MySpace: The worst HTML/CSS web application design ever?
Believe it or not, I don’t think there’s any hyperbole in this post’s title. How is it possible that MySpace could be worth so much—but have so little contributed to making it attractive, flexible, and well-designed—especially after its whirlwind success over the past 12 months?
I guess the first thing I would question is: Is MySpace actually profitable? My guess is “not yet”—after all, this is hardly the first site to try to provide the same kind of community that the old school BBSes used to have (does anyone remember The Globe, and how they bellied up?). Maybe the time is just right now: The right technologies, the right mix of people, and a big wad of cash at News Corporation to float the site for as long as it needs to to be profitable.
Call me an idiot, but I think MySpace could be designed better from a user interface perspective. I’m not suggesting the whole thing be popped into a rubbish tip and start from scratch; but come on. Using non-breaking spaces to position titles? 4+ nested tables for a design that should be using absolutely no more than one table, and for MSIE-compatibility reasons only? Naming your classes based on the intended appearances of the items they describe followed by obscure numeric references? Not only does 1999 want its “best design practices” back, but they wish the MySpace designers could have at least used them consistently throughout—especially on users’ profile pages, unarguably the most important single document template on the entire web site.
I don’t seriously believe I (or anyone else that isn’t already employed by News Corporation) would ever have a stab at redesigning any of MySpace’s templates, but if anyone out there is listening, the following would be nice:
Valid, structurally sound and semantic-markup XHTML templates
Leave the design tables right out of the profiles and let people design and organize the pages the way they want them to look—after all, they’re doing it already, why not let them go the whole nine yards and actually have the chance at making it look decent to boot? Hey, at least there aren’t any tags …
Clean, simple, transparent CSS
Yes, of course MySpace wants to establish their brand—that’s their right; the site belongs to a corporation, not the users that populate it. Sure, we users make the content, and the content makes the site—but someone at News Corporation has access to the power switch, not you or me.
But the style sheets can still be exposed without risking MySpace’s brand. In fact, they hardly have one on every page anyway. Why not make the CSS for user profiles simple and clear so it’s easy to alter and/or extend to the user’s preference? Don’t worry about complex compound statements: plenty of designers out there have already figured out the extents of what they can get away with in designing MySpace templates already, and they can do it again.
Accessible Design
I know how we should all be grateful when we accomplish something that satisfies 80% of our ‘customers’—the so-called “80% rule” popular in design circules.
If you’re one of the millions whose ability to interact with a computer are affected by a disability, you might even find yourself in the single-digits percentage wise—which is fine for designers, but not for you, since you’re always in that low percentile when a website is designed inaccessibly.
So, you don’t have a disability now, and can’t see the point or why those who are disabled are so whiny? Just wait another few years, and you might: One of the most common ‘disabilities’ covered by WAI/Section 508 is low/impaired vision. Wouldn’t it be nice to be able to trivially resize MySpace to suit your level of vision? How about making it fit the width of your browser instead of all the wasted space to either side of the window if your browser is wider than 800 pixels (I think most of us now have monitors that do higher resolution than that, thank you very much), or be able to easily override someone’s colour preferences to be able to view their profile? Which is the greater impairment to your freedom of expression: having someone supercede your design choices to be able to read and understand your content, or have that person not be able to read your content at all?
Separation of common interface elements from those that are user-customisable
It is almost painful to have to ask: Why can user profiles impact what should be “protected” user interface elements on the page, i.e. navigation bars, branding, and—perhaps most importantly—advertisements (you know, those annoying things that pay for your otherwise gratis use of the site? Yes, I hate them too, but then I wouldn’t be a member of the site at all if MySpace wanted to charge me a red cent to maintain my membership, and probably neither would you)?
I can think of at least three distinctly different means of protecting these interface elements from abuse/user alteration, some of which would be nearly trivial to implement. Again, maybe I misunderstand the value of the above, but if it’s annoying to me that these elements can (and are) frequently interfered with, surely I’m not the only one.
ColdFusion?
This doesn’t really have much to do with the UI, but … ColdFusion? Maybe that’s why so many MySpace features seem to be busted all the time. I’d recommend a more robust technology, personally—Yes, I know CF is backed by Java now, but the whole thing is FuseBoxed—I think they could do better for one of the most popular sites in the world. I’d suggest a different architecture for the back-end, particularly something that’s a bit more scalable and has considerably improved separation of interface code from programming… I’m pretty sure a lot of MySpace’s design woes are caused by the classic failure, “A programmer designed it.”
Please feel free to share your thoughts on this subject or direct me to any similar-minded (or alternative-viewed) posts of your own.