The mastodon and lemmy content I’m seeing feels like 90% of it comes from people who are:

  • ~30 years old or older

  • tech enthusiasts/workers

  • linux users

There’s nothing wrong with that particular demographic or anything, but it doesn’t feel like a win to me if the entire fediverse is just one big monoculture.

I wonder what it is that is keeping more diverse users away? Is picking a server/federation too complicated? Or is it that they don’t see any content that they like?

Thoughts?

  • Freesoftwareenjoyer@lemmy.worldEnglish
    1·
    1 year ago

    It’s not based on the same 30 years old design that all popular operating systems are. So it might take some time to learn how to use it. Is that what you meant or do you think that it’s badly designed?

    Here is an interesting video about this topic: https://youtu.be/GkxAp2Gh7-E

    • NathanUp@lemmy.mlEnglish
      1·
      1 year ago

      Yea, I think they make honest to god bad design decisions that hurt usability. A common thread is hiding features / reducing discoverability while making no attempt at progressive disclosure, requiring memorization to complete certain tasks with the interface. This isn’t only bad UX, it’s an accessibility issue for users with attention and/or memory deficits. Creating a new paradigm is one thing, but with that comes the task of building affordances that help users with the transition, like skeumorphism did back in the day (…and to an extent, skeumorphism should have never been abandoned in the way it was…), something that the GNOME project simply does not do. They also ignore common accessibility recommendations, for example, by using icons without text in their applications, and utilizing mystery-meat navigation methods like hamburger menus, and ignoring long-established patterns for even very basic tasks, like allowing titlebars to become cluttered with interface elements leading to confusion when the user wants to move the window and widgets are in the way. I don’t think it’s bad at all that the GNOME project is trying to build their own paradigm, but they do so without consideration for the most fundamental usability guidelines.

        • NathanUp@lemmy.mlEnglish
          2·
          1 year ago

          There are many options; it just takes a bit of creativity, and it’s better to involve the designer in the early planning stages to nail down what needs to be immediately accessible, and what can be revealed via progressive disclosure. I did a website for a beverage company a few years back that used a bottom-aligned series of pills for the navigation that scrolled horizontally - a shadow appeared on the side to indicate that it can be scrolled. (We used JS to add shadow to any side which had overflowing elements. See below for a very rough little wireframe.) Twitter, like many sites and apps, also used to have no hamburger menu if you can remember.

          A mockup illustrating the navigation system described above

      • Freesoftwareenjoyer@lemmy.worldEnglish
        1·
        1 year ago

        I see. I haven’t thought of that before, but when I look at Windows 11 file explorer, the one in GNOME seems way easier for me to understand despite it having the flaws that you mentioned. Maybe I just got used to it.

        • NathanUp@lemmy.mlEnglish
          3·
          1 year ago

          Yea, I get you. That’s why usability research is so important - these interfaces may be simple for you and I, but small issues / contributors to increased cognitive load make a big difference for the large majority of people who do not live and breathe computer interfaces. I get that not everyone can afford to conduct their own studies, but that’s why small orgs (and even modestly funded FLOSS projects) need to be making use of the work of players like the Nielsen Norman group who release their reports very affordably, even if they have an expert on hand. I always try to remember that every interface is contrived, so there is no such thing as an “intuitive” computer or machine interface; interface usability lives or dies with discoverability.