The FezGuys
Updating the Book of Style
[ No. 60 - October 2001 ]

We're not going to capitalize the word "internet" any more. Yeah, it made sense at the beginning. Five years ago there was something exciting about using a computer network, created by military and academic engineers to safeguard data in case of atomic obliteration, for the whimsical purpose of having friendships with people we'd never meet who heard our songs in the comfort of their homes half a world away. We had a new technological toy, one that would (this time for real) actually make a difference in our lives. A tool for "disintermediation," to "level the playing field" and "skip the middle man." A tool so compelling we could almost envision a harmonious and equitable global community. Every resource would be available for free in every hut, home and hovel. No one left out, no one excluded. Our "internet" was to be the Great Equalizer.

But our equalizer is only the internet, after all. We sift through 500 emails and only three need answering. Our public discussion board must have filters to lock out spammers. Our choice of creative tools is reduced to one or two pieces of unintuitive and expensive bloatware (courtesy the "purchase and bury" method of corporate competitive tech advantage). When humor and humility are replaced by hurricanes of hype, it becomes adamantly clear our great new tool is only a hammer. In the right hands it can build something worthwhile.

All this doesn't mean your plucky FezGuys have gone AWOL. We know where the useful things are. We know what's a good common-sense tool and what's a piece of crap. And we won't be shy about saying what we think, either. In spite of the megalo-maniacal hubris surrounding us like leech-infested mud flats, us Fezzes will continue to seek out and promote Good Hearts and Fair Minds, and to expose deceit and foolish behavior.


In the early 90s (when Netscape was still Mosaic) you'd be hard-pressed to find a company with a huge internet budget, yet couldn't toss a floppy without hitting one of many freely available programs coded by good-meaning individuals. Many of these folks found themselves scrambling to turn a good idea into a way to support themselves as the world became enamored with the Web. Within the space of a few weeks, it seemed, the mad rush for VC millions became commonplace. Most of these companies died; hypnotized by unrealistic expectations or just lacking a workable business model. Some were swallowed whole by larger companies. A few carved out a successful niche and are now jettisoning employees to survive lean times. Many companies' stories are the chaotic story of the internet itself. To celebrate column #60 (the FezGuys Five Year Anniversary!) we're diving into one example of that chaos - a company that began as merely a neat and useful idea, a "small tool done good," that saw big money acquisition and later negative public opinion, a company using litigation to stay alive and still grinding on, hoping to survive. It's a typical story of the internet.

Created by Steve Scherf and Ti Kan in California in 1995 as a "fun and easy way to identify and categorize their CDs" CDDB began as a casual tool. You could instantly identify song, artist and album info for CDs played in your computer from an online database. Encoding and player apps would map your CDs' TOC (table of contents) to the actual track names in the database and, if someone had entered and submitted the CD's information, the track titles (plus other info) would appear in the player window. Nice, simple and free. Anyone (individuals, label employees, the folks at CDDB) and everyone supplied the information to create the CDDB database. We got used to updating our own CD releases (we're talking extreme indie stuff here) to CDDB and also playing our own one-off CDs to see what the CDDB database would make of it. Sometimes it could be very amusing. We'd put in a recent mix of some song we're working on and CDDB would tell us it was Tori Amos and give us album tracks in order. No dice, but amusing, nonetheless. A simple idea isn't always so simple to pull off.

From the beginning the founders thought it would be nice to make a living from something they had created and that others found useful. When CDDB started running out of their own resources, they moved on, borrowing bandwidth and diskspace from appreciative folks online in true grassroots fashion. The code that ran CDDB (programmed by Scherf) was released as open-source under the GNU Public License (GPL), but Scherf retained the copyright to his work (GPL gives the right to modify and use the software as long as the source code of the new application is, in turn, also released under the GPL). At this time, the founders also included the complete current database along with the code so anyone could easily (and helpfully) put up their own mirror. A lot of people were now using CDDB and it was clearly time to get the business in order. Though CDDB code and use were still free, it was also clear that (the GPL notwithstanding) the service could be stolen, modified and co-opted. They stopped including the full database in new version source code distribution in February, 1998.

An investor named Scott Jones purchased CDDB in July 1998 and incorporated it into the Escient company name. In November, 1998, Jones also acquired Ion (a CD-ROM company), that controlled technology patents CDDB quietly made use of, making everything legally clean. Escient/CDDB had money to pay actual employees. It was time to start licensing their system. Still wanting to keep things free for the majority share of small users, the founders-turned-company-men opted for a single license allowing free use to anyone distributing 250,000 units or less. This allowed independently distributed shareware/freeware application development to continue using CDDB while having the new big kids (RealNetworks, Winamp, Audiocatalyst, etc) pay for each copy of their software sold. At this point everything is all wine and roses: a useful product, free software, developer support, larger users paying an equitable fee, and people who love what they do making a living doing it.

In 1999, a site called appears. Built from Scherf's code and an old 1997 database, the site is possibly a response to Escient's pronouncement of ownership of the CDDB database (created and shared by all) and now requiring licenses for its use. Several months later FreeDB's database converts from the old 1997 CDDB database version to a current version, leaked from one of the authorized mirror sites. All anyone can seem to find out about FreeDB is that it's operated primarily out of Germany. How to handle this conundrum? Yes, FreeDB is clearly using a copy of the database proven to be owned by CDDB. But is it legal (and ethical) to copyright a database primarily created by the public? Many open-source supporters (the Electronic Frontier Foundation included) say absolutely not. For now (along with other companies like - whose real assets are provided by the public at large) Escient holds the upper hand.

Escient, renamed Gracenote in July, 2000, currently represents itself as doing "everything it can to support continued use by the open-source community." Recent licenses (April 2001) include a revised commercial version providing free use for under 75,000 units and a non-commercial version allowing free use provided no money is made from the related product. Commercial license fees are a mild six cents per user per year (far cheaper than another internet audio technology company - MP3Pro - demands of its licensees).

According to Gracenote CEO David Hyman, response has been strong. Over 2,200 non-commercial licenses were granted in a recent four month period and about 100 a week continue to come in. But there's that little exclusivity clause that's bugging a lot of people. The open-source community has withheld the Gold Star rating for Gracenote that it once gave to CDDB. A tiny, yet oh-so-important detail in the new non-commercial license states: "you use the Gracenote CDDB Database and Gracenote CDDB Client in your Licensed Application as your sole source of Data from the Internet(sic) that is based on reading the TOC Data of any CD, Enhanced CD, CD-ROM, Tag ID or DVD media with your Licensed Application." Simply put: it's their way or the highway. This rather harsh stance rubs open-source supporters the wrong way. The open-source community is worth listening to. They often lead the way to popular adoption of technologies and applications. Naturally Gracenote wants this. But Scherf gives the impression the exclusivity clause is the result of lawyers "perhaps not quite getting things quite right [in trying to prevent simultaneous use of the FreeDB service]." Hyman goes further, stating the company "should seriously consider re-evaluating that clause."

But Scherf appears ambivalent about FreeDB's existence, saying that: "competition is good...the more CD database systems out there, the better." Although he felt "betrayed" when later versions of the database were leaked to FreeDB, Scherf thinks it will work itself out. In his mind; if FreeDB continues to use the Gracenote database or violates their software patents things will move to litigation. If FreeDB switches to its own data and doesn't infringe on patents, then it's simply healthy competition. FreeDB and others in the field know that Gracenote isn't shy about going to court. Digital music player manufacturer Roxio Inc. was sued by Gracenote earlier this year for breach of contract after letting their license expire and moving to a new CD database system that Gracenote claimed violated their patents. Roxio responded with a counter-suit and things have been quiet for the past month or so. Just another spasm in the internet audio technology story.

What's all this to us? Obviously, if you start something (and want to make money from it) protect your work cleanly and completely from the start. Sound familiar, songwriters?

Gracenote's future is focused on granting licenses and moving their database system into home consumer components. Their entire database of over 11 million song titles can be compressed (lots of song titles include "baby" and "love") down to 300MB (half a single CD-ROM). This "Gracenote Data Disc" is set for release in January, 2002, and the data is planned to be integrated into multi-disc CD carousels, with regular updating via mail. Today in Japan there are people with very expensive car CD changers using voice recognition systems integrated with the Gracenote CD database. We can visualize drivers shouting song names they want to hear while careening around highways. Scherf says he hopes to see co-operation between his and other existing CD database companies (, MUZE and Microsoft's new service).

Naturally, the FezGuys think that it only makes sense for everyone to work together. Open standards in technological infrastructure are good. Open standards lead to things that work for the consumer and aren't co-opted by constipated corporate bureacracies. Sure Gracenote made a few judgement errors in handling their licensing and paid dearly in public opinion. We hope they stand behind their words. Hey, FreeDB could end up a Gracenote licensee. Everybody wins, right?

The FezGuys welcome your comments. <>



About the authors:

Jon Luini is a working technophile, a musician (bass player/singer) with full-blown facility and extensive experience on the Web and no free time. He is a co-founder of IUMA and MediaCast, co-creator of Addicted To Noise, and runs an Internet and music consulting and technology company, Chime Interactive (formerly Evolve Internet Solutions). <>

Allen Whitman is a working musician (bass player/singer/producer) with a keen, real-world interest in the practical use of the Web. Music credits include: The Mermen, "Brine-The Antisurf Soundtrack, biL, Deep Field South, Doormouse, Delectric and Drizzoletto. He has written for the San Francisco Examiner, Wired, EQ, Revolution, Yahoo Internet Life, Prosound News, Surround Professional, Replication News and others. <>

©1996-2003 The FezGuys™