I feel like I want to agree but I just can’t get behind this especially because the wording is so nebulous that it isn’t really talking about anything in particular. It would have helped if they...
I feel like I want to agree but I just can’t get behind this especially because the wording is so nebulous that it isn’t really talking about anything in particular. It would have helped if they named one concrete example of what they are arguing against.
Part of it might be because of the design choices for the page. Webpages that change as you scroll can all burn in a trash fire. I hate it so much. If I have to scroll back because something was changing where it looked like it was just a bunch of white space before, I’m already in the hate zone.
But yeah, I’m also tired of interfaces so slick that you can’t do anything with them. I’m tired of controls I have no idea what they do because we decided we only need arbitrary symbols for them. I’m tired of everything being hidden in menus nested further in a difficult to notice hamburger menu. I’m tired of forms that don’t just let me type in the info and force me to click a dozen times just to enter a date. I’m also sick of apps removing capabilities just to make them more streamlined. Especially when it comes without warning. I’m tired of desktop applications needing to be full screen because they are using mobile-first interfaces that take up way too much space.
Every single last one of these. Maximizing screen real estate is overrated: put all your dang buttons in WORDS and make them visible at all times. Making them go away or even hidden behind icons...
Every single last one of these.
Maximizing screen real estate is overrated: put all your dang buttons in WORDS and make them visible at all times. Making them go away or even hidden behind icons or a stupid kebab/burger should be the optional next step for AFTER I have already become familiar with the interface
Have I talked about how much I love Tildes' and Three Cheers' UI? Nothing flashes, nothing moves around and hide themselves. Just click the word that has the thing you want to do
This is one of the areas in which I believe the semi-lost-art of a proper menubar shines. There, a dev can put items for each program function in plain language unconstricted by UI layout,...
Maximizing screen real estate is overrated: put all your dang buttons in WORDS and make them visible at all times. Making them go away or even hidden behind icons or a stupid kebab/burger should be the optional next step for AFTER I have already become familiar with the interface
This is one of the areas in which I believe the semi-lost-art of a proper menubar shines. There, a dev can put items for each program function in plain language unconstricted by UI layout, organized into logical sections and submenus, using a pattern that’s easily learned and universal across desktop software. It’s so much better than trying to shove everything in a hamburger or burying slightly niche functions three modals deep, as is currently in fashion.
Of course, menubars have their limits too and can be made difficult to use, but very few programs have enough functionality to be at risk of this.
Tildes sidebar is called sidebar. I am not expected to long press or swipe or tap invisible sections of a screen in hopes of (1) finding an option I can understand and (2) not magically perform an...
Tildes sidebar is called sidebar.
I am not expected to long press or swipe or tap invisible sections of a screen in hopes of (1) finding an option I can understand and (2) not magically perform an action I don't know how to recover from. Give me 10 layer deep menus and incessant nag "are you sure" modal dialogues instead.
Edit: yes it's definitely a lost art, but I would strongly prefer even the ugliest and clunkiesf of menus over this hellscape
Found this on Lemmy — félt like something worth discussing here. I'm at least partially on the authors side, there's something frankly quite cold and unwelcoming of a lot of modern UI, and I miss...
Found this on Lemmy — félt like something worth discussing here.
I'm at least partially on the authors side, there's something frankly quite cold and unwelcoming of a lot of modern UI, and I miss some of the more skeumorphic elements of UI from my childhood. I accept however that some of this is a romanticism for the past, there are aspects of contemporary design that is frankly better (particularly in the accessibility front)
The author here is writing a rather fun / cute call to action, but to what end?
I think I agree, though the old skeuomorphic elements sometimes went overboard creating a cluttered mess that was difficult to navigate from that perspective. Having said that, I am not sure that...
there's something frankly quite cold and unwelcoming of a lot of modern UI
I think I agree, though the old skeuomorphic elements sometimes went overboard creating a cluttered mess that was difficult to navigate from that perspective.
Having said that, I am not sure that modern UIs are what they are as a countermovement to those skeuomorphic interfaces. Some of them certainly did start that way, but these days I think there are a variety of others reasons. Though a lot of it can be boiled to down to cost and efficiency.
A lot of UI these days, is done through a lot of boilerplate and other abstractions providing the developers with ready-made elements, icons, etc. Making a lot of these applications or websites not only very flat but also very very similar as they share a basic "design language".
This has been something that has been going on for some time now. Bootstrap for example is over a decade old now and one of the earlier examples I can think of. When it took off I noticed for the first time how large parts of the internet suddenly started to look very similar.
As I said earlier, there are many reasons for this. But the main factor, to me, seems to be cost and convenience as it is just a lot easier to combine ready-made building blocks compared to doing something yourself entirely.
Even ignoring all the frameworks, doing flat design simply is easier and therefore more cost-effective.
Anyway, back to the article. I love the visual presentation of it all, certainly the yarn world that unfolds in front of us. The example interface shown doesn't have me outright convinced, but that might just be because mindmapping in the way they do there feels chaotic to me. And maybe also because it more felt like an AI hype demo and not an interface demo.
It is a bit ironic that when I navigated away from the article the website itself lacks any features and is mostly a flat white pane. A white pane where I had trouble orienting myself as everything is just thrown on there without any separation. (Also, turns out the demo is indeed an AI tool)
VSTs for music are one of the rare places that is resisting the trend. It's largely the norm to have unique UIs that evoke physical hardware or futuristic things. There will be knobs and buttons,...
VSTs for music are one of the rare places that is resisting the trend. It's largely the norm to have unique UIs that evoke physicalhardware or futuristic things. There will be knobs and buttons, and they'll look interactive.
The exception to this norm is Ableton, which has a much more flat and uniform design scheme for all of its included "devices." Even the mixing faders don't look like physical ones, which is a fairly typical skeumorph for DAWs.
This is because making music is almost always a hands-on physical act. A human doesn’t go in and say “this sound is harsh, it needs a 127kHz cutoff filter,” they say “this sounds harsh, maybe a...
This is because making music is almost always a hands-on physical act. A human doesn’t go in and say “this sound is harsh, it needs a 127kHz cutoff filter,” they say “this sounds harsh, maybe a cutoff filter? I’ll adjust it until it sounds right and if that doesn’t sound right I’ll try the next thing”.
A lot of musicians I have known are gear heads, too, so the skeuomorphic design of VSTs help scratch that itch while making it understandable for musicians who are used to such interfaces.
I thought it was a good read. You might also enjoy this recent release series on Youtube regarding UI paradigms. https://www.youtube.com/playlist?list=PLsfH1Ahi4SzE-QmrsrtyZubGmi66iP45V I found it...
I thought it was a good read. You might also enjoy this recent release series on Youtube regarding UI paradigms.
I found it a compelling listen, covering a wide range of topics beyond just UI design. The first video provides a historical overview of how the modern computer interface evolved from the classic office environment metaphor, and how this paradigm has been refined over time without truly being reimagined. It also explores how our tools shape the way we think and create. While it presents an engaging critique of traditional UI structures, it did fall short in offering concrete alternatives, leaning more toward philosophy than practical solutions. Still, it’s a thought-provoking watch that challenges assumptions about how we interact with computers.
Their second brain app looks like a fun project to build. I struggle with certain interfaces like Figma because everything ends up in a giant ball, and the UI doesn't "tell" you what to do. But...
Their second brain app looks like a fun project to build. I struggle with certain interfaces like Figma because everything ends up in a giant ball, and the UI doesn't "tell" you what to do. But you "telling" the UI what to do sounds promising. Even a bad design could be usable with that kind of functionality.
Totally off-topic; I don't really care about the text contents of the article itself, but the imagery was breathtaking! This might be one of the first cases of AI generative art (it has to be,...
Totally off-topic; I don't really care about the text contents of the article itself, but the imagery was breathtaking! This might be one of the first cases of AI generative art (it has to be, right?) that I've found genuinely beautiful.
It doesn't have to be, there is decades of precedence where graphic designer and 3d artist make very similar images to illustrate their website. More often than not their portfolios. If you were...
It doesn't have to be, there is decades of precedence where graphic designer and 3d artist make very similar images to illustrate their website. More often than not their portfolios. If you were around in the early 2000s you might remember those incredibly elaborate and intricate flash websites.
To be clear, it is very possible ai is involved. But it doesn't have to be at all. In fact that you might think so might be a result of the flat featureless designs we have gotten so used to.
Yeah, I understand it doesn't have to be AI generated, per se. I have seen wool/yarn art before that was somewhat similar (E.g. Simon Brown 'The Gentleman Felter', "crochet art" on Etsy, etc.)....
Yeah, I understand it doesn't have to be AI generated, per se. I have seen wool/yarn art before that was somewhat similar (E.g. Simon Brown 'The Gentleman Felter', "crochet art" on Etsy, etc.). But considering how many of them there are in this blog post, and how detailed some of them are (E.g. 1, 2, 3, 4), none of which are credited to any particular artist, or show up in reverse image search, it kinda does have to be AI generated, in this particular case... which is more what I meant. ;)
I think your instinct is correct. There are a couple of them that show "text" and it's all incoherent (but letter-like) scribbles. That was the tell for me. There are yarn art LoRAs and custom...
I think your instinct is correct. There are a couple of them that show "text" and it's all incoherent (but letter-like) scribbles. That was the tell for me.
There are yarn art LoRAs and custom models that produce stuff like this. With some decent prompting ("cute, fuzzy, tilt-shift, warm desaturated tones", that sort of thing) you can get similar results.
I mean, yes in this instance it is likely AI generated. I am just saying that in a more general sense it doesn't have to be. As I have seen artists create similar detailed websites with no other...
I mean, yes in this instance it is likely AI generated. I am just saying that in a more general sense it doesn't have to be. As I have seen artists create similar detailed websites with no other credit other than you being on their website.
Including similar art styles, again, often done by 3d artists.
Again, I am not denying this might be AI generated. But I am pushing back to the notion that because it is so intricate and detailed for just one page it has to be AI generated. Because that doesn't feel right to me given the amazing pieces of digital art I have seen over the decades.
The yarn art/feltwork definitely wasn't physically crocheted/felted -- as someone who sometimes shops for crochet patterns, GenAI has meant having to learn to keep an eye out for the signs that...
The yarn art/feltwork definitely wasn't physically crocheted/felted -- as someone who sometimes shops for crochet patterns, GenAI has meant having to learn to keep an eye out for the signs that pictures of yarnwork aren't representing a real piece someone could make by hand. I think these pictures are almost certainly AI-generated, but from my background/perspective there's not much difference if it is a human artist making these digitally, since that would cause the same problems in the places AI-generated yarnwork is most pervasive (i.e., to sell fake patterns on Etsy). I'm not familiar with any digital artist who works in this style, though, so I'd love for you to link me to someone if you know of any human artists who specialize in digital art that looks like yarnwork.
Look, I already acknowledged that in this case it likely is AI generated. Just that in a more general sense, over the years, I also have seen ellaborate websites with equally elaborate digital art...
Look, I already acknowledged that in this case it likely is AI generated. Just that in a more general sense, over the years, I also have seen ellaborate websites with equally elaborate digital art just created because someone wanted to create it. I mean, you are active on the tildes minecraft server and have seen the massive projects people undertake just for the sake of it. That is my entire point, people have been creating this sort of stuff as well so it isn't a given that it must be AI generated from that perspective. Again to be very clear, in this instance there are AI tells, but that is not the point I am trying to make.
I would love to link you to some 3d artists over this. But it isn't as if I am keeping track of every showcase someone posted in one of the many 3d rendering communities or other digital artist spaces. I just know that I have seen some pretty damn cool 3d renderings of yarn over the years. But also that isn't my point as, as far as I am concerned, the subject matter could have been anything else elaborate.
There are many people doing cool things with AI. I think we passively get exposed to the slop so that's the first thing we think about, but if you dig enough you'll find cool stuff. A couple...
There are many people doing cool things with AI. I think we passively get exposed to the slop so that's the first thing we think about, but if you dig enough you'll find cool stuff. A couple examples:
I think it's related to the toupee fallacy. You can spot bad AI generated content easily as it's bad. Good AI generated content doesn't have the tell tale signs of poorly generated content.
I think it's related to the toupee fallacy.
You can spot bad AI generated content easily as it's bad.
Good AI generated content doesn't have the tell tale signs of poorly generated content.
I was not sure if the art was AI generated or not, but I had a sneaking suspicion that it had to be as well. I appreciate the artistic elements, fabric work has a certain tactility to it which I...
I was not sure if the art was AI generated or not, but I had a sneaking suspicion that it had to be as well. I appreciate the artistic elements, fabric work has a certain tactility to it which I really appreciate (in my free time, I do enjoy crocheting a lot).
I scrolled the whole post thinking "please let it not be AI", sad reality we're living. Anyway, I'm not very good detecting this but I think I found proof that it's AI. Look at the "guesture while...
I scrolled the whole post thinking "please let it not be AI", sad reality we're living. Anyway, I'm not very good detecting this but I think I found proof that it's AI. Look at the "guesture while talking" girl, she has 6 fingers. Congrats to the artists AI stole to learn how to create this, beautiful post nonetheless :/
edit: not judging Amelia BTW, big fan of her work.
I really think a lot of this has to do with graphics designers not being valued enough. As others have mentioned, we have these packages set up for developers to use for creating web pages, so we...
I really think a lot of this has to do with graphics designers not being valued enough. As others have mentioned, we have these packages set up for developers to use for creating web pages, so we use those and set everything up in ways that make sense to us. Many of them have a lot of accessibility set up so that nobody feels the need to think about it, and the end result looks like everything else, so it must be fine.
Designers don’t just make software pretty. They do a lot of research before they make any decisions and they conduct user interviews to make sure users don’t struggle to use the software. I have yet to work at a company willing to assign a designer to internal software, so employers end up wasting a lot of time struggling with the software or refuse to use it altogether.
Interesting read! I had a similar thought to Akir, I felt like I both understood and didn't quite understand what the author was getting at. I think this came from the lack of concrete examples....
Interesting read! I had a similar thought to Akir, I felt like I both understood and didn't quite understand what the author was getting at. I think this came from the lack of concrete examples. They gave one example with the organizing tool, though details were sparse. I wanted to learn more!! It didn't help that some text was not visible hue to being cropped by the images.
The thoughts of the author are somewhat related to similar thoughts I have on a continuing interest in input perephrals for computers. Everything is a mouse and keyboard/buttons (or worse, screen taps). I have to click or press buttons repeatedly to adjust volume, where did our volume control knobs go? Those are so nice! You can even have one act as a mouse scroll wheel! Or use a SpaceMouse or trackpoint/nub as a pointer device, that's rare outside laptops. Or those IMU/accelerometer devices you wave around to move the pointer during presentations, one could have it for everyday use even. Let's bring back trackballs! Even for keyboards, there's many options out there, split, 60%, ergonomic, etc. The physical layout is so wack by being staggered, it's not great. Even button layouts like qwerty, other ones feel so much nicer to type with! There's Frogpad and others for one-handed use, so some other ways to type. There's probably good reason many of these aren't mainstream, but there's no reason to restrict ourselves to one mode of input. But still, most are meant for you to type one letter at a time. It doesn't necessarily have to be that way, one could chord their inputs! I'm sure there's mere. What if we were to start from scratch, could thece be a better way to "type"? I don't even know what that would look like.
I definitely agree with your thoughts as well as Akir, frankly part of why I posted this was my struggles with understanding and simultaneously not understanding (I was wondering if it was just my...
I definitely agree with your thoughts as well as Akir, frankly part of why I posted this was my struggles with understanding and simultaneously not understanding (I was wondering if it was just my brain being stuck in French today, and I am somewhat relieved that is not the case lol)
And input methods are really important! I guess this is more so tangentially related, but I have been meaning to try more input methods — specifically ones usually relegated to accessibility, like eye tracking. Part of it is perhaps more so my desire to learn how to use a computer without using all of my faculties, just ot have a better understanding of how users interact with their computers. But I also hope to find use for it where it excels beyond the traditional mouse / keyboard input. As it stands now, I already do have a somewhat multi-modal input system, I handwrite some things and convert them to text later, or my varried usage of trackpoint / trackpad / mouse depending on what application I am using.
What if we were to start from scratch, could thece be a better way to "type"? I don't even know what that would look like.
I think there likely would be a better way to type. As it stands now, keyboards are not the most ergonomic of input methods, and even optimized layouts falter in multilingual usecases. I feel like a truly optimized setup would be some type of chorded input, but even then that might not be the only one as sometimes it might just be easier to use voice recognition or other methods.
I've seen this article making the rounds but I had never actually read it. I agree with some minor things but disagree with the general idea. The minor things I agree with: human-computer...
I've seen this article making the rounds but I had never actually read it. I agree with some minor things but disagree with the general idea.
The minor things I agree with: human-computer interfaces will evolve to better integrate our senses (animations, sounds) and the physical space, both as pure tactile feeling (hardware, haptics) and certain digital spaces trying to mimic what we feel and perceive in our real-world spaces.
The general idea, or what I thought it was based on the prototype shown at the end: de-digitalizing spaces and tools is generally a good path. I mostly disagree with this because the digital space is not bounded by our senses and precisely that's its good thing; we don't need to feel or touch, we can just think. And when dealing with information that is highly valuable. We can get information streamlined and distilled to its most pure form, and have a representation of it appropriate for the specific moment we might be in.
I honestly don’t think animations help interfaces in any way. Instant feedback is way better for me. If I tell a computer to do something, I want it to do it now, not after it’s done a...
I honestly don’t think animations help interfaces in any way. Instant feedback is way better for me. If I tell a computer to do something, I want it to do it now, not after it’s done a loop-de-loop to congratulate me for telling it what to do. That kind of thing in particular I find really patronizing.
There are some animations that are acceptable. When I am moving things with the mouse, for instance, but that’s part of that instant feedback thing. But the one time I think animation is actually helpful is when it’s trying to teach me something. By all means, catch my eye and bring me to the thing I need to know about. It saves me from having to search for it. But after I’ve learned it, I don’t want it to bug me.
This immediately brought to mind two ideas for me. The first was the Pattie Maes TED Talk from 2009 (barely two years after the first smartphone!) with an incredibly clunky prototype...
This immediately brought to mind two ideas for me.
The first was the Pattie Maes TED Talk from 2009 (barely two years after the first smartphone!) with an incredibly clunky prototype camera+projector device that was supposed to be a more intuitive gesture-based way of interacting with the digital world.
The second is the current state of VR, which other than refresh rate and number/density of pixels, really doesn’t seem to have progressed much further than this 2009 tech demo with regards to interface. My experience is likely coloured by Meta’s Quest 3 being basically my only VR UI experience which I’ve heard is even less imaginative than other VR, although not my much.
I wonder if better pass through AR and a complete projector wall might push for better ways and types of interacting with data? Because at the moment it seems like there are too many people who get motion sick from VR (even with high frame rates and low latency) for it to ever be truly ubiquitous.
I had such high hopes for VR to be used for office work. I'd love to replace my screens with a virtual office where my physical screen size is no limitation.
I had such high hopes for VR to be used for office work.
I'd love to replace my screens with a virtual office where my physical screen size is no limitation.
I’ve never tried using VR for productivity because the resolution for text just isn’t there yet (or is too expensive for a consumer product) but even with two external monitors plus my laptop at...
I’ve never tried using VR for productivity because the resolution for text just isn’t there yet (or is too expensive for a consumer product) but even with two external monitors plus my laptop at work, I still run into limits when I’m comparing details across multiple documents and databases, and dream of the day that VR/AR can give me infinite screen space!
I feel like I want to agree but I just can’t get behind this especially because the wording is so nebulous that it isn’t really talking about anything in particular. It would have helped if they named one concrete example of what they are arguing against.
Part of it might be because of the design choices for the page. Webpages that change as you scroll can all burn in a trash fire. I hate it so much. If I have to scroll back because something was changing where it looked like it was just a bunch of white space before, I’m already in the hate zone.
But yeah, I’m also tired of interfaces so slick that you can’t do anything with them. I’m tired of controls I have no idea what they do because we decided we only need arbitrary symbols for them. I’m tired of everything being hidden in menus nested further in a difficult to notice hamburger menu. I’m tired of forms that don’t just let me type in the info and force me to click a dozen times just to enter a date. I’m also sick of apps removing capabilities just to make them more streamlined. Especially when it comes without warning. I’m tired of desktop applications needing to be full screen because they are using mobile-first interfaces that take up way too much space.
Every single last one of these.
Maximizing screen real estate is overrated: put all your dang buttons in WORDS and make them visible at all times. Making them go away or even hidden behind icons or a stupid kebab/burger should be the optional next step for AFTER I have already become familiar with the interface
Have I talked about how much I love Tildes' and Three Cheers' UI? Nothing flashes, nothing moves around and hide themselves. Just click the word that has the thing you want to do
This is one of the areas in which I believe the semi-lost-art of a proper menubar shines. There, a dev can put items for each program function in plain language unconstricted by UI layout, organized into logical sections and submenus, using a pattern that’s easily learned and universal across desktop software. It’s so much better than trying to shove everything in a hamburger or burying slightly niche functions three modals deep, as is currently in fashion.
Of course, menubars have their limits too and can be made difficult to use, but very few programs have enough functionality to be at risk of this.
Tildes sidebar is called sidebar.
I am not expected to long press or swipe or tap invisible sections of a screen in hopes of (1) finding an option I can understand and (2) not magically perform an action I don't know how to recover from. Give me 10 layer deep menus and incessant nag "are you sure" modal dialogues instead.
Edit: yes it's definitely a lost art, but I would strongly prefer even the ugliest and clunkiesf of menus over this hellscape
Found this on Lemmy — félt like something worth discussing here.
I'm at least partially on the authors side, there's something frankly quite cold and unwelcoming of a lot of modern UI, and I miss some of the more skeumorphic elements of UI from my childhood. I accept however that some of this is a romanticism for the past, there are aspects of contemporary design that is frankly better (particularly in the accessibility front)
The author here is writing a rather fun / cute call to action, but to what end?
I think I agree, though the old skeuomorphic elements sometimes went overboard creating a cluttered mess that was difficult to navigate from that perspective.
Having said that, I am not sure that modern UIs are what they are as a countermovement to those skeuomorphic interfaces. Some of them certainly did start that way, but these days I think there are a variety of others reasons. Though a lot of it can be boiled to down to cost and efficiency.
A lot of UI these days, is done through a lot of boilerplate and other abstractions providing the developers with ready-made elements, icons, etc. Making a lot of these applications or websites not only very flat but also very very similar as they share a basic "design language".
This has been something that has been going on for some time now. Bootstrap for example is over a decade old now and one of the earlier examples I can think of. When it took off I noticed for the first time how large parts of the internet suddenly started to look very similar.
As I said earlier, there are many reasons for this. But the main factor, to me, seems to be cost and convenience as it is just a lot easier to combine ready-made building blocks compared to doing something yourself entirely.
Even ignoring all the frameworks, doing flat design simply is easier and therefore more cost-effective.
Anyway, back to the article. I love the visual presentation of it all, certainly the yarn world that unfolds in front of us. The example interface shown doesn't have me outright convinced, but that might just be because mindmapping in the way they do there feels chaotic to me. And maybe also because it more felt like an AI hype demo and not an interface demo.
It is a bit ironic that when I navigated away from the article the website itself lacks any features and is mostly a flat white pane. A white pane where I had trouble orienting myself as everything is just thrown on there without any separation. (Also, turns out the demo is indeed an AI tool)
VSTs for music are one of the rare places that is resisting the trend. It's largely the norm to have unique UIs that evoke physical hardware or futuristic things. There will be knobs and buttons, and they'll look interactive.
The exception to this norm is Ableton, which has a much more flat and uniform design scheme for all of its included "devices." Even the mixing faders don't look like physical ones, which is a fairly typical skeumorph for DAWs.
This is because making music is almost always a hands-on physical act. A human doesn’t go in and say “this sound is harsh, it needs a 127kHz cutoff filter,” they say “this sounds harsh, maybe a cutoff filter? I’ll adjust it until it sounds right and if that doesn’t sound right I’ll try the next thing”.
A lot of musicians I have known are gear heads, too, so the skeuomorphic design of VSTs help scratch that itch while making it understandable for musicians who are used to such interfaces.
I thought it was a good read. You might also enjoy this recent release series on Youtube regarding UI paradigms.
https://www.youtube.com/playlist?list=PLsfH1Ahi4SzE-QmrsrtyZubGmi66iP45V
I found it a compelling listen, covering a wide range of topics beyond just UI design. The first video provides a historical overview of how the modern computer interface evolved from the classic office environment metaphor, and how this paradigm has been refined over time without truly being reimagined. It also explores how our tools shape the way we think and create. While it presents an engaging critique of traditional UI structures, it did fall short in offering concrete alternatives, leaning more toward philosophy than practical solutions. Still, it’s a thought-provoking watch that challenges assumptions about how we interact with computers.
Their second brain app looks like a fun project to build. I struggle with certain interfaces like Figma because everything ends up in a giant ball, and the UI doesn't "tell" you what to do. But you "telling" the UI what to do sounds promising. Even a bad design could be usable with that kind of functionality.
Totally off-topic; I don't really care about the text contents of the article itself, but the imagery was breathtaking! This might be one of the first cases of AI generative art (it has to be, right?) that I've found genuinely beautiful.
It doesn't have to be, there is decades of precedence where graphic designer and 3d artist make very similar images to illustrate their website. More often than not their portfolios. If you were around in the early 2000s you might remember those incredibly elaborate and intricate flash websites.
To be clear, it is very possible ai is involved. But it doesn't have to be at all. In fact that you might think so might be a result of the flat featureless designs we have gotten so used to.
Yeah, I understand it doesn't have to be AI generated, per se. I have seen wool/yarn art before that was somewhat similar (E.g. Simon Brown 'The Gentleman Felter', "crochet art" on Etsy, etc.). But considering how many of them there are in this blog post, and how detailed some of them are (E.g. 1, 2, 3, 4), none of which are credited to any particular artist, or show up in reverse image search, it kinda does have to be AI generated, in this particular case... which is more what I meant. ;)
I think your instinct is correct. There are a couple of them that show "text" and it's all incoherent (but letter-like) scribbles. That was the tell for me.
There are yarn art LoRAs and custom models that produce stuff like this. With some decent prompting ("cute, fuzzy, tilt-shift, warm desaturated tones", that sort of thing) you can get similar results.
I mean, yes in this instance it is likely AI generated. I am just saying that in a more general sense it doesn't have to be. As I have seen artists create similar detailed websites with no other credit other than you being on their website.
Including similar art styles, again, often done by 3d artists.
Again, I am not denying this might be AI generated. But I am pushing back to the notion that because it is so intricate and detailed for just one page it has to be AI generated. Because that doesn't feel right to me given the amazing pieces of digital art I have seen over the decades.
The yarn art/feltwork definitely wasn't physically crocheted/felted -- as someone who sometimes shops for crochet patterns, GenAI has meant having to learn to keep an eye out for the signs that pictures of yarnwork aren't representing a real piece someone could make by hand. I think these pictures are almost certainly AI-generated, but from my background/perspective there's not much difference if it is a human artist making these digitally, since that would cause the same problems in the places AI-generated yarnwork is most pervasive (i.e., to sell fake patterns on Etsy). I'm not familiar with any digital artist who works in this style, though, so I'd love for you to link me to someone if you know of any human artists who specialize in digital art that looks like yarnwork.
Look, I already acknowledged that in this case it likely is AI generated. Just that in a more general sense, over the years, I also have seen ellaborate websites with equally elaborate digital art just created because someone wanted to create it. I mean, you are active on the tildes minecraft server and have seen the massive projects people undertake just for the sake of it. That is my entire point, people have been creating this sort of stuff as well so it isn't a given that it must be AI generated from that perspective. Again to be very clear, in this instance there are AI tells, but that is not the point I am trying to make.
I would love to link you to some 3d artists over this. But it isn't as if I am keeping track of every showcase someone posted in one of the many 3d rendering communities or other digital artist spaces. I just know that I have seen some pretty damn cool 3d renderings of yarn over the years. But also that isn't my point as, as far as I am concerned, the subject matter could have been anything else elaborate.
There are many people doing cool things with AI. I think we passively get exposed to the slop so that's the first thing we think about, but if you dig enough you'll find cool stuff. A couple examples:
And also: Entangled others, an artistic duo that started using deep learning way before LLMs.
I think it's related to the toupee fallacy.
You can spot bad AI generated content easily as it's bad.
Good AI generated content doesn't have the tell tale signs of poorly generated content.
I was not sure if the art was AI generated or not, but I had a sneaking suspicion that it had to be as well. I appreciate the artistic elements, fabric work has a certain tactility to it which I really appreciate (in my free time, I do enjoy crocheting a lot).
I scrolled the whole post thinking "please let it not be AI", sad reality we're living. Anyway, I'm not very good detecting this but I think I found proof that it's AI. Look at the "guesture while talking" girl, she has 6 fingers. Congrats to the artists AI stole to learn how to create this, beautiful post nonetheless :/
edit: not judging Amelia BTW, big fan of her work.
I really think a lot of this has to do with graphics designers not being valued enough. As others have mentioned, we have these packages set up for developers to use for creating web pages, so we use those and set everything up in ways that make sense to us. Many of them have a lot of accessibility set up so that nobody feels the need to think about it, and the end result looks like everything else, so it must be fine.
Designers don’t just make software pretty. They do a lot of research before they make any decisions and they conduct user interviews to make sure users don’t struggle to use the software. I have yet to work at a company willing to assign a designer to internal software, so employers end up wasting a lot of time struggling with the software or refuse to use it altogether.
Interesting read! I had a similar thought to Akir, I felt like I both understood and didn't quite understand what the author was getting at. I think this came from the lack of concrete examples. They gave one example with the organizing tool, though details were sparse. I wanted to learn more!! It didn't help that some text was not visible hue to being cropped by the images.
The thoughts of the author are somewhat related to similar thoughts I have on a continuing interest in input perephrals for computers. Everything is a mouse and keyboard/buttons (or worse, screen taps). I have to click or press buttons repeatedly to adjust volume, where did our volume control knobs go? Those are so nice! You can even have one act as a mouse scroll wheel! Or use a SpaceMouse or trackpoint/nub as a pointer device, that's rare outside laptops. Or those IMU/accelerometer devices you wave around to move the pointer during presentations, one could have it for everyday use even. Let's bring back trackballs! Even for keyboards, there's many options out there, split, 60%, ergonomic, etc. The physical layout is so wack by being staggered, it's not great. Even button layouts like qwerty, other ones feel so much nicer to type with! There's Frogpad and others for one-handed use, so some other ways to type. There's probably good reason many of these aren't mainstream, but there's no reason to restrict ourselves to one mode of input. But still, most are meant for you to type one letter at a time. It doesn't necessarily have to be that way, one could chord their inputs! I'm sure there's mere. What if we were to start from scratch, could thece be a better way to "type"? I don't even know what that would look like.
I definitely agree with your thoughts as well as Akir, frankly part of why I posted this was my struggles with understanding and simultaneously not understanding (I was wondering if it was just my brain being stuck in French today, and I am somewhat relieved that is not the case lol)
And input methods are really important! I guess this is more so tangentially related, but I have been meaning to try more input methods — specifically ones usually relegated to accessibility, like eye tracking. Part of it is perhaps more so my desire to learn how to use a computer without using all of my faculties, just ot have a better understanding of how users interact with their computers. But I also hope to find use for it where it excels beyond the traditional mouse / keyboard input. As it stands now, I already do have a somewhat multi-modal input system, I handwrite some things and convert them to text later, or my varried usage of trackpoint / trackpad / mouse depending on what application I am using.
I think there likely would be a better way to type. As it stands now, keyboards are not the most ergonomic of input methods, and even optimized layouts falter in multilingual usecases. I feel like a truly optimized setup would be some type of chorded input, but even then that might not be the only one as sometimes it might just be easier to use voice recognition or other methods.
I've seen this article making the rounds but I had never actually read it. I agree with some minor things but disagree with the general idea.
The minor things I agree with: human-computer interfaces will evolve to better integrate our senses (animations, sounds) and the physical space, both as pure tactile feeling (hardware, haptics) and certain digital spaces trying to mimic what we feel and perceive in our real-world spaces.
The general idea, or what I thought it was based on the prototype shown at the end: de-digitalizing spaces and tools is generally a good path. I mostly disagree with this because the digital space is not bounded by our senses and precisely that's its good thing; we don't need to feel or touch, we can just think. And when dealing with information that is highly valuable. We can get information streamlined and distilled to its most pure form, and have a representation of it appropriate for the specific moment we might be in.
I honestly don’t think animations help interfaces in any way. Instant feedback is way better for me. If I tell a computer to do something, I want it to do it now, not after it’s done a loop-de-loop to congratulate me for telling it what to do. That kind of thing in particular I find really patronizing.
There are some animations that are acceptable. When I am moving things with the mouse, for instance, but that’s part of that instant feedback thing. But the one time I think animation is actually helpful is when it’s trying to teach me something. By all means, catch my eye and bring me to the thing I need to know about. It saves me from having to search for it. But after I’ve learned it, I don’t want it to bug me.
This immediately brought to mind two ideas for me.
The first was the Pattie Maes TED Talk from 2009 (barely two years after the first smartphone!) with an incredibly clunky prototype camera+projector device that was supposed to be a more intuitive gesture-based way of interacting with the digital world.
The second is the current state of VR, which other than refresh rate and number/density of pixels, really doesn’t seem to have progressed much further than this 2009 tech demo with regards to interface. My experience is likely coloured by Meta’s Quest 3 being basically my only VR UI experience which I’ve heard is even less imaginative than other VR, although not my much.
I wonder if better pass through AR and a complete projector wall might push for better ways and types of interacting with data? Because at the moment it seems like there are too many people who get motion sick from VR (even with high frame rates and low latency) for it to ever be truly ubiquitous.
I had such high hopes for VR to be used for office work.
I'd love to replace my screens with a virtual office where my physical screen size is no limitation.
I’ve never tried using VR for productivity because the resolution for text just isn’t there yet (or is too expensive for a consumer product) but even with two external monitors plus my laptop at work, I still run into limits when I’m comparing details across multiple documents and databases, and dream of the day that VR/AR can give me infinite screen space!