this post was submitted on 08 Jul 2023
1953 points (94.8% liked)
Fediverse
27910 readers
1 users here now
A community to talk about the Fediverse and all it's related services using ActivityPub (Mastodon, Lemmy, KBin, etc).
If you wanted to get help with moderating your own community then head over to [email protected]!
Rules
- Posts must be on topic.
- Be respectful of others.
- Cite the sources used for graphs and other statistics.
- Follow the general Lemmy.world rules.
Learn more at these websites: Join The Fediverse Wiki, Fediverse.info, Wikipedia Page, The Federation Info (Stats), FediDB (Stats), Sub Rehab (Reddit Migration), Search Lemmy
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
This is not a problem with people, but with UX design.
We don't need a corporation to have usable interfaces. Right now, if you visit join-lemmy.org, the main focus is for people wanting to host an instance, which is only a small part of the advanced user base. The common user won't care about the fact Lemmy is made with rust or that there's a docker image.
I don't think it's only an issue with Lemmy, lots of open-source projects lack user-friendliness and onboarding.
I agree so much. I feel like your typical user does not need to know 1% of all that. Hell, I don't even think that they need to know much about the whole instances shebang. Scroll through a feed, see a bunch of users from various federated instances say things, the end. The fediverse aspect is something that could be relegated to a help section or something along these lines. Complexity scares people away.
I seriously wish that Lemmy and the Fediverse in general would boom in popularity because this kind of "free" social media is what people deserve, but I just can't see it happening with how things are now.
People need to know what an instance is. They need to know what defederating means too. They also need to know why the Fediverse is better than centralized alternatives, because otherwise they won't care.
Can I ask you to educate me on why people need to know? I'm struggling to see your point. Surely people will benefit in some way from knowing how the Fediverse works under the hood, but not everyone needs to know, and as you can see this heavy emphasis on the Fediverse's inner workings is, instead, turning people away.
People are just very used to having things magically work, and I think that it's very natural for them to not want to deal with things that they need to make work. Many people never knew how Twitter or Reddit works under the hood, many never cared, but in the Fediverse it's suddenly brought up as a necessary thing. And I just can't see how that's necessary thing to use such a tool. All I'm saying is that UX-wise, many apps in the Fediverse could feel more seamless.
I hear you on the Fediverse being better in many many cases, but I also feel like many of the Fediverse's features are all but marketable to the average person. Add in the fact that people feel like there is a learning curve, and what you get is that people lose interest very fast.
In order to be able to choose an instance or change it in the future, a user needs to know what it is first. Maybe we could choose it for them randomly (would that be ethical?), but they still need to know where their account is hosted. If their instance gets defederaded by other instances, then they also need to know what defederation is and how it affects them. Those are just basics needed to fully use the Fediverse and the freedoms it gives you. I'm not saying that people should know exactly what ActivityPub protocol is and how it works. But they need to know the basics about how federation works. This is nothing complicated, so I don't think I'm asking for much.
In centralized platforms users don't have freedom, so they don't have to worry about any of this. If they aren't willing to learn a few basic things in order to have freedom, then there is nothing we can do for them. We can and should try to educate them, but many people don't see a problem with being abused by corporations. This is not a new problem in software. It has existed for at least 30 years since the Free Software movement was created.
Lemmy has many UI and UX problems and they certainly need to be fixed. But it's possible that decentralized platforms will always be more complicated to use than centralized ones. If people aren't willing to learn how to use it, that's a problem with them, not with this technology.
Okay, you bring very good points, especially on the "changing instance" one. I think I was seeing things from an ideal point of view, where instances just work and there are no reasons to defederate from other instances (but even though I've not been on the Fediverse for long, I've already seen I think two "big" such cases :( ).
I would say that choosing it for them is not the way, and assigning it randomly isn't either as there are definitely problems associated with that as well. The best thing that comes to mind is to maybe have some "special" instance (or just an application, kinda like what Mastodon's Android app does - at least with the new update) whose purpose is to guide users through sign up and choosing an instance. I think this would kill two birds with one stone. Guiding users through instance-selection, maybe briefly explaining what an instance is and eventually pointing to more user-friendly docs, could already be much more manageable for everybody, and could feel like a more seamless experience, similarly to traditional social media.
At the end of the day I feel even more like the Fediverse is almost inherently harder than centralized services, maybe it will take time before people settle with the idea of using something like this. People eventually got how to use the now traditional kind of technology, but I don't know if it's because enough time has passed or because it became a necessity (socially speaking).
That's a very good idea. The website https://join-lemmy.org could have something like that or we could make a new website for it.
People are already adjusting to the decentralized services, but it's possible that they will never become popular enough for most people to have to learn how to use them.
Good points. I do think the default community filter should be All instead of Local though.
I agree that it should be the default option. Thankfully you can change it from your account's settings, which is very nice.
You are probably right about that. There are other UI improvements that we could and should do too, but Lemmy will probably always be more difficult to use than the centralized alternatives.
Yes, Lemmy's UI is very bad. It would be pretty easy to improve it, if only developers understood this. But I think the part that new users complain about the most is federation. At least I've seen many posts and comments saying that it's too confusing.
I don't think it's necessarily the job of the developers, the main issue IMO is that there's not enough involvement from other specialists such as designers in open-source communities.
Sometimes I try to help, but unfortunately not everyone is willing to listen. I've noticed there are multiple reasons why UI might be bad in a Free Software project:
I believe in Lemmy's case it's mostly the 2nd point.
IDK about Lemmy devs, but point 2 is so, so common. Making a point about UX or accessibility in 99% of FLOSS project discussion spaces is incredibly stressful; you can have user research, industry best practice, and years of experience on your side, but you're inevitably met with dismissal and argument. Devs often treat designers as though they're a bunch of artsy crystal-healing crusties, despite the fact that good UX people base their work on actual research and theory grounded in human behavior and psychology. (Calling use of basic design principles "eye candy" for example) Of course, if a dev makes a decision on technical grounds, it must be treated as scripture as far as any remaining designers on the project are concerned. It's no wonder so many FLOSS projects have abominable UX.
True. I feel like usually anything I say about UI is interpreted as just my opinion, which in developer's minds is just as valid as anyone else's opinion. It maybe kinda makes sense, since those developers don't really know me, but there is nothing I can say to change their minds. So they remain stuck with a bad, inefficient design for no good reason.
GNOME desktop environment is a good example that having a good UI is possible.
In Lemmy's case users are forced to take matters into their own hands: [email protected]
Exactly. I once made a point about excessive indicators of visual / information hierarchy increasing cognitive load without additional benefit on a subreddit and got downvoted to oblivion. That was not my opinion; that's what industry research indicates!
Got to say though, I think GNOME is pretty, but a usability nightmare.
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
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.
What is an alternative to a hamburger menu for a mobile layout?
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.
Hmm, that's interesting, thank you ☺️ I bet you're expensive
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.
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.
That is true, good point.
Here is an alternative Piped link(s): https://piped.video/GkxAp2Gh7-E
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source, check me out at GitHub.
Also, I always have to have a chuckle when I see what clients produce on their own. Like, they'll cheap out and produce absolute garbage that they're somehow happy with, but the minute you, a design professional, submit a concept or mockup that is far beyond their ability to produce, the client is absolutely full of feedback: "Make the logo 5x bigger!" "Let's use this barely visible shade of blue!"
Hi there! Looks like you linked to a Lemmy community using an URL instead of its name, which doesn't work well for people on different instances. Try fixing it like this: [email protected]