5 ms·

We have army of ux professionals these days. But I think engineers do better UX than most so called ux professionals and I say this as a ux professional myself

by pcurve 7y ago

We have army of ux professionals these days. But I think engineers do better UX than most so called ux professionals and I say this as a ux professional myself. we've invented a colossal industry with good intent in the beginning, but over past 10 years, I've seen it degenerate into desperation for relevancy by constantly introducing new things. (things aren't much better in SE) Can we get back to just doing things? It is extremely frustrating working in software design space, from start to finish. Everything sucks. Process. Tools. Speed. Complexity. Politics. Pretending. Is it just me?

martin_drapeau 7y ago
I agree with engineers doing a better job. Especially over inepexerienced UX professionals who want to try the latest fad. Ultimately users don't care - they want it to work consistently and be intuitive.
- Aeolun 7y ago
  The most usable application I’ve seen recently (including the ones I’m working on :( ) has been our 2000ish era work time entry system. My initial gut reaction when I saw it was ‘oh god, how ugly’, but then I noticed how pleasant it was to use.
  - mixmastamyk 7y ago
    A usable, predictable app needn't be ugly and vice-versa.
- pcurve 7y ago
  " Ultimately users don't care - they want it to work consistently and be intuitive." Yes. And if you get these 3 aspect experience right, you can basically get away with murder: Quick performance, accurate info/data, and forgiving/recoverable from error/mistake. What is looks like, for the most part is inconsequential.
omniscient_oce 7y ago
I don't get these fake persona things I see in portfolios and on Behance and stuff. Are these a legit thing or just something that is taught in school so students do and put on their portfolio but isn't used in practice. I get the basic idea behind it but a lot of the time it comes off as quite pretentious, or filler work.
- WillPostForFood 7y ago
  I think they are a symptom of an underlying problem: building a product not for yourself, but for some imagined customer and need. It puts your product in peril because it adds guesswork. But if you have to do it, personas at least challenge you to put yourself in someone else's mindset.
- MattGaiser 7y ago
  I have seen the persona thing everywhere from startup accelerators to books on effective pitching as a consultant. It is a useful sales tool as then the people people management types who can easily relate to the persona.
- GuiA 7y ago
  They’re a bit like the UML diagrams of design. They have some utility in certain contexts but they are taught as a ready made solution without that context to students who aren’t able to relate it to real experience and end up overdoing it and misapplying it as they tend to do. It doesn’t hurt that doing it makes you feel like you’re doing something useful, it gives you a shiny deliverable, ... but in the end there likely are better ways to answer the kind of questions you really want to answer, and many of them involve making the actual thing vs making things that only serve the planning of the making of the thing.
- pcurve 7y ago
  Not just persona. There's journey mapping, storymapping, design thinking, storybrand exercise, designops, formative/summative/ generative research, ethnographic study, service design, heuristics, lean ux, contextual inquiry. I'm not saying these are all useless, but many are just old things re-packaged. And I've noticed that UX people are rarely challenged by other stakeholders because no one can possibly keep up with what any of these mean. It's part FUD and part obfuscation. If you you have good handle on who your customer base is (current and target), then no, you don't need personas. Just use real data. Where I become dumbfounded is when large companies with mountains of customer data with complicated segmentation and profiles continue to rely on just 5-6 same personas.
wott 7y ago
No, it's the same with urbanism for example. If I look at what was built in my country before, say, WW2, it all makes sense, even if we just consider private housing development: numerous streets around small blocks, which guarantee many communication options and no concentration of traffic on single exits, streets on the outside of the blocks, so that it can be used by outsiders too, and so that it allows future extension without compromising any communication. This could be made simply for working-class neighbourhoods or company towns, and yet it was well designed, and has aged well. But in the last 40-50 years we have had gazillions of educated, graduate, certified, professional urbanists and architects, and those have produced all those closed or almost closed subdivisions, which are bad to almost everything related to communication by any mean (foot, bike, but also by car, because it unbalances the traffic between places), and forbid any kind of evolution (extension, inclusion, change of destination). The same also have validated the opposite: totally un-organised subdivisions, which produce the exact same result: for each group of 2 or 3 lots, you have a lane to the main road, perpendicular to it, and of course making a dead end on their side. With the additional penalty of blocking the vision for anyone going along the road, by a never-ending streak of artificialisation. Example: https://www.google.fr/maps/@43.4593112,1.3756329,1321m/data=!3m1!1e3?hl=fr https://www.google.fr/maps/@43.4593112,1.3756329,1321m/data=... And they kept on doing that despite the evidence that it has almost no pro and plenty of terrible cons (otherwise said, it is utterly stupid shit). And they keep on doing it everyday, still. ------------ In my opinion, those situations have a degree of similarity: the advent of a horde of graduate professionals, who do worse than what has already been done, despite having been exposed to more experiences and more results of those decades of experiences, and having studied them.
- pcurve 7y ago
  dumb question... when there's a lot of space to spread out, what would be a better way to lay out subdivisions off of main road?
  - barrkel 7y ago
    There shouldn't be a lot of "space to spread out"; that's car culture, not human scale development.
a1369209993 7y ago
> engineers do better UX than most so called ux professionals Yes, exactly. Beyond a bare minimum of having a UI (which, to be fair, many engineers forget) good UI design is about not doing things.
neop1x 7y ago
They are designing more for eye candy, less for usability. For example it is common to have gray text on white background, hard to read. Black themes for programming IDEs often have dark gray background, reducing contrast. Scrollbars are now thin, hard to see, often disappearing automatically and sometimes it is hard to distinguish which part is the scrollbar handle and which the background. Sometimes it is the same with switches - on/off. Is that switch on or is off? Sometimes hard to see as both states are similarly bright just different color.

by pcurve 7y ago

martin_drapeau 7y ago

I agree with engineers doing a better job. Especially over inepexerienced UX professionals who want to try the latest fad. Ultimately users don't care - they want it to work consistently and be intuitive.

Aeolun 7y ago
The most usable application I’ve seen recently (including the ones I’m working on :( ) has been our 2000ish era work time entry system. My initial gut reaction when I saw it was ‘oh god, how ugly’, but then I noticed how pleasant it was to use.
- mixmastamyk 7y ago
  A usable, predictable app needn't be ugly and vice-versa.
pcurve 7y ago
" Ultimately users don't care - they want it to work consistently and be intuitive." Yes. And if you get these 3 aspect experience right, you can basically get away with murder: Quick performance, accurate info/data, and forgiving/recoverable from error/mistake. What is looks like, for the most part is inconsequential.

omniscient_oce 7y ago

I don't get these fake persona things I see in portfolios and on Behance and stuff. Are these a legit thing or just something that is taught in school so students do and put on their portfolio but isn't used in practice. I get the basic idea behind it but a lot of the time it comes off as quite pretentious, or filler work.

WillPostForFood 7y ago
I think they are a symptom of an underlying problem: building a product not for yourself, but for some imagined customer and need. It puts your product in peril because it adds guesswork. But if you have to do it, personas at least challenge you to put yourself in someone else's mindset.
MattGaiser 7y ago
I have seen the persona thing everywhere from startup accelerators to books on effective pitching as a consultant. It is a useful sales tool as then the people people management types who can easily relate to the persona.
GuiA 7y ago
They’re a bit like the UML diagrams of design. They have some utility in certain contexts but they are taught as a ready made solution without that context to students who aren’t able to relate it to real experience and end up overdoing it and misapplying it as they tend to do. It doesn’t hurt that doing it makes you feel like you’re doing something useful, it gives you a shiny deliverable, ... but in the end there likely are better ways to answer the kind of questions you really want to answer, and many of them involve making the actual thing vs making things that only serve the planning of the making of the thing.
pcurve 7y ago
Not just persona. There's journey mapping, storymapping, design thinking, storybrand exercise, designops, formative/summative/ generative research, ethnographic study, service design, heuristics, lean ux, contextual inquiry. I'm not saying these are all useless, but many are just old things re-packaged. And I've noticed that UX people are rarely challenged by other stakeholders because no one can possibly keep up with what any of these mean. It's part FUD and part obfuscation. If you you have good handle on who your customer base is (current and target), then no, you don't need personas. Just use real data. Where I become dumbfounded is when large companies with mountains of customer data with complicated segmentation and profiles continue to rely on just 5-6 same personas.

wott 7y ago

No, it's the same with urbanism for example. If I look at what was built in my country before, say, WW2, it all makes sense, even if we just consider private housing development: numerous streets around small blocks, which guarantee many communication options and no concentration of traffic on single exits, streets on the outside of the blocks, so that it can be used by outsiders too, and so that it allows future extension without compromising any communication. This could be made simply for working-class neighbourhoods or company towns, and yet it was well designed, and has aged well. But in the last 40-50 years we have had gazillions of educated, graduate, certified, professional urbanists and architects, and those have produced all those closed or almost closed subdivisions, which are bad to almost everything related to communication by any mean (foot, bike, but also by car, because it unbalances the traffic between places), and forbid any kind of evolution (extension, inclusion, change of destination). The same also have validated the opposite: totally un-organised subdivisions, which produce the exact same result: for each group of 2 or 3 lots, you have a lane to the main road, perpendicular to it, and of course making a dead end on their side. With the additional penalty of blocking the vision for anyone going along the road, by a never-ending streak of artificialisation. Example: https://www.google.fr/maps/@43.4593112,1.3756329,1321m/data=!3m1!1e3?hl=fr https://www.google.fr/maps/@43.4593112,1.3756329,1321m/data=... And they kept on doing that despite the evidence that it has almost no pro and plenty of terrible cons (otherwise said, it is utterly stupid shit). And they keep on doing it everyday, still. ------------ In my opinion, those situations have a degree of similarity: the advent of a horde of graduate professionals, who do worse than what has already been done, despite having been exposed to more experiences and more results of those decades of experiences, and having studied them.

pcurve 7y ago
dumb question... when there's a lot of space to spread out, what would be a better way to lay out subdivisions off of main road?
- barrkel 7y ago
  There shouldn't be a lot of "space to spread out"; that's car culture, not human scale development.

a1369209993 7y ago

> engineers do better UX than most so called ux professionals Yes, exactly. Beyond a bare minimum of having a UI (which, to be fair, many engineers forget) good UI design is about not doing things.

neop1x 7y ago

They are designing more for eye candy, less for usability. For example it is common to have gray text on white background, hard to read. Black themes for programming IDEs often have dark gray background, reducing contrast. Scrollbars are now thin, hard to see, often disappearing automatically and sometimes it is hard to distinguish which part is the scrollbar handle and which the background. Sometimes it is the same with switches - on/off. Is that switch on or is off? Sometimes hard to see as both states are similarly bright just different color.