7 ms·

It seems like most people who scroll use touch, wheel, or other non-scroll-bar interfaces, and the scroll bars are just a visual indicator of document size and

It seems like most people who scroll use touch, wheel, or other non-scroll-bar interfaces, and the scroll bars are just a visual indicator of document size and position while navigating it. I know I do— and have for 20 years unless I was using a shitty laptop without a scrolling input method. Maybe unless I'm scrolling through literally millions of lines of text, which is infrequent enough that I'd prefer to have them hidden and revealable by scrolling a bit. Not having them take up screen real estate and working with most people's usage styles while being configurable for others is the right decision. The problem is systems and applications that don't let you configure them. People talk about accessibility and their preferences like they are absolute truths— it's a lot more complex. Not having scroll bars sucks for people that never got comfortable with other scrolling input methods and people who use sight interfaces, for example. They couldn't possibly be less relevant to people who use screen readers, and as an input method rather than an Univision visual indicator, only slightly more relevant to people on phones and tablets. Phone and tablet usage has outpaced regular computer usage for quite some time, and that informs users interaction style. Adding visible UI elements adds cognitive load, and for people who never worked with scroll bars out of necessity, they're just another bulky animated distraction on the screen. Most younger users would probably think them about as useful as an always-visible on-screen keyboard. I guarantee you— the designers at Apple who chose to hide those scroll bars probably thought more about scrolling more than every one of these commenters combined, and that was before they did formal usability studies.

bluGill 3y ago
Does anyone apply formal usability studies anymore? (not just do them - do they apply the results). I was trained in formal UX long ago, and many UIs today outright violate common rules we learned back then. High contrast was considered important for low vision users even if it looked ugly then. Now most UIs are lower contrast and while they look nicer, they are less usable. Of course I learned long ago and so I'm willing to accept that what was best practices back then might be wrong given new research that I'm not aware of. However this seems unlikely.
- asoneth 3y ago
  Formal usability testing is still practiced in enterprise, military, and professional products. From what I've seen from colleagues in the consumer space they've predominantly switched to A/B testing and statistical clickstream/log analyses, though large companies sometimes still do run lab studies for important decisions and to rationalize the giant one-way mirror they installed.
  - chefandy 3y ago
    It depends on how big/important/popular of a product you're talking about, but most software that your average joe uses (operating systems, native communication apps, national websites) likely either conduct significant usability testing or largely inform their designs using great usability testing from people like the mozilla foundation and the nielsen norman group. And no-- neither of those organizations have to be perfect for that to be a perfectly valid approach, and if you follow that guidance, you'll never end up with a 'very low contrast' interface.
- williamcotton 3y ago
  I go to the store to buy a hammer. I’m a hammer user and I use the hammer. Or I go to the store and I buy heroin. I’m a heroin user and I use the heroin. These UIs we speak of, are they hammers, or are they heroin? Because I’m currently looking at a high contrast screen using a mixture of Unix tools created from the late 60s to yesterday. I’m looking at an Excel spreadsheet made by a team of lawyers that is in a UI. The lawyers themselves highlighted and handled the contrast of the document. Beyond the table itself, Excel’s menu bar is easy to read for the basic operating needs of a spreadsheet. I mean come on, the pivot table UI is clearly masterful. (If Excel was a native plaintext tabular format and you could pipe stdin/out to the GUI app? Gimme!) However, a lot of software is basically useless these days. It’s not designed from the beginning to be productive in any shape or form. Some stoned dance by a junkie under a highway overpass is definitely some kind of expressive art form. How much more productive is the expressive art forms promoted by TikTok, Twitter or Twitch? There’s clearly a scale at play. I’m as big fan of Twitch as much as I’m a big fan of the Buffalo Bills. I’ve always appreciated the sociological aspects of sports. To add some context to my definition of productivity, I would say that group sporting events are overall productive from a sociological perspective, be it IPL cricket or Fortnite tournaments. TikTok is clearly and quite obviously towards the junkie dance scale of things. From an experiential position there was not much of a difference between my observations of a woman filming herself doing a weird little dance on the sidewalk while completely oblivious to the rest of the world and the junky on the next block doing a weird little dance on the sidewalk while completely oblivious to the rest of the world. This basic object-level assessment of the actions that we and others experience with our physical bodies is core to the very notion of “meaning”. The “meaning” is just the stoned dance. I know of no junkies famous for only being junkies but there are definitely TikTokers famous for only being TikTokers so don’t confuse this for some strict equivalence. I’m just trying to establish a kind of framework for discussing the relationship between productive UI and productive software tools.
lozenge 3y ago
The days of usability studies are long gone.* It's now change for the sake of change. Spot the indication that not all settings sections are shown in this screenshot, in all display settings and lighting conditions that your device might be used in. https://imgur.com/a/pMUKwjT https://imgur.com/a/pMUKwjT https://imgur.com/a/2ADNH2B https://imgur.com/a/2ADNH2B * Were they ever there? I can't believe this dialog background - https://arstechnica.com/gadgets/2001/10/macosx-10-1/4/ https://arstechnica.com/gadgets/2001/10/macosx-10-1/4/ - was the "winner" in any usability study. This was removed in 2003-11 with the release of 10.3.
- chefandy 3y ago
  a) That scroll bar decision was made a long time ago b) Problems with specific dialog boxes or dynamically created features is evidence of problems with those dialog boxes or dynamically created features. As someone who's worked both as a developer and a designer, often the engineers win. c) Assuming that these topics are simple enough for off-the-cuff reckonin' to validate or invalidate them are exactly why the engineers shouldn't win when it comes to interface decisions.
Sakos 3y ago
> the designers at Apple who chose to hide those scroll bars probably thought more about scrolling more than every one of these commenters combined, and that was before they did formal usability studies. This blind appeal to authority shouldn't convince anyone.
- chefandy 3y ago
  If you've got an actual point to make instead of just poking at one little part of what I said, I'd respond.
  - Sakos 3y ago
    That point is how you justify all your conjecture. Rethink what you wrote because all you're really doing is relying on anecdotes. I can literally say the opposite (I always use the scrollbar for scrolling and everybody I know does too) and we'd have made zero progress. You haven't actually made an argument worth arguing about. It's devoid of value and it's not a productive addition to the conversation. And don't bother wasting any more of my time. I don't care. You clearly have nothing to say.
    - chefandy 3y ago
      Yes. This is exactly the sort of 90 word response made by someone who doesn't have time to waste on arguments they don't care about.
CRConrad 3y ago
> Phone and tablet usage has outpaced regular computer usage for quite some time, and that informs users interaction style. "Informs"?!? Deforms! WTF does phone usage have to do with sabotaging the UI on a computer???
- JohnFen 3y ago
  Because designers and devs don't want to have to do multiple designs. Everyone's aiming for one design that works for everything. The problem is that such a thing is impossible.
  - chefandy 3y ago
    Vollying back your flipness-- This sort of reductive assumption about software design is why companies that need people who aren't developers to actually use their products hire interface designers.
    - deprecative 3y ago
      Most employers I've had are all about getting engineers to be everything. Designers, infrastructure, QA, UI/UX, and even product/project managers.. everything.
      - chefandy 3y ago
        Sounds like you need new employers. The only people who want that are shortsighted managers who don't care that they know nothing about users' needs, and that's certainly not a universal disposition in the software business.
        deprecative 3y ago
        This has been true of every employer I have ever had. It's true outside of "tech" as well but nearly all of my experience has been as a software engineer.
indymike 3y ago
> Phone and tablet usage has outpaced regular computer usage for quite some time This does not excuse making terrible user experiences elsewhere.
- chefandy 3y ago
  I think you're missing the point. Usability is not prescriptive. You implement what your users have the easiest time using. Touch interfaces change how people interact with interfaces. If you care about usability, you work with people on that and not against them.

bluGill 3y ago

Does anyone apply formal usability studies anymore? (not just do them - do they apply the results). I was trained in formal UX long ago, and many UIs today outright violate common rules we learned back then. High contrast was considered important for low vision users even if it looked ugly then. Now most UIs are lower contrast and while they look nicer, they are less usable. Of course I learned long ago and so I'm willing to accept that what was best practices back then might be wrong given new research that I'm not aware of. However this seems unlikely.

asoneth 3y ago
Formal usability testing is still practiced in enterprise, military, and professional products. From what I've seen from colleagues in the consumer space they've predominantly switched to A/B testing and statistical clickstream/log analyses, though large companies sometimes still do run lab studies for important decisions and to rationalize the giant one-way mirror they installed.
- chefandy 3y ago
  It depends on how big/important/popular of a product you're talking about, but most software that your average joe uses (operating systems, native communication apps, national websites) likely either conduct significant usability testing or largely inform their designs using great usability testing from people like the mozilla foundation and the nielsen norman group. And no-- neither of those organizations have to be perfect for that to be a perfectly valid approach, and if you follow that guidance, you'll never end up with a 'very low contrast' interface.
williamcotton 3y ago
I go to the store to buy a hammer. I’m a hammer user and I use the hammer. Or I go to the store and I buy heroin. I’m a heroin user and I use the heroin. These UIs we speak of, are they hammers, or are they heroin? Because I’m currently looking at a high contrast screen using a mixture of Unix tools created from the late 60s to yesterday. I’m looking at an Excel spreadsheet made by a team of lawyers that is in a UI. The lawyers themselves highlighted and handled the contrast of the document. Beyond the table itself, Excel’s menu bar is easy to read for the basic operating needs of a spreadsheet. I mean come on, the pivot table UI is clearly masterful. (If Excel was a native plaintext tabular format and you could pipe stdin/out to the GUI app? Gimme!) However, a lot of software is basically useless these days. It’s not designed from the beginning to be productive in any shape or form. Some stoned dance by a junkie under a highway overpass is definitely some kind of expressive art form. How much more productive is the expressive art forms promoted by TikTok, Twitter or Twitch? There’s clearly a scale at play. I’m as big fan of Twitch as much as I’m a big fan of the Buffalo Bills. I’ve always appreciated the sociological aspects of sports. To add some context to my definition of productivity, I would say that group sporting events are overall productive from a sociological perspective, be it IPL cricket or Fortnite tournaments. TikTok is clearly and quite obviously towards the junkie dance scale of things. From an experiential position there was not much of a difference between my observations of a woman filming herself doing a weird little dance on the sidewalk while completely oblivious to the rest of the world and the junky on the next block doing a weird little dance on the sidewalk while completely oblivious to the rest of the world. This basic object-level assessment of the actions that we and others experience with our physical bodies is core to the very notion of “meaning”. The “meaning” is just the stoned dance. I know of no junkies famous for only being junkies but there are definitely TikTokers famous for only being TikTokers so don’t confuse this for some strict equivalence. I’m just trying to establish a kind of framework for discussing the relationship between productive UI and productive software tools.

lozenge 3y ago

The days of usability studies are long gone.* It's now change for the sake of change. Spot the indication that not all settings sections are shown in this screenshot, in all display settings and lighting conditions that your device might be used in. https://imgur.com/a/pMUKwjT https://imgur.com/a/pMUKwjT https://imgur.com/a/2ADNH2B https://imgur.com/a/2ADNH2B * Were they ever there? I can't believe this dialog background - https://arstechnica.com/gadgets/2001/10/macosx-10-1/4/ https://arstechnica.com/gadgets/2001/10/macosx-10-1/4/ - was the "winner" in any usability study. This was removed in 2003-11 with the release of 10.3.

chefandy 3y ago
a) That scroll bar decision was made a long time ago b) Problems with specific dialog boxes or dynamically created features is evidence of problems with those dialog boxes or dynamically created features. As someone who's worked both as a developer and a designer, often the engineers win. c) Assuming that these topics are simple enough for off-the-cuff reckonin' to validate or invalidate them are exactly why the engineers shouldn't win when it comes to interface decisions.

Sakos 3y ago

> the designers at Apple who chose to hide those scroll bars probably thought more about scrolling more than every one of these commenters combined, and that was before they did formal usability studies. This blind appeal to authority shouldn't convince anyone.

chefandy 3y ago
If you've got an actual point to make instead of just poking at one little part of what I said, I'd respond.
- Sakos 3y ago
  That point is how you justify all your conjecture. Rethink what you wrote because all you're really doing is relying on anecdotes. I can literally say the opposite (I always use the scrollbar for scrolling and everybody I know does too) and we'd have made zero progress. You haven't actually made an argument worth arguing about. It's devoid of value and it's not a productive addition to the conversation. And don't bother wasting any more of my time. I don't care. You clearly have nothing to say.
  - chefandy 3y ago
    Yes. This is exactly the sort of 90 word response made by someone who doesn't have time to waste on arguments they don't care about.

CRConrad 3y ago

> Phone and tablet usage has outpaced regular computer usage for quite some time, and that informs users interaction style. "Informs"?!? Deforms! WTF does phone usage have to do with sabotaging the UI on a computer???

JohnFen 3y ago
Because designers and devs don't want to have to do multiple designs. Everyone's aiming for one design that works for everything. The problem is that such a thing is impossible.
- chefandy 3y ago
  Vollying back your flipness-- This sort of reductive assumption about software design is why companies that need people who aren't developers to actually use their products hire interface designers.
  - deprecative 3y ago
    Most employers I've had are all about getting engineers to be everything. Designers, infrastructure, QA, UI/UX, and even product/project managers.. everything.
    - chefandy 3y ago
      Sounds like you need new employers. The only people who want that are shortsighted managers who don't care that they know nothing about users' needs, and that's certainly not a universal disposition in the software business.
      - deprecative 3y ago
        This has been true of every employer I have ever had. It's true outside of "tech" as well but nearly all of my experience has been as a software engineer.

indymike 3y ago

> Phone and tablet usage has outpaced regular computer usage for quite some time This does not excuse making terrible user experiences elsewhere.

chefandy 3y ago
I think you're missing the point. Usability is not prescriptive. You implement what your users have the easiest time using. Touch interfaces change how people interact with interfaces. If you care about usability, you work with people on that and not against them.