Oliver Nassar

URL Patterns: Underrated and tough to get right

July 27, 2011

The recent Hacker News-discovered story Even a URL tells a story has shown light on something I've been thinking about for a while: how hard it is to generate a thoughtful, distinct and informative url architecture.

I've always held apple.com as the epitome for this, as does the article linked to above. Reilly Brennan showcases an example of Sony, whereby the URL http://store.sony.com/webapp/wcs/stores/servlet/CategoryDisplay?catalogId=10551&storeId=10151&langId=-1&categoryId=16178 references a set of MP3 players. A comment on the Hacker News link brings up the fact that while both Apple and Sony are most likely using similar technologies (Java server), they go about managing their url schema entirely different.

What I respect so much about this is the UX it leads to. A comment on HN made a note that the primary way people will find either a Sony MP3 player or an Apple iPod will be via a search engine (which is arguable), and therefore the actual URL of the resource doesn't matter all that much, but for me, it's about the visibility of those URL's.

Google isn't the only way URL's are exposed. Whether you're sending them on IM, via email, viewing them on posters, or looking at your address bar, within milliseconds you know what the resource is, and what the parent resource is (eg. at http://www.apple.com/ca/macbookair/, I know that I'm looking at a Macbook Air on the Canadian site).

Beyond the UX, though, I'm equally impressed by the managerial requirements that it must have taken to get to that point. Organizational overhead is common in every company, let alone the world's largest. The emails, excel documents, and countless rebuffs it must have taken from IT to get to that point isn't easy. It tells me that there is a level of organization at that company (even if just on the customer-facing side) that other companies should strive for.