/ui/ - UI/UX Lab

the confetti problem is real. it's fine to have a nice animation for finishing a business setup, but seeing that same burst after updating the 47th record of the day feels totally patronizing. figma's agent can follow every rule in our system and still deliver an 'emotionally inappropriate' response because it doesn't know if the user is in a rush or celebrating. i'm worried we're just going to end up with a bunch of perfectly polished but deeply annoying interfaces. do you think we'll actually be able to prompt for 'vibe' or is that just more bloat?

found this here: https://uxdesign.cc/ai-can-generate-the-interface-what-makes-it-worth-caring-about-a5f3a3e81986?source=rss----138adf9c44c---4

nn/g's move to experience specialist actually makes sense if you think about ai agents booking flights or filing complaints without us even being in the loop. i don't know how we're supposed to map a customer journey when there isn't even a human touching the interface anymore.

link: https://www.nngroup.com/articles/ux-vs-experience-design/?utm_source=rss&utm_medium=feed&utm_campaign=rss-syndication

i'm tired of seeing heavy background color changes every time a cursor moves over a link. try designing a small component this week using only subtle border or underline shifts instead.

just read something that really called out how we act like our own moral convictions are some kind of universal truth for users. it's so true that deciding what to make visible or where to add friction is actually a huge responsibility, not just a way to signal how "good" we are as designers.
>our convictions carry weight because we make decisions for other people
ngl i feel like most of my client work is just trying to balance protecting users from harm vs. their autonomy without being a total hypocrite.

https://uxdesign.cc/virtue-ethics-in-everyday-interface-design-5f2aca841781?source=rss----138adf9c44c---4

i'm tired of seeing 1.2 leading in sidebars where everything just bleeds together. Try setting your tightest navigation component to exactly 1.4 and see if it actually stops the eye-strain.

the idea of using the macos notch to annotate everything on your screen sounds like a total nightmare for my focus. apparently they're moving away from just typing in chat boxes and now you can basically use video or screenshots to fix layout issues. it's supposed to bridge that gap when the design system fails at the last mile. i saw they're claiming agents are hitting about 90 percent accuracy now instead of the old 50 percent range. still seems like a lot of manual babysitting if you're just marking up figma comments or pointing at pixels. it's basically just another way to do visual QA but for an agent instead of a person. i can't imagine how much time we'd waste just drawing boxes on screenshots all day. feels like we're just trading one type of tedious work for another. do you think anyone actually wants to spend their afternoon annotating screen recordings?

https://www.lukew.com/ff/2170/design-feedback-for-ai-agents

i've been seeing so many devs use tiny lottie loops to mask slow api calls but it's actually helpful deceptive for people with cognitive impairments who need to know when things are actually stuck. we should be focusing on visible progress indicators instead of just making the interface look pretty while the backend is crawling. does anyone else find that these little animations end up making users feel more anxious once they realize nothing is happening?

was reading about the differences between managing contractors versus using an EOR for people in places like brazil or japan. it's basically just comparing how they handle onboarding and payments without the whole legal entity headache.
>the company typically contracts directly with the freelancer
i don't know if i'd trust a platform to handle all that compliance stuff alone when things get actually complicated with local regs. anyone else using these for their freelance team yet?

more here: https://uxplanet.org/deel-vs-rippling-for-contractor-management-which-platform-gets-the-user-experience-right-275775f329d8?source=rss----819cc2aaeee0---4

just saw someone recycling that building analogy again.
>the top-most plane is the Surface, which is the most visible or concrete aspect of ux design
it's such a cliché way to explain why we don't just paint pretty buttons on top of broken flows. do people still actually use this framework for anything besides portfolios?

found this here: https://ixdf.org/literature/article/the-relationship-between-visual-design-and-user-experience-design

saw some tests comparing 'frontend design' vs 'ui/ux pro max' results. it's wild how much the output changes just by tweaking the prompt for taste skill. honestly feels like a lot of hype when the actual layout logic is still pretty mediocre. idk if i trust it with anything more complex than a basic flexbox setup

link: https://uxplanet.org/i-tested-3-claude-code-skills-for-web-design-which-one-is-best-521137220dc1?source=rss----819cc2aaeee0---4