Ok so the post was about the Grateful Dead and I tried to make it abundantly clear I don't wanna talk about them too much but... c'mon just one more little post about Deadheads. Just a little old post about Deadbase, a series of pretty boring books. It's gonna be fun I promise just one more more teensy post about Deadheads. For now!

This is the first post in a series on the history of music metadata. It's also the first in a less formal series on text documents from early Internet communities: zines, NFO files, handbooks, etc.

Everyone thinks if you're going to heist a giant jewel or rob a bank you have to have all this fancy tech to see invisible laser beams or wear a Richard Nixon mask as a disguise or create an elaborate diversion.

I'm offering all of you aspiring felons out there a nice, simple trick that will save you time, money, and stress.

Just waltz straight up to the security guard and say, "Hey, I've been researching the history of music metadata from the late 1980s through the early 2010s and I'd love to tell you all about it." The typical person will instantaneously snap into a fugue state and become totally unaware of their surroundings for as long as you keep droning on. You could pluck their hairs off their head and they'd be perfectly bald by the time they come to.

Well, this past July I was lucky enough to get invited by my colleague Virginia Partridge to give a talk to the Center for Data Science summer program at my place of employment UMass Amherst, where people were actually really excited to hear me talk about the history of music metadata. Thank you Virginia, I had a lot of fun. For everyone in attendance, my apologies for just implying you're not a "typical person" because you enjoyed it, but hey, it's nice to feel special right?

The history of music metadata is really a history of how really dedicated music fans used the Internet to collaborate with each other in the name of something they loved. The cool thing about it is they came up with a bunch of really ingenious projects that seem Rube Goldberg-esque in retrospect.

Dead-adata

One such very early genius and ridiculous artifact is the Deadbase series: effectively the equivalent a sports stats book but for Grateful Dead shows. Each edition focuses on the preceding year, cataloguing all of those tours’ sets lists from as well as tons of metrics about the music played at that show: most common songs, least common songs, how long those songs were, what songs they most commonly followed in a set.

They also had photos, selected concert reviews, and some really great illustrations. I particularly love the cover of Deadbase '90 with the upright, symmetrical horse skeletons.

All of the early editions contained amazing computer-generated illustrations depicting blending computer and Grateful Dead iconography.

If metadata is "data about data" then it's not quite accurate to call the information in Deadbase books metadata, since it was really data about concerts. But I do believe these Deadheads presaged how community-powered metadata projects I'll discuss in future posts, such as MusicBrainz, would create gigantic stores of metadata by harnessing dedicated volunteers and strict norms. I'm not super confident saying Deadbase influenced those later projects, but we can see this as a kind of primordial version.

I think I've made a case for elaborate. But what of this Rube Goldberg machine aspect? Well, imagine if people today used the Internet to pool their collective intellect and instead of posting the fruits of their labor, printed the stuff out and mailed it around. It's actually not a bad idea! But it does demonstrate that at some moment in time, the Internet was not operating on the feedback loop logic that's so common to us today. In fact, it made more sense to use the Internet to create something people could hold in their hands and find on their bookshelves whenever they needed some information.

And it was wise. It's not possible as lay Internet user to find archives of discussions from 1980s about Grateful Dead set lists, but you can still find all that data about 1988 in the 1988 edition of Deadbase, which I have on my desk right now.

There also ended up being this now hilariously convoluted online interaction with Deadbase books. Comb archives of the rec.music.gdead newsgroup, and you'll find people for all intents and purposes commenting on Deadbase, remarking on how nice the photos are and wondering if the 5/6 "I Know You Rider" was really 12 minutes long or asserting "I've heard the tape and it is definietly NOT The Eleven, but instead King Solomon's Marbles/Stronger Than Dirt (which is in 7, not 11)." 

So basically, a lot of the data in Deadbase was collected with the help of the early Internet, then that data was put into print, then someone took the time out of his day to walk down to the campus computer lab, access Usenet through a terminal with an slow Internet connection, and show everyone that he knows more about both the Grateful Dead and time signatures than the people who compiled the book. To be clear, we made it easier, not harder, for those type of people to use the Internet.

Deadbase Origins

Let' start simple here. As I've described before, the Grateful Dead played a lot of shows and were constantly changing things up. They were improvising on stage constantly, which means basically no song was played the same way twice. Even a standard country tune like "Me and My Uncle" would see Phil Lesh doing some kind of interesting harmonic bass noodling that was always evolving.

As I've also described before, one of the earliest music-oriented activities people used digital network technology to collaborate on was figuring lyrics to live-only Dead songs by sending drafts back and forth over ARPANET.

That would be been the mid-’70s, and just a few years later, new proto-World Wide Web Internet technologies like BBS and Usenet were proliferating, and attracting Deadheads. Again, I've said this before, but lots of Deadheads were on a BBS called the WELL. And as I've also also said, the WELL had a kind of outsized place in the public imagination of the primordial Internet.

Case in point was the popular Usenet group rec.music.gdead. The best way to understand Usenet is that it was Reddit before the Web. You'd log onto it via a terminal, very likely at a university, and access the Usenet newsgroups that met your interests, specifically using a piece of software called a newsreader.

There were a lot of newsgroups and they were set up in a hierarchy structure. So with rec.music.gdead, rec was the top level category that included stuff like rec.skydiving and rec.video and, of course, rec.music. Then the rec.music category had all sorts of groups like rec.music.reggae and rec.music.celtic and, brace yourself, rec.music.gdead. So imagine if the way Reddit functioned is each time a group of people in a subreddit wanted dedicated discussion to a related topic, they just created a forking branch from that subreddit.

You had to pay to join the WELL but all you needed for Usenet was access to a terminal. Both had vibrant communities of Deadheads, and a lot of those Deadheads wanted information about Dead shows. A big part of the demand came from the tape trading community among fans. You could judge which concert recordings you wanted to hear based on setlists you saw online.

But the actual people to start the Deadbase projects were simply, to use a technical term, total nerds. Stu Nixon, in his introduction to Deadbase 50 (not the 50th Deadbase, but the Deadbase released on the 50th anniversary year of the band in 2015), recounts that he used to collect Grateful Dead setlists and just wanted to publish a book with them. He also recounts nervously asking Jerry Garcia if—I'm paraphrasing here—that made him some kind of compulsive weirdo and Jerry graciously told him that he used to keep lists of songs he saw at bluegrass shows he attended when he was younger.

Stu Nixon was an early Internet user, as was the band's bassist Phil Lesh, who frequented the WELL with the legitimately very cool username "kilowatt." Lesh recognized Nixon from online, and was instrumental in earning the Deadbase series the band's official blessing, as well as giving it access to the band's archivist Dick Latvala.

What's in the Deadbase

I bought a lot comprising six of these books off eBay for this research, covering the years 1988 through 1993 so that I could try to get a more authentic experience with them than just looking at scans on the Internet Archive. What struck me most is the creativity of the types of data catalogued.

The 1988 edition sets the template with three basic sections: data, visual matter, and concert reviews by fans.

The concert reviews are selected writings by some of the more influential contributors to the Deadbase project, where they write up a single show they saw. Those reviews are in chronological order, and while they don't review every Dead show from 1988, they review enough to give a reader a basic sense of how the music panned out over the year.

The visual matter comprise scans of ticket stubs with fun illustrations and high quality photos of the band performing, taken by Deadheads.

The data is of course the meat of the thing. There are of course the setlists themselves, but those only run form pages 1 through 8. What follows is a variety of statistics about songs and set structure that try to make sense of both the bigger picture of that year's performances as well as the fine-grained minutiae of what fans actually heard.

There's a "Songs Played" section which charts all songs played in alphabetical order alongside their average length, their maximum length, their minimum length, and their songwriters. There's a section with lengths for every song in a given set list. There'a a lists of total duration of each show, total duration of sets in each show, total duration of extended drum soloing. There's a list of the longest songs with the date they were played and how long they were played.

I think my favorite is a list of total time a song was played. For example, "Terrapin Station" was played for 3 hours, 50 minutes, and 15 seconds in aggregate, over the course of 20 performances while "Louie Louie" was played for a total of 25 minutes and four seconds across 5 performances. It makes a lot of sense to try to define a band in a given year by what they spent most of their timing playing.

If I were interested in delving a little deeper to that, I might look at "Louie Louie" and ask myself "wow how much time did they dedicate in total to letting Bob Weir play totally fucking miserable covers" then I could tally those covers and assess, with data, how much it would have felt like the Grateful Dead were punishing me personally in 1988. Considering they dedicated 31 minutes and 43 seconds to playing "Hey Jude" as an encore, we're already at minimum one hour.

But the Deadbase project changed pretty quickly in a fundamental way. In the 1989 edition, the authors added a "Feedback" section which reflected the results of a survey anyone could mail to a PO box in New Hampshire. From then on, a sizeable portion of the Deadbase books would be dedicated to understanding Deadhead culture in a given year. Early on, these surveys were more tightly focused on the Dead and the shows themselves, querying people's favorite venues, wishlists of cities they would play, favorite items to bring to Dead shows, that type of thing.

Fast forward to the 1993 edition, and the survey is dense with plenty of subject matter that has noting to do with the Dead. There's a chart for favorite and least favorite movies, a chart about specific feelings about Bill Clinton, a chart with weighted average ratings on statements like "David Koresh was solely responsible for the Waco tragedy." and "I was in favor of the Brady Bill." There's a "who's hot and who's not in 1993" chart, where Snoop Doggy Dogg ranks below the Menendez Brothers and Republicans, while Kramer ranks above Michael Jordan and Stephen Spielberg (and Jerry Seinfeld).

The Dead show-specific areas are more robust too, with really fun lists of favorite t shirts and stickers seen at Dead shows ("I've tripped and I can't get down") and biggest problems at Dead shows ("Not enough nudity", "seat scammers", "gate crashers", "not enough garbage cans," "vomit").

In Conclusion

I think there's something really funny about the Deadbase books, a kind of fundamental quandary. These are volumes created by incredibly passionate fans, which means the specific music they're so passionate about is some of the freest rock ever made (“free” as in free jazz), where the line between composition and improvisation is about as blurry as it gets. So many of the statistics in the book are only possible because the thing the Dead did live is so mutable, so committed to chance that the singers weren't always committed to remembering their own lyrics.

But these are books are an attempt to groom the shaggy dog, to quantify and qualify and analyze a body of work that seems much more interested in finding moments of the ineffable. I think the way I can personally identify with the impulse is there are a lot of times I've tried to take a photo of the moon on my phone and it comes out looking like a parody of my experience of the thing—a tiny glowing dot in a vast black sky, nothing like what it feels to be there in the moment but a sure reminder of humanity’s smallness.

But here's the thing: these aren't the work of a solitary Deadhead. They’re a collective endeavor, and my hunch about these books is that more than an attempt to rationalize the irrational, they're an attempt to make something new that was a lot more difficult to imagine before the advent of the Internet.

I end up seeing the Deadbase porject as really interesting artifact, preserving an attempt by early Internet enthusiasts to figure out what this network technology could be used for.

Sure, I think they cared about data itself, and thought it was really important to collect and present this data cleanly. But there isn't much evidence of that impulse existing among Deadheads prior to the 1980s, and I think that's really because people got on Usenet and the WELL to talk about the Dead and started wondering what else was possible because writing little concert reviews and trading tapes with each other.

What I’m listening to

This incredible Sugar Minott digital dancehall track:

The Hawkwind live album Space Ritual:

And lots of Rich Homie Quan. Big RIP to someone who made some of the best hooks of the century.