[jenius]

Justifying design decisions is often extremely difficult. Although there is a science buried deep behind design decisions, unlike programming, design decisions are often very subjective and opinion-based in a debate, unless you have done A/B tests extensively before said debate. Most of the time, a designer will do something because 'it looks better' or 'it feels right', and redesign something because 'it looks bad'.

I can see what you are saying here, but at the same time you are proposing something very difficult for this designer. In my past, whenever I have been in a debate over design decisions, there is no objective criticism involved. Even if I've done extensive research beforehand and am citing what I imagine users will do and artistic theory that backs the decisions I made, it can always be disagreed with. "Yeah that might be the case, but it doesn't look as good." or "No, users would do this instead, and want this." Yes, surprisingly that flies in design debates.

I don't think that Kyro should attempt this unless he is going to run A/B tests so that he has actual data to back his reasons. Debate based solely upon opinion are fruitless, never-ending, and are bound to come up all the time in such a public and critical forum as hacker news.

That being said, I don't even think that he should be running A/B tests, and shouldn't need to provide any additional justification. Not only is this a ton of work for something he is doing for personal practice in his free time, but it's often a waste of time. A good designer must be trusted. You hire a designer if you like his work and trust his design decisions, not if you are going to nitpick and tell him to justify and test every shade of color he has used.

I think the most famous case of this happening is with with Douglas Bowman quitting at google to move to twitter as the lead visual designer. Mr. Bowman is quite clearly one of the best design and ux guys out there, and is widely respected. This article provides a lot of insight into his decision, and it came down to them questioning his decisions and asking him to prove everything, which was in the end a waste of time.

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

[bmelton]

While I agree with the sentiment of your message, I think that the designer loses something when people give him props for a good looking page that doesn't work. "Not working" is a relative term of course, but assumptions should always be based on something, and decisions shouldn't be arbitrary.

I think it would be worth it to the designer to justify the decisions to himself, and let us try to tear them down, as that will further his learning. I'm assuming that by doing these exercises, he's trying to grow, and that puts him in an entirely different situation than with Bowman.

The flip side of that coin is that no design is proven until it's A/B tested. You can have awards for the site's beauty, but if a significantly uglier design increases sales, then you've lost the plot. Keats writes really great poetry. The writing is beautiful. I would no sooner put a Keats poem as my website copy than I would an Ansel Adams photograph as my website's visual design.

This is why website design is harder than people think. This is what separates the good from the great.

The designer is of course under no obligation to us to do this, but it may help himself. Even if he doesn't bother to justify things out loud, sometimes the act of thinking through justification in your head is enough to solidify (or compel change to) a design decision.