The Googlish way to do Directories
Sunday, June 2, 2002 by Dave Winer.
Good morning sports fans. A Silicon Valley story. Went to the grocery store to get some coffee and food and wine. At the checkout counter the clerk says "Oh the tourists." I said "Tourists? I didn't know we got tourists here." It turns out we do. They stop in and ask if there's anything to see. She says "Oh yeah, there's Highway 280, go check it out."
I had the same experience when I came to Silicon Valley for the first time in the late 70s. I kept driving around looking for something to see. There's nothing to see. Lots of freeways, Denny's, and not a whole lot more. Stanford has a couple of museums. Winchester Mystery House.
The week before last I took a trip to Google. First I took the tour. Lots of geek toys. Lots of press clippings. A nice graph showing flow. Then we sat down and talked software.
I pushed six things and gave them a heads-up on an seventh. Here they are. 1. Spell-checker web service. 2. Pings from CMSes for more currency. 3. Google On The Desktop. 4. An API to access page rank. 5. OPML and directories (instead of two or three directories, millions). 6. RSS feeds for their news flows. 7. Gnutella as a decentralized distribution method.
This article is about #5, the Googlish way to do Directories.
I heard someone say the other day that they don't use bookmarks anymore, they just go to Google and type the name of the site and it takes them there. (And gives them a few more ideas on the way.)
Driving in the car the other day I found myself wishing I had Google so I could complete a thought I was working on. Had to wait till I got to my friend's house. On arrival I asked if I could use his browser. A funny new social construct.
So why does Google work so well? Because it's open and fair and seems smart (it's just a computer of course). Its algorithms figure out what the Web has to say on a given subject. Collectively we believe in Google, it's our memory, it's the way we share. And it works very nicely with the tools we sell which are for creating things to share, and more often, every month, the sharing happens on Google.
So what does Googlish mean? Open, fair and smart.
Yahoo started as a directory, before there was a Google, before Alta Vista even.
The directory has a top level, where the world is divided into categories, sub-categories, and so on. To find what you're looking for, pop in and out levels, and read. Each level in the directory has an editor, who works with other editors to build a hierarchic map of the Web.
There are two main directories: Yahoo and DMOZ. Yahoo is maintained by employees and volunteers, although the editorial process is not publicly documented. DMOZ is all-volunteer, but both approaches centralize control of categories.
They are not Googlish. The Web doesn't have two or three home pages. It has millions. The right way to do it is the same way Google's search engine does. Open, fair, and smart. Lots of competition. No gatekeepers.
Now I'm going to get technical.
Imagine a new format, like HTML, but for hierarchies.
It's called OPML, which stands for Outline Processor Markup Language, an XML-based format I designed in Y2K.
You edit OPML files with an outliner. Several of them support the format now, including the one that UserLand includes in Radio. Eventually, I believe (and hope) all outliners and many other kinds of programs, ones that create and understand hierarchies, will support the format.
You can save OPML files to the Web, just like HTML files, and browse them in lots of interesting ways. Recently we came out with a tool that turns an outline into a PowerPoint-like presentation, but in HTML so you can browse it. It's cool because weblogs are showing up at conferences, and of course so do presentations. Now people can use the same tool to edit both, and achieve interesting integration.
Another thing outlines are good for is authoring directories, like Yahoo and DMOZ. Everyone can edit their own outlines. The next step is for Google to understand them and browse them like directories.
Millions of people can use this tool. It's not hard. That's key, because what we want to do is enable people who have deep knowledge of important areas to gather resources, organize them, and reorganize, as the world changes. Teaching Google how to parse this information, and then present it to users of Google is an important next step for the Web, I'm sure of it.
I've sold hundreds of thousands of outliners. I've met hundreds of people who love them, and get emails every day from people who use them. They are smart people, but generally not the kinds of people who have been writing for the Web because writing for the Web has required so much technical knowledge.
So this little Web trick, that Google can help with, can bring more minds onto the Web faster, something we absolutely need if we're to meet the challenges from government and the entertainment industry. There's really no time to waste.
1. When you encounter an OPML file, index it.
2. When an OPML file shows up in a search, offer to browse it as a directory.
Now I get even more technical. Sorry, but it's important. ;->
OPML directories can link to other directories, they can even (theoretically) link into other directories. When this happens, the linked-to directory is "included" in the other. At the bottom of the page, the author's name is different, and the suggest-a-link feature sends an email to the included directory's author, but most readers won't notice. It's almost seamless.
Now, instead of having two or three all-encompassing directories, anyone with an outliner and some server space can compete to be the authority on any subject. That's how Web directories become Googlish.
Have a great Sunday!!
PS: No patents. ;->