[codingdave]

I think you are confusing Graphic Design with UX -- Just using your first example, Google has an awesome UX. One field, one button, and everything is about the results. Its simplicity is its design. UX is not about how pretty a site is... it is about how smooth of an experience the end user receives.

UX is required for success. You can have the best engineering in the world, but if your customers don't understand your product, cannot work with it, or get frustrated by it, all that beautiful elegant code means nothing. Products need to solve a problem, not add new pain to existing problems.

That does not mean that engineering is less important. Trying to put one above the other is like answering whether the chicken or the egg came first. They are two components that complement each other.

[wintom]

Your first paragraph about Google having great UX, one button one form.

That was built by engineers not by designers or by UX guys who do "research".

So my argument stands.

The best UI/UX is the type that gets out of your way. That is If there was a way to do away with UI and Ux all together and simply control our machines with thoughts or electrical impulses that would be the best UI and UX there is. Engineers know this that all of that stuff just gets in your way, and if they don't know it they quickly learn it through seeing their users interact with their software.

We don't need a bunch of people standing around doing "research" telling us that.

[SapphireSun]

I would argue that the UX was designed by engineers with a solid sense of what they wanted and what the customer would want. Whether Larry and Sergey were aware of it or not, they were acting as a designer. I've seen many engineers with exceedingly poor design sense.

For instance, a valid choice for Google would have been to clutter the landing page with advanced search options or ancillary material as so many of its contemporaries did. In this case, I would argue that the engineer was skilled at a small set of design tasks.

For what it's worth, I'm an engineer, but I'm close friends with designers.

[trisomy21]

Exactly. The uncluttered search experience was very much a user-centered design decision and it was indicative of the fact that Google valued my needs as a user.

[kagamine]

> That was built by engineers not by designers or by UX guys who do "research"

Except that what Google did was look at existing search engines and think 'how can we differentiate?' and the result was an (originally) ad-free, no links to other services site. They stripped all the bad design from thier industry competitors and left a service in which the new or naive user was in no doubt about what to do there. No distractions, no fiddling, no - anything except full focus on Google Search.

The search results are what you get after you use the search functionality, that those results were good could be argued as secondary to what got users to switch from Yahoo, Ask Geeves etc in the first place.

[sizzle]

You have a huge chip on your shoulder, geez. Your argument is missing some logic. The engineers at Google are building from UX design specs.

"if they don't know it they quickly learn it through seeing their users interact with their software."

Engineers don't do ethnographic research to avoid these problems in the first place. Why subject users to painful workflows when UX teams can solve these issues early on in the SDLC and save the dev teams from doing MORE work later on fixing these usability issues?

[codingdave]

Google has been performing A/B testing on its design since 18 months after it launched. If you go look through archive.org, you'll see that the UX has been tweaked over the years, adding fields to the home page early on for language options, removing it later, tweaking how many links show on the front page, moving options around, etc.

If you think 15 years of UX tweaks, driven by A/B testing, is not research, then please clarify what kind of research you are arguing against?

[kriro]

"""We don't need a bunch of people standing around doing "research" telling us that.""" As an engineer I very much disagree with this sentiment. I usually prefer to base my decisions on quantitative methods. Sure you can say "it's obvious that design should get out of the way". Even if I agree with you I'd still prefer to run some tests using actual data instead of taking your word for it.