The Google user-interface perspective

Written by Adrian Holovaty on October 21, 2002

In a interview, Google product manager Marissa Mayer describes her philosophy on Google's usefulness:

I think Google should be like a Swiss Army knife: clean, simple, the tool you want to take everywhere. When you need a certain tool, you can pull these lovely doodads out of it and get what you want. So on Google, rather than showing you upfront that we can do all these things, we give you tips to encourage you to do things these ways. We get you to put your query in the search field, rather than have all these links up front. That's worked well for us. Like when you see a knife with all 681 functions opened up, you're terrified. That's how other sites are - you're scared to use them. Google has that same level of complexity, but we have a simple and functional interface on it, like the Swiss Army knife closed.

This absolutely nails how I think a news site should function. Don't scare me away with hundreds of headlines and links. Do make it easy for me to find the news I want, at a detail level of my choosing.


Posted by Ben on October 21, 2002, at 2:13 p.m.:

I was recently thinking about how well the google interface is designed (at least when compared to some competitors). As mentioned in the article a HCI background high up in the company has had a beneficial effect.

My rough and ready analysis is the centrality of the search function. Google focus on this throughout the site. They use paged views effectively (as per J.Nielson) to present different views of the same query (whether in groups, news or web pages).

I think that news sites need to understand their basic function. Their interface design should flow from this understanding, unfortunately I think it is all too common for advertisers and purely visual designers to be in control of the "design" of a site. I think google is a good example of how focusing on utility and providing an effective service quickly is ultimately a business benefit.

Posted by Jay Small on October 21, 2002, at 4:42 p.m.:

I won't pretend to defend the many news site index pages that have turned into seas of blue, underlined text. Something must be done! :-) But we have to recognize there's a difference between building an interface for one key function -- search -- and building an interface that represents editors' choices of a wide swath of news content.

Google's interface works because it promotes only that one key function, tabbed in the context of a few useful aggregated databases (Web pages, the open directory, groups and news). If only we could just put a simple search window on our news home pages and expect people to find what they wanted, yeah, wouldn't that be nice?

But it won't work.

At UIEast last week, we had it beaten into our heads that search doesn't work all that well for navigating types of content that cannot be placed into commonly known categories. Search works great if you're looking for the song "All That Heaven Will Allow" by Bruce Springsteen in a database of CD titles. Expected query, expected results.

But it begins to fail much more often at delivering expected/desired results (more than half the time at first pass, overwhelmingly at second or third passes) if you're looking for news articles about, say, cancer. Worse, the usability folks say people use search for navigation only after link-based navigation has failed them -- typically, three or four pages into a site.

News sites' answer to this dilemma is to just keep putting more links out there, either navigational or headline-style, rather than attempting to provide more meaningful metadata and categorization of news articles to improve navigation AND search.

Reason? I suspect it's cost. It takes skilled librarians to annotate an article -- newspapers pay them to do so for electronic archives. But very few sites have the means to push over the annotated versions of stories in anything resembling real time, and many news librarians do that markup hours after the fact.

We have much to do on the back end -- in meta-markup and user research -- before we can really enable meaningful workflows for news consumers. Costs for annotation and user research aren't trivial, and the returns on investment can be difficult to prove to an impatient industry.

Sooner or later, though, the largest news sites will have to do something, or their home pages will become absolutely unusable.

Posted by Ben on October 21, 2002, at 7:12 p.m.:

I just want to clarify that I don't believe news sites and search sites are the same thing. I was advocating user centred design rather than design driven by other interests. Online news sites are similar to the development search engines were taking a few years ago.

What is needed is a fundamental change in how many news sites deliver the content to their users. You are right to mention the failure of search engines in sites, I am not suggesting search engines are the answer, what I am proposing and promoting is the centrality of good HCI in good online design.

Without innovative thinking on the part of news sites all we are going to get are front pages filled with "blue links". Unfortunately the thinking is not always being done by the best people for the job.

I don't know what the answer(s) for effective online news distribution is, but I know that usability will play a key role in distinguishing between competitors.

Posted by Jay Small on October 21, 2002, at 8:23 p.m.:

Good points. I didn't assume you meant the two were the same. I see the "why can't news sites be as simple as Google" discussion from time to time, and feel compelled to point out reasons -- good and bad -- why they aren't.

Comments have been turned off for this page.