[chc]

A coder will generally have an idea about how to do so in the same way that a designer can understand how to architect and code a 900kloc enterprise app — it will be very superficial. In general, other people's jobs always look easy because you don't realize how much you actually suck at them.

[mootothemax]

In general, other people's jobs always look easy because you don't realize how much you actually suck at them.

That's an ironically sweeping statement to make. For example, are all UX specialists designers... or does it just seem easy to designers ;-)

[chc]

1. UX does not "just seem easy" to designers. Most are no good at it.

2. "Are all UX specialists designers?" depends entirely on how we define our terms. If we define UX as a design discipline (as is common, since it basically is), then yes, by definition. If we define it as something else, then no.

[nirvdrum]

I assume for the purposes of this discussion, design and usability have been melded into a single concept. However, sability principles are certainly something a developer can learn and scientifically derive. The SIGCHI proceedings are a great resource.

Making that attractive is a completely different skillset and one I certainly don't have. However, based on the amount of flowery yet wholly unusable crap I come across, it seems many designers don't really have a good handle on usability either. Getting someone great at both is a rare find indeed, but it's not as if there's this huge cognitive gap between the two.

[chc]

This is kind of what I'm talking about. There is some subset of usability principles that you can "learn and scientifically derive" without actually delving in and becoming a usability expert, but that just makes you the UX equivalent of a cowboy coder hired because his uncle heard he knew computer stuff — you might get stuff done, but you will not be a substitute for an expert unless you actually become one.

[nirvdrum]

Well, on the other hand, I've grown weary of "UX" experts that think that means just making things look pretty. The term HCI shouldn't be foreign to them. Usability studies shouldn't just be grabbing three colleagues and asking them what they think (few ever even get that far). There is a lot of psychology that goes into usability. I'll take someone that studies that over someone that only reads other usability blogs (i.e., an echo chamber) any day of the week.

And again, I contend there are people that do both extraordinarily well. And then there's a ton that think they can, on both sides of the fence.

[nirvdrum]

And I perhaps chose the incorrect term in saying "derive" earlier. I don't think conducting a usability study is going to magically surface a design for you. You need some sort of background to influence your hypotheses. But you can test and measure usability and you don't need to be well-versed in CSS 3 or Photoshop to do so was more my point. And you can draw upon well-established principles to get yourself started.

[danilocampos]

(Good) Design is much more about how it works than how it looks.

[Wolf_Larsen]

And if a designer had a working product after the "architect" part, you'd see a lot more designer-only apps being built.

The fact is that a coder CAN get a working product up without the design-equivalent of "coding". They can lay out a UI just as well as an average professional designer (I would know), but the execution will certainly fall short of an average professional designer (I would know).

With that said, I'd always prefer to hire an excellent designer before building a UI, but I rarely will because I like to build more than I like perfect UIs.

[chc]

If the execution falls short, I wouldn't say they did it just as well. Would you say a coder who creates a bug-ridden ball of spaghetti is just as good at programming as one who writes clean, correct, bug-free code, but just doesn't execute as well?