Skip to content →

Author: cgerrish

Unemployed philosopher

Google and Mahalo: People Don’t Scale (But do they scale enough to be useful?)

Rich Skrenta assumes Google. He says that “everyone starts at Google,” therefore no one will start at your site. No one will type Wikipedia, Mahalo or Amazon. I grant that no one will type a complex URL filled with query strings and other detritus intended for reading by machines. But people will type simple names they can remember.

It’s the simple Investment / Value equation. Does Mahalo, or any site, provide more value than the cost of remembering its name and learning how to type it (or bookmarking it)? Today Mahalo probably doesn’t (Unless you want to learn how to speak French). Dave Sifry notes that people come to Technorati organically. That means they type the URL or use a browser bookmark. That’s because Technorati has dominance in its domain. If you want to know what the Blogs are saying on a topic, Technorati has the best results.

Mahalo has the best results for the SERPs that it has. As I’ve mentioned before, Mahalo responds to people who have questions. Google solves the I can’t remember the URL problem. It’s a tool to helps you track down the half remembered, the lost friend, the incomplete query. Through berrypicking behavior Google can lead you to your destination. However Mahalo’s results pages are better organized and more useful to people with questions. Their issue is can the write, edit and maintain enough pages to make the service interesting in the long run. Calacanis thinks they can maintain 25,000 pages. The interesting thing about this is it’s a talent and people management problem, not writing a better algorithm problem. Wikipedia has a similar structure, but it requires the management of volunteers—a more difficult task. The technology community doesn’t believe in this approach, programs are written to solve this kind of problem. People don’t scale. Developers solve problems for zero, one and infinity. Mahalo posits that there’s a number less than infinity that’s useful to people with questions.

Comments closed

Nobody Reads, Nobody Scrolls, Except For Jack

On The Road

There was a nice appreciation of Jack Kerouac’s On The Road by Luc Sante in Sunday’s NY Times. I haven’t read it in years, but at one time it was an important book to me. A book that started on the margins of society has been widely translated and published around the world.

Scroll of On The Road

To read the original, you’ll need to do a lot of scrolling. It was originally transcribed onto a 120 foot long scroll. Kerouac taped together 12-foot sheets of tracing paper to create the original manuscript. On the Road was written in a burst of creative effort—three weeks from start to finish.

It’s hard to imagine a writer doing a reading on a modern television talk show. Steve Allen deserves credit for inviting Jack to read. Something about it seems too tame and too clean, but it’s a beautiful portal to the past.

Comments closed

The Bees of the Invisible

“We are the bees of the invisible,” Rilke wrote to Hulewicz, his translator, “Frantically we gather in the honey of the visible—to store it in the great golden hive of the invisible.”

The process of linking one thing to another on the Web may someday contain a kind of poetry. Today we link things together because they have a certain kind of coherence or utility. But, in truth, in life, the kind of links that stay with us are the ones the produce unexpected connections.

Comments closed

Everyone and No One

In this new media world, Jeff Jarvis thinks that “everyone” can provide the necessary checks and balances to the conscious or unconscious editorial bias on Mahalo’s edited search page results. He calls the idea of an Ombudsman very Old Media.

There’s a sense in which Mahalo is very old media. It employs editors to filter the Web and determine what’s important and what’s not. It’s not a Wiki and it’s not a UseNet Group— the public can suggest editorial content, but the editors make the decisions. Mahalo appears to be structured like a Wiki, but it’s operated by paid professionals on our behalf.

Jason Calacanis commented recently that Mahalo wasn’t a product built for the leading edge of users. It’s a service that aims for the early majority, not the innovators and early adopters. This is one of the reasons that leaving the task of challenging editorial bias to “everyone” probably won’t work.

Of course, challenging editorial bias on Mahalo only becomes an issue if the company and the service is successful. If Mahalo itself is in the margin in the world of Web search then the Web itself provides the counter argument.

Comments closed