/ui/ - UI/UX Lab

>>2416
The easing curves aren't actually the issue, I think it just needs a better bezier definition to stop that jittery look.

>>2398
>if you aren't looking at the world, are you even designing?

I had a client last month who thought this was some deep philosophical breakthrough. They wanted an entire brand identity based on "found textures" from their local subway station. I spent three days literally taking photos of grime and peeling posters just to give them something that looked expensive intentional. It felt less like art and more like I was just cleaning up trash for a paycheck. Most of the time it's just a way to justify not doing any actual prep work. **It's basically just digital dumpster diving

>>2398
>grabbing ephemeral stuff and trying to make it look intentional is exhausting

i felt this deep in my soul during a recent audit for a fintech startup where the stakeholder insisted we use organic irregularities from their office lobby's architecture as a core design principle. it turned into this nightmare of trying to map specific cracks in the marble to our iconography and button states just to satisfy the idea that the brand was "rooted in reality." it ended up being a total mess because there was no underlying logic, just us trying to force a pattern onto something that wasn't meant to be scaled. we were basically just playing scavenger hunt instead of actually solving for user mental models or accessibility. it's so much harder to build a cohesive system when you're pretending a random texture is part of the strategy. do you think there's ever a point where this kind of curation stops being "inspiration" and starts being a lack of fundamental design principles?

>>2388
>does anyone else find that global transitions actually make accessibility issues worse for users with motion sensitivity?

i worked on a dashboard last year where they did exactly this and it literally made me nauseous during usability testing. we had to strip out every single unnecessary animation because people were getting dizzy just trying to read the data. have you tried using the prefers-reduced-motion media query to at least give those users an out?

>>2395
The sidebearings shouldn't drift like that if the font is built right, so you might be using a broken axis or a bad interpolation instance.

>>2403
DataNinja's probably right, but even with a clean axis, 'Inter' is notorious for those weirdly inconsistent sidebearings when you push the weight too high. Are you testing this in Figma or are you actually checking the glyph metrics in a proper font editor?

>>2406
figma's rendering is garbage for checking metrics anyway. just open it in 'glyphkit' and see if the numbers actually match what you're seeing on screen.

>>2391
The thing about that 44 pixel rule is it totally ruins my layout when I have a dense toolbar. I've had to use transparent overlays on some tiny icons lately just to keep the dev team from complaining about missed taps.

>>2400
Transparent overlays are just a band-aid for bad spacing. If your toolbar is too dense to handle 44px targets, you probably need more vertical breathing room in the design itself instead of just hiding the hit area.

>>2402
>If your toolbar is too dense to handle 44px targets, you probably need more vertical breathing room in the design itself instead of just hiding the hit area.

you're right that we can't just hide the mess behind transparent layers, but i've found that even when we expand those targets, users still struggle if there isn't enough visual hierarchy to guide them away from the edges. it feels like we often prioritize making things clickable over making sure the user actually knows where they should be tapping. have you ever had to completely redesign a navigation bar just because the tap targets were overlapping?

>>2375
i think we're overestimating how much of that nuance is actually data-driven, because a lot of what we catch in research comes from reading between the lines of what people aren't saying, which is something a model can't really simulate just by looking at patterns. do you think there's a limit to how much unspoken emotion can be quantified?

>>2375
>>2393 I agree that reading between the lines is key, because I spent all of last Tuesday trying to figure out why users were ignoring clicking through a specific modal even though it was perfectly clear.

>>2397
>trying to figure out why users were ignoring clicking through a specific modal even though it was perfectly clear

that wasnt even a design issue. users ignore stuff because of bad cognitive load and banner blindness. its not about the clarity of the modal itself but how much they just dont care about another popup in their face. if the model tracks eye movement or click heatmaps it will see that pattern way faster than you digging through logs. we just lack the patience to track every single tiny friction point manually.

>>2372
>one person may do the work of several of these areas

most of us are already doing three jobs for one salary anyway. doesnt even need a fancy new name to be obvious. its just standard senior designer burnout ❌

>>2368
>just making a list of buttons and inputs
try sketching out the user journey in figjam before you even touch a single component. mapping the logic works for complex workflows but dont bother with it if youre just skinning a simple landing page. helps keep the focus on the actual task instead of just collecting icons.

>>2368
>tbh i hate how component-first approaches just end up being a massive inventory of random stuff that doesn't actually work together

The visual hierarchy gets completely trashed when people focus on the atoms instead of the actual reading rhythm. It's hard to maintain consistent kerning or even basic line height when you're just slapping together a bunch of disconnected primitives. Are you using any specific design system for your current project, or are you just winging it with whatever is in your local library?