Over the weekend I watched Wait Until Dark, the 1967 film starring Audrey Hepburn, Richard Crenna, Alan Arkin, and the awesomely named Efrem Zimbalist, Jr. Here's the summary from Netflix:
"Audrey Hepburn was nominated for an Academy Award for her portrayal of Susy Hendrix, a blind woman terrorized by a criminal (Alan Arkin) who's after a stash of heroin that was planted in her apartment -- inside a doll -- without her knowledge. He manages to isolate Susy, and she must find a way to defeat him in a thrilling, chilling battle of wits."
Based on a play from 1966, Wait Unitl Dark is a textbook example (and that is not meant to be derogatory ) of the idea "If you introduce a gun in the first act, be sure to have it fire in the third." Through out the movie, small elements and conversation pieces grow and become important during the climax. Characters expose motivations at the end, all in relation to previous set pieces and dialogue.
And while there aren't any guns, there sure are a lot of knives.
"But Jones", you say, "what does this have to do with design? What made you even think this could be a post?"
Simple: If you introduce a gun in the first act, you should fire it in the third. Stop making arbitrary design decisions just because something seems cool.
I think all the elements of a design should work towards relaying information to a specific audience in a specific manner. As a matter of fact I think that is almost rule number one of design. Rule number 2 may be "Masses be damned" but that's a post for another day.
It's so easy to fall back on current design trends and superficial elements. I've done it plenty of times (especially when I'm under a tight deadline). But when you remember that design is meant to deliver information to a user in a very specific way, every single visual choice should go back to your original goals. Basically, all of your elements should serve the 'story' of your design. If it doesn't, take another look at it:
Do I really need that 15th photoshop texture layer? Do I really need a shiny reflection under my logo? That gradient I want to make... It should have it relate to the information via color or degree of fade. Do you want to use trompe-l'œil? What effect will the 'realistic' images on the page bring to your design?
These types of questions can and SHOULD be asked as you create a design. In Wait Until Dark, every chair, every door, every kitchen utensil is in a very specific place for a very specific reason. It's economical set use let's you get all the information you need, but in a way that is specifically unique to the movie.
I tend to stick a lot on function and form working in tandem, as opposed to function over form. Just take a few extra minutes to ask if the element you're working on speaks to the information and goals of the site as a whole. That pause can keep a site from falling in to the current world of homogeneous Web design and becoming just another Michael Bay flick.
#1: Efrem Zimbalist, Jr. is, indeed, awesomely named.
#2: Comparing anything to a Michael Bay flick is one of the hardest put-downs ever. If you said that about something I made I would cry.
#3: Your questions seem like good ones, if only because I get excited wanting to answer them. I think that feeling of excitement may be one of the key ways to tell whether you're trying to solve design problems via spackle (just throwing @#$% on the walls 'til the problem goes away).
There are good answers, after all, to questions like "Do I really need that 15th photoshop texture layer?" Maybe your web page is all about what that awesome feeling of nostalgia and familiarity you get when viewing a photograph from fifteen years ago and it has all the wear-and-tear of age. Or maybe your shiny buttons are all shiny because the app you're designing is so awesome it's freaking from the future. But, yeah, if you ask yourself the question and there's no zeal there, no real excitement... probably time to peel it back. Tuck it in a drawer and use it at a better time.
I think the excitement of design, as well as throwing things at the wall to see what sticks, are essential for problems to get solved.
What I'm thinking about is how people use those choices and if in 75% of the cases it actually helps advance the design.
If a site decides to use trompe-l'œil then SWEET. A well done trompe-l'œil site is something I really enjoy. But just adding a coffee stain and a texture without thinking if it truly adds to a feeling you want a user to have while interacting with your content? Then you've got a shaky cam in the middle of Public Enemies type of thing (more nerd movie references that only Nate may get).