UI designers in companies like Apple don't mind this because 'everyone already knows how to use their phones' and Macs / Windows. This ignores the young -- but we often ignore the young and expect them to learn -- and it ignores the elderly, those with bad eyesight, those who have not before now had the financial opportunity to own something, and so forth. There are lots, and always will be lots, of new people.

And I think that designing such that it can be used only by those who already know how to use it is a failure. Not just because there are always those who don't know, but because designing such that it can be used by anyone is more optimal. And finally, that designing such that it can be be used by _anyone_ means _everyone_ can use it. Phrased another way, another golden quote I cannot attribute is, 'accessibility is for everyone.' Which is a way of saying, designing for the non-average user who may have difficulties makes usage much easier for the average user too.

Re: Designing Windows 95’s User Interface (2018)

#27

Too bad they didn't read these notes before gutting the taskbar in Win11. It's so unusable, so slow when you can't "ungroup" stuff. Made me actually reinstall and go back to Win10.

Or the task bar grouping in Windows 7 that I've found most people actually don't prefer when you reveal that it's optional. A lot of people also like it, maybe as much as 40% or so, but when I look over someone's shoulder (who had been using 7 for ages already) and I wondered out loud how they manage to work with that task bar, they were usually happy to learn of a way to just see open programs instead of having this submenu structure plus a lot of unused task bar space, even if that meant getting used to a new way of working again.

Or the joke of a start menu from Windows 8 that presumably Ballmer liked a lot. You even had to make some hidden gesture (swish and flick, professor flitwick would say) to get to the settings only to find the 'shut down' option.

Windows 9x was a bit gray but otherwise mostly made sense.

Windows 10 seems to have gone back to the state of things at 7, except now the settings are at 200% zoom/whitespace and all over the place and I've heard everyone from beginner to guru wonder where the heck they hid xyz.

Windows 11 you've already covered.

Precisely this sort of thing is why I wondered in another comment (https://news.ycombinator.com/item?id=33776750) whether this type of testing happens anymore at all, or whether it is constrained in some way.

Re: Designing Windows 95’s User Interface (2018)

#28

Back then you could test your UI on people that were a completely blank slate, having possible never used a computer before. You could focus on making the best possible UI there is. Now, users come already with certain habits and expectations developed by years of of computer use. This means, even if you really cared about the user experience of your customers you would only be able to make incremental improvements t…

The beautiful thing is, you can still just make a UI that mimics Windows 95, and get all of the benefits of that research.

Re: Designing Windows 95’s User Interface (2018)

#29

I think windows 95/98 era was the peak of UI discoverability. As a kid, I and my friends could learn and explore how a program worked just by expanding all the menus and opening the various dialogs. Dialogs which were filled with fields and their descriptions! Before then, it was the world of DOS and you had to know what you were doing ahead of time. And now, UIs are dynamic, initially hiding stuff from the user and…

I think the big key in your comment is "As a kid,..". Maybe you think this is the easiest to learn/discover UI because it is the UI that you used to learn and discover. I see "kids these days" navigating around flat, touch-based interfaces with ease, in the same way that I found the desktop paradigm of the late-90s to be incredibly natural.

I think a lot of us came of age during that time, and as such see it as the ultimate UI.

Re: Designing Windows 95’s User Interface (2018)

#30
Again, weirdly, not a single word on how Windows 95 copied the title bar design and buttons from NeXTSTEP.

See, for instance, here: https://guidebookgallery.org/screenshots/openstep42#full

The “close window” (“×”) button is identical, and Windows 95 re-purposed NeXTSTEP’s “minimize” button into a “maximize” button. But you can still see why NeXTSTEP’s “minimize” button looks that way; it resembles a NeXTSTEP application icon, seen at the bottom on the above screenshot.

Windows 95 did not, sadly, copy the very useful behavior of the “close window” button in NeXTSTEP: It changed shape if the contents of a window was not saved, as seen here: https://guidebookgallery.org/screenshots/openstep42#textedit...

Post reply on HN