It is kind of like answering on "Jeopardy" and having to remember to phrase your answer as a question.

ChrisL

I agree with the second comment - there's no need for a meeting if something can be metrically proven, just send out a spreadsheet - end of story.

By the way you all can stop developing fonts, we have empirically proven that Verdana is superior.

Philosophically I prefer "I like that" over data driven solutions and academic justifications.

Confucianism sucks.

pbc

It might make sense to say something more informative than "I like the way the screen looks".

Saying that "this new design let users find what they wanted 10% faster" is certainly useful, and if the point is to make the screens work better, while leaving the aesthetics to one designer recognized as competent - because aesthetics are subjective, there's no point debating those in committee - this approach may not be all that bad.

But it's an approach to keep things tightly focused for an organization that is busy and has chiefly utilitarian goals. This would not be the way to effectively decide on a new look for Avatar II - because here the meeting would be one of creative designers... not of computer programmers providing amateur creative input.

What you say is: "I think people will like ...."

hhp

I'm all for collecting data whenever possible and using it to inform design wherever possible. But statements such as "The experimentation on the site shows that his design performed 10% better" sound like hog-wash. No doubt it's taken out of context, but as it stands in the article, it begs the question "10% better than what on what metric!?".

In this particular situation (without the proper context of course), a statement like "I like the way the screen looks" is actually more informative. At least I know that someone likes what's there. I suppose I'm just disgruntled with the amount of jargon and meaningless word salad I hear on a daily basis.

It's much like people saying something is "value added" without explaining where the value has been added. As though uttering the phrase actually adds value in some mysterious way. Similarly, the uttering of a vague statistic doesn't imbue that statistic with meaning or utility.

I suppose it depends on the kind of meeting as well—is it internal within a design department, or is it an external presentation to a group who's natural habitat is money, metrics and quantifiability?

Sounds a bit like a naive desire from unsteady leadership to impose a mathematical level of order and clarity on a discipline that requires a strong coherent aesthetic vision, even in the most user-driven testing environments.

Of course, I've never really thought of Google as a thought-leader or haven of design excellence or process-driven innovator when it comes to design—at least not at the level of groups like Tim Brown's IDEO.

In generating excellent, ambitious ideas, Google is a fish in water—but at the level of design execution (or even really user-interface development) I've always thought of google as still having a bit of a vestigial "engineer's aesthetic."

Reminds me of this:
http://www.diacenter.org/km/index.html

(found here: http://typophile.com/node/41104)

Art and science are not in conflict. They are complimentary.

Websites like Google and Amazon are in unique position to quickly measure their own success. Google wants people to select the website that will likely have what the user is looking for and Amazon wants people to buy things. Both can try different designs for very short periods of time to see if people click on a link at a higher rate or buy more things respectively.

In order to try different designs, someone had to design the new design. Design didn’t go away. The experimentation on the site measures whether the new design has a higher click rate or buy rate than the previous design. And the story doesn’t end there, because new design ideas are still needed to keep improving the click rate and buy rate.

I don't see ‘science’ as involved in this situation at all. It is a matter of art and utility, which may be either complimentary or in conflict in any given situation. Art and utility are often complimentary because humans value art -- by which I mean the extra-utile component of things -- and respond positively to it, enhancing the utility of a thing by making more people want to engage with it. Art and utility will be in conflict, however, if a design that people find more attractive aesthetically actually functions less well than another, less attractive design.

There is also the matter of branding to consider, because while a company may generally favour utility over art when they are in conflict, this may be moderated if the better functioning design is inconsistent with or undermines an established visual identity in which the company has invested.

I am interested in individuals, not polls.

Do they work? Yes.

Do I like it? No. (Oops, let me rephrase that, polls are at least 10% more prone to error, of course that depends on the individual). :)

pbc

polls are at least 10% more prone to error

Paul, that made me laugh out loud. Thanks.

douglas bowman, the first "visual designer" hired by
google, in 2006, left google less than 3 years later:

> Yes, it’s true that a team at Google
> couldn’t decide between two blues,
> so they’re testing 41 shades between
> each blue to see which one performs better.
> I had a recent debate over whether a
> border should be 3, 4 or 5 pixels wide, and
> was asked to prove my case. I can’t operate
> in an environment like that. I’ve grown tired
> of debating such miniscule design decisions.
> There are more exciting design problems
> in this world to tackle.

> http://stopdesign.com/archive/2009/03/20/goodbye-google.html

also see here, for comments from the public:
> http://www.zeldman.com/2009/03/20/41-shades-of-blue/

-bowerbird

Hadn't caught that bit from Bowman on why he left Google before...interesting read. Someone should write a book on the engineering-aesthetic. It would deserve a chapter on remote control designs that smack usability in the face.

I heartily echo the conviction that, "I won’t miss a design philosophy that lives or dies strictly by the sword of data."

> I had a recent debate over whether a
> border should be 3, 4 or 5 pixels wide

this is exactly what you should be concerned with. Sure there shouldn't be a debate over it (probably diversionary tactics by colleague to change something else), but consideration over such matters means that you care about what you are doing.

/m

The question is not between "I like ..." and "The user will like ..." but for a person to provide reasons and rationale _why_ they find a feature preferable. So "I like ... because ..." is the better way to make the statement.

This will be a bit difficult for a senior designer, who has developed good taste. But in most cases tracing the reason why the design works (or works better) can be found and explained to the layperson.

Subjective and objective critique are important. Google is great at objective critique. They're so good at it, we tend to forgive them for their lack of subjective critique (and subsequent lack of refined aesthetics).

I say "I think XYZ will do the job better because…".
It's about functionality, and our job is to make it happen.

Sure it depends on context.* At least in some contexts (and certainly the ones I'm in), «we could prove that this performs 10% better» sounds about as credible (or as pertinent) as an ad for hair removal cream on QVC.

An artist likes and a scientist describes why he likes it. In a corporate meeting I can see how being a scientist would be more productive.

...................
Get My Fonts Under Freebies

An artist likes, a graphic designer likes and describes why he likes it and why other people should like it, and a scientist tries to determine empirically whether people like it.