Hacker News Reader

Comment by bluGill | Hacker News Reader

Does anyone apply formal usability studies anymore? (not just do them - do they apply the results). I was trained in formal UX long ago, and many UIs today outright violate common rules we learned back then. High contrast was considered important for low vision users even if it looked ugly then. Now most UIs are lower contrast and while they look nicer, they are less usable.

Of course I learned long ago and so I'm willing to accept that what was best practices back then might be wrong given new research that I'm not aware of. However this seems unlikely.

View on HN

Formal usability testing is still practiced in enterprise, military, and professional products.

From what I've seen from colleagues in the consumer space they've predominantly switched to A/B testing and statistical clickstream/log analyses, though large companies sometimes still do run lab studies for important decisions and to rationalize the giant one-way mirror they installed.

It depends on how big/important/popular of a product you're talking about, but most software that your average joe uses (operating systems, native communication apps, national websites) likely either conduct significant usability testing or largely inform their designs using great usability testing from people like the mozilla foundation and the nielsen norman group. And no-- neither of those organizations have to be perfect for that to be a perfectly valid approach, and if you follow that guidance, you'll never end up with a 'very low contrast' interface.

I go to the store to buy a hammer. I’m a hammer user and I use the hammer.

Or I go to the store and I buy heroin. I’m a heroin user and I use the heroin.

These UIs we speak of, are they hammers, or are they heroin?

Because I’m currently looking at a high contrast screen using a mixture of Unix tools created from the late 60s to yesterday.

I’m looking at an Excel spreadsheet made by a team of lawyers that is in a UI. The lawyers themselves highlighted and handled the contrast of the document. Beyond the table itself, Excel’s menu bar is easy to read for the basic operating needs of a spreadsheet. I mean come on, the pivot table UI is clearly masterful. (If Excel was a native plaintext tabular format and you could pipe stdin/out to the GUI app? Gimme!)

However, a lot of software is basically useless these days. It’s not designed from the beginning to be productive in any shape or form.

Some stoned dance by a junkie under a highway overpass is definitely some kind of expressive art form. How much more productive is the expressive art forms promoted by TikTok, Twitter or Twitch?

There’s clearly a scale at play. I’m as big fan of Twitch as much as I’m a big fan of the Buffalo Bills. I’ve always appreciated the sociological aspects of sports. To add some context to my definition of productivity, I would say that group sporting events are overall productive from a sociological perspective, be it IPL cricket or Fortnite tournaments.

TikTok is clearly and quite obviously towards the junkie dance scale of things.

From an experiential position there was not much of a difference between my observations of a woman filming herself doing a weird little dance on the sidewalk while completely oblivious to the rest of the world and the junky on the next block doing a weird little dance on the sidewalk while completely oblivious to the rest of the world. This basic object-level assessment of the actions that we and others experience with our physical bodies is core to the very notion of “meaning”. The “meaning” is just the stoned dance.

I know of no junkies famous for only being junkies but there are definitely TikTokers famous for only being TikTokers so don’t confuse this for some strict equivalence.

I’m just trying to establish a kind of framework for discussing the relationship between productive UI and productive software tools.

> These UIs we speak of, are they hammers, or are they heroin?

A hammer is useful for doing stuff, which modern UIs increasingly aren't. So it seems they aren't hammers.

A)Yes. I've been in the software business in some capacity for the better part of 25 years and software usability right now is miles beyond what it used to be. Thinking otherwise is a function of nostalgia.

B) if you're doing dev work, you're probably not entirely cognizant of how different your software selection and usage patterns are than most people's, even for non-dev things.

C) there's a whole lot of interfaces out there and the great ones work so naturally and intuitively that you don't even notice them. Those are the ones for which many assume the design decisions were so obvious that they didn't even need designers. That's so so not true.Look at the most popular user-facing software: web browsers are all obsessed with usability (mozilla actually does some of the best open usability research out there.) Instagram, Tik Tok, Facebook, Twitter, et al... Incredibly clear and usable for their target market. Can't get much higher contrast than you do in email applications and other communication apps. There are tools to guage if your text color/background color/ and text size make for readable text according to the WCAG and I have never had a job, either as a designer or as a developer, where AA compliance wasn't a minimum for all functional elements. Most common user-facing interfaces that regularly use light or white text on colored backgrounds, like the iMessage interface or signal, benefit from OS-level accessibility enhancements like increased text size, which are very commonly used among the people who need them. My non-tech-savvy elderly relatives all knew about them through the setup process on their devices when they first bought them, and never looked back. Looking through every communication or other popular user-facing software package on my Galaxy U22 using a default theme shows nothing even remotely low-contrast.

Reply on news.ycombinator.com