Music + Technology + Random Nonsense from the Music Industry by Ethan Kaplan, VP Product, Live Nation

What’s Next for Artist Websites

When I started at WBR five years ago, I entered a department where the creation of a website was as curated, and isolated as creating a billboard. It was meant to represent a moment and an experience in that moment, not anything for someone to inhabit. We had sites where the navigation elements were floating balloons for instance. The extent of “information” was two feeds for news and tour that were piped in from a very rudimentary content management system. Home grown, of course.

I made my first band website in 1996, for R.E.M. It was unofficial of course, as at that time R.E.M. had no website. The original tag line of the site was “Come in, login, and live.” I operated under the presuposition that to make websites more than just static pages on screen, they needed to bridge the gaps between screens by creating the notion of shared experiences. The term I always use to describe this is commorancy, which is a legal term describing a place of dwelling. A commorant is the inhabitant.

To create this notion, I organized the site as a hierarchal tree that consisted of “rooms.” Each forum was a room, threads were rooms within them, etc. This extended to the entire site, including news pages, lyrics etc. Based on the tree structure, every page gave indication to user presence, such as “There are 10 people here with you and 15 people next to you” and also flags on every topic in a forum listing showing how many people are in the conference.

The goal here was to abut the simulation of physical presence with virtual, and change a site from being sets of pages into hierarchal notions of space. The experiment was a success, enough so that it was nice to see similar features roll into discussion board systems in later versions. Now its something that users of my site take for granted.

The Internet today is vastly different than it was when I did this “rooms” system on The concept of shared spaces, and shared references to virtual experiences is something that is pervasive across sites. Facebook chat is an example of this, transforming Facebook from a series of (often slow loading) pages into a communal experience with no spatial referents other than presence. The key difference with Facebook is that space is being ascribed to the entire site rather than pages. This is mostly caused by the notion of “pages” evolving into the notion of “screens” with the advent of AJAX and asynchronous page loading.

And of course, realtime has become a part of this. While “instant” is a buzzword and mostly a way to show increases in data processing, more interesting are sites like Quora and Twitter and their use of long-polling/comet/websockets and non-blocking servers like Node.js and Tornado to provide realtime multiclient integration between the server and client. The client in a sense is becoming a thin-client processing engine off a server, creating a hub/spoke computing model between interaction and server resident data. Pretty cool.

And yet, our artist sites are stuck back in 2005.

Sure, they have evolved. They are connected to the social graphs on various sites, and often times are leafs in open graph structures for Facebook. But really, they are two or three column textual representations of whats going on for an artist. Ultimately they are newspapers with basic level interactivity, not too different from the “pages as rooms” mentality that I created back in 2001. Artist sites provide news, photos, music and a modicum of community. Community in this case is the ability to give feedback and content back to the site and its “official” content.

I wonder though, could there be more? What is it that draws people to an artist site, much less create them? What is the experience supposed to engender?

I created for R.E.M. because I wanted a place to feel “safe” being a fan of the band. I didn’t want to go on a board where I’d be judged, ridiculed or not listened to. I wanted a home as it were, filled with like minded friends who I could talk to about R.E.M. or anything else. We had a common referent, everything else was additional.

Fan websites still exist, and often serve the same purpose. The difference is that in the last five years, the walls that bound artists and their online representations from their fans have come down, to the point where the artist sites look like the fan sites. Their aim is a bit more complicated than to provide a safe place. They are also in it to provide the artist revenue through upsell’s of the artist. Fanclubs, merchandise, direct-to-fan programs, etc.

But ultimately, both fansites and artist sites are striving to create an online place rooted around being a fan.

Now, in 2010, with so much technology at our disposal to make experiences for the web that transcend the concept of a “site,” what does this space become? How does fanatacism, consumption and passion for music manifest itself in a 2010 era artist website?

Here are some of my loose ideas.

Realtime Aggregation

This really isn’t a new feature, as it was done in sites like Friendfeed. However, as applied to artist websites, it could serve to bridge relative activity during quiet periods. Artists operate in a “light/dark” model in terms of presence out there. Usually between releases, they are “dark” but the fans lives are certainly not. If you consider all activity that goes on in a fan’s life that has a referent point to that artist, it is substantial, and unique. It is not like Friendfeed, where the aggregate activity was a reflection of individual life. Instead this is aggregate activity as a reflection of being a fan.

I’m a huge R.E.M. fan (obviously), so if I was to filter my daily activity online and off based on that reference point I’d end up with something like the following:

- 10:10 AM – played a song from Out of Time

- 10:15 AM – played a song from New Adventures in Hi-Fi

- 11:00 AM – logged into

- 1:00PM – played a video on YouTube

- 1:30PM – posted on


On a day when R.E.M. was playing a show, it could also include me updating the setlist on their artist app, listening to tracks from the new record, checking in at the show, checking in with friends, etc.

The power here is not the singular effect of one fan, but the cumulative of all fans. It’d be great to go to the R.E.M. site, see what others are playing, watching and doing. If a fan finds a news story about the band and tags it, it’d show up there. If they quote a lyric, it’d show up there. Or a video just uploaded to YouTube. Think the REM tour site, amplified.

As a fan, I would love to open the website for the National, or R.E.M. or any other band, and see immediately that things were happening within that band’s ecosystem, whether or not the band is currently “active.” Enjoyment of an artist shouldn’t be singular, and the site should be the hub and router for the transmission of collective experience.

Tap into the potential of curation

The currency of fanaticism isn’t in the fact that we like a band. That’s a given. What is a currency among the die hards is the depth to which their fanaticism has manifested itself through externalization out of one’s self. I might be the world’s biggest R.E.M. fan (maybe), but that fact can’t be something that belongs to myself, otherwise it is only a self described trait. What sets me apart, or sets other fans apart is the degree to which they curate things around them that manifest their fandom.

Curation of course is a huge culture online. Among music sites, especially those for bands, the curatorial tendencies are more a matter of social currency than just accumulation. There are sites dedicated to curation (dimeadozen) and a lot of fan sites have this notion. has it in a few forums, and used to also have its own torrent tracker for live shows.

As a fan, curation is huge for me. My dream scenario would be an artist site that had a very innovative data browser, letting me browse through tons of content and self-organize a taxonomy based on what I own, want, like or want to share. Everything the band does should be dumped into the corpus: live shows, interviews, EPK’s, press photos. For R.E.M., there are literally warehouses of this stuff. I know at WBR that we have more stuff made than we could ever use.

This is made a lot easier as of late with data storage systems like MongoDB, as from experience, designing a schema for things like this was a mess. The ultimate for a fan, I think is to create a collection of artifacts around themselves as a method of defining themselves as a fan. This transcends the notion of a “profile” into a living and breathing record of passion and interest.
Location, location, location

Agents of the Free.png

Given that most phones now have GPS internally, it seems obvious that connecting location to the artist site experience is a natural outgrowth. However, not many have done so. The natural use of location of course is with live shows, but I think that minimizes its potential impact. Location could transcend just live shows to become a real world historical record of an artists career. Not only places they will play, but places they did play. Places important to the band. The window where an album cover was shot, or what other fans are in the area.

Certain bands lend themselves to this more than others, Radiohead and R.E.M. are the easiest examples. But every band has toured, travelled and has historical roots scattered throughout the world.

Live shows present other opportunities. Back in 2002 I did a project where I self organized meetups at shows based on social activity on Murmurs (left). Similar concepts could exist, using location awareness and context as a way of creating serendipitous encounters between fans, the band or the environment of a show. Ultimately, location awareness must serve to strengthen the identity of a fan on the site and elsewhere.


I’ve been toying a lot with Mongo DB recently for various projects. It presents something of a dream to me in the sense that the biggest challenge working with data around a band is trying to normalize the table structures to support the data. An example being setlists from shows. Insanely insane, and you end up with kind of nuts queries. Songs could be in the encore or in the main set, or in a case like Phish, the 50th set. Multiple versions of songs could be played (R.E.M. has two versions of the song Drive). Cover songs have to be accounted for, but not pollute the discography, so you can’t click through to necessarily a “Song” page for that.

When you add in discography, it gets even more crazy. At WMG, the database to store repertoire information needs many sheets of paper to print the table map by necessity.

A dream of mine has been, for R.E.M. through at least, to have the defacto datastore of their career. Every song, album, show, video, press clipping, news article, statistic, all in one store. Mongo makes this possible to a large degree. I’ve seen some really good experiments as of late with Mongo and realtime web frameworks to power the datamining, notably some work out of Development Seed, who are more commonly known as a Drupal shop.

Imagine a system like this for amorphous, entropic artist data. Imagine being able to browse data to get specific on all shows a band has played where a song was played in the main set, using graphing libraries like Protovis. Or imagine a data browser where tweets mentioning the band flow in, with automatic contextualization into a rich and growing data store to link to song info, etc.

Those are some loose ideas. I’m interested in any feedback as to what others think, and the plan is during the next year to do some experiments around these difference concepts.

If you enjoyed this post, please consider leaving a comment or subscribing to the RSS feed to have future articles delivered to your feed reader.

3 Responses

  1. Pierre says:

    Best post ever Ian. I agree with everything you said here.

    I’d like to throw out another point to ponder. A person’s favorite band is like a lightning rod for their taste. If a person is a huge fan of that band they probably like similar bands, and also the classic bands that influence them. They also probably have similar style tastes.. they may shop at similar stores as other fans, and lifestyle tastes as they probably like similar scenes and travel destinations.

    All of that creates a huge opportunity to become leaders of the culture that surrounds them. is an example of how that works. They created a site based around 50 Cent and his hardcore fans, but it isn’t just about 50, it’s about what he presumably likes and endorses. That makes the site sticky and gives the opportunity for a lot of revenue while always making sure people come back often enough to promote his own projects. I think does the same. They become destination sites that always, most importantly, stay true to the artist and their tastes.

    A band’s site doesn’t always have to be as commercial as those two, but bands need to stop being so insular and start appreciating their role/opportunity online to be cultural ambassadors instead of just using their sites for tour updates and amazon/itunes landing pages.

  2. Great post, Ethan. I really like your ideas on the potential of curation.

    We currently have built out two tour archives for artist websites that we are always striving to make better. Fans can see the entire touring history of a band, then go in depth on the page with media from the show (audio, photo, video) and connect with other fans that were at the show. There are still a number of possibilities to develop towards the full potential of this, and we are still working on pushing the limit on it.

    Here’s an example of one we built out for Matisyahu:

  3. You are right on the money about the data being created. I think there is a real opportunity here for artists to create semantic data. Being from the Drupal world a long standing dream of mine is a standard set of feature modules to create a database of songs, releases and concert performances. Mark this up with RDF as per This would be a treasure trove of information of historical/geek value.

    With technologies such as MongoDB, RDF and Drupal maturing the timing feels right for something like this. I know that here in London exciting “hackathon” work is being done with data made available by musicbrainz, BBC and others. It would be great for WBR to be a leader in this space, especially if the configuration was released as open source for any band or fan site to use.

    A few other links of interest in this space:, (I WISH this data was available in a structured form). There are many more.