Friday, April 15, 2011

Metadata, Microformats and Tomahawk... Oh My!

Yesterday I wrote a post over on the Tomahawk Blog showing how with just a little html markup, any page could be made to work with Tomahawklet.

Today, I wanted to just do a little bit more showing how you could make song info in sentences/paragraphs playable. It's actually the same exact thing, but the idea I am toying with is how bloggers can add more value to their context and keep their writing more fluid and natural if they didn't have to jam a big flash widget (e.g. YouTube, MP3 player, etc.) in the middle of all their paragraphs.

Please excuse my inane drivel, this is why we need real bloggers and reviews to provide better context than my example below... (hint: if you have dragged the Tomahawklet button to your browser toolbar, click it now)
I'm digging the new song "Lippy Kids" by Elbow. It's pretty mellow, but still sticks to your ribs. You don't like that?
OK, maybe give Grouplove's "Colours" a spin. A nice upbeat ditty, no?

Too much? Alright maybe we can bring it back down a bit with
Sadie from Fences.
I'm thinking a Wordpress widget that let users just highlight/indicate what word is an artist name, and which track belongs to it, could be pretty cool....


microformats

(Just click "close" and the shade will close but continue to play.  To reopen, just click the bookmarklet button in your browser toolbar again.)


Related articles
Enhanced by Zemanta
Posted by at 1 comment:
Labels:

Monday, March 28, 2011

Why I think Tomahawk is the future of music consumption...

If you follow me on Twitter then you know I have been very vocal lately about the new media player (nay, music player) called Tomahawk. Perhaps some of you hopped over to the site, looked at the screenshot and thought... "oh, yet another spreadsheet for music files". If you did, I urge you to look further as it is so much more.




Instead of me explaining the features (you can get better descriptions of those at www.gettomahawk.com), I'm going to talk about some of the value propositions that I hope to see Tomahawk deliver. Specifically, a day in which music fans can easily share their tastes across a range of music service providers. Today, if you subscribe to Rhapsody and I get my music from eMusic, we can't easily share our music. And no... I'm not talking about sharing your music files, I'm talking about your music taste and discoveries.


Today we live in a world of seemingly endless number of content silos, your playlists and taste data largely locked up within their walls with no easy way to get them out or share them. Twitter and Facebook have, in many ways, made this specific problem much worse. For example, let's say you post a link to your favorite new song. Unless I subscribe to the same service (or you have linked to a free source) then I have to click the link, see what the song was (maybe listen to a snippet), then go and search for it from whatever source/service/store I use. Not really the poster child for an easy user experience. The increasing fragmentation of the music provider market means that every day my Twitter stream is filled with virtual spam for music services I don't use (and that's saying something because I use many). I don't blame the people I follow, they just want to tell their followers about some great new discovery, but the link is largely noise as it provides no value to me - unless I establish a relationship with whatever source/service/store they use.  In my mind, this is one of the main contributors to YouTube (and MP3 links) becoming the defacto music sharing standard. Yes, they are free, but more importantly they "just play".


Last.fm (which I love), is the closest thing to a music taste translation layer we have today, but it doesn't solve the problem in the way Tomahawk has the potential to.  While I can get loads of great data around other peoples' (as well as my own) tastes, that data is not easily actionable.  I can visit their profile pages, or subscribe to an RSS feed of a friends' data, but then I'm stuck having to manually resolve that information against my music sources.  Why can't I subscribe to this information as a dynamically updated playlist in my media player, and have each of those tracks automatically resolved against all of the content I have the rights and ability to access?  


This brings me to my next pain point.  I have the rights, and ability, to access content from a large number of sources - yet no single interface that lets me easily do so.  Content from my home computer, my laptop, streaming tracks from my  Ex.fm library, SoundCloud, Topspin widgets, band sites, label sites, music blogs.  The list is endless.... and unfortunately so are the required number of interfaces one has to use.  Not to mention that many of those web-interfaces provide (in my opinion) a suboptimal user experience.  

  • "What tab is that music coming from?!"  
  • "Bah, I accidentally navigated away from the page/closed by browser and interrupted playback.

Now imagine a world where we share music metadata, and the logic of how that audio gets rendered is determined by the consumer of the data, not the sharer.  Under the hood that user can can add plug-ins/content resolvers for all of the music sources that they have rights to (local machine, remote machine, subscription service, web sources, etc.).  Spotify users sharing with iTunes users sharing with Rhapsody users, sharing with Mog users.  Tastes and curation freely flowing. Music discovery propagates like never before.


OK, so now we are all able to speak the same raw musical language (metadata), but what about the context?  But wait... (as they say on bad TV commercials)... there is more!  Tomahawk has a web server, API (see the awesome Playdar for more info), and authentication mechanism built-in.  Now bloggers, writers, curators, reviewers and more can provide context around music without having a way to provide the files/streams too (or go through the legal/licensing gauntlet required to do so).  Instead of providing links to a half dozen music providers, or a bunch of MP3 links, or embedded YouTube videos - these curators can rely on you to bring your (accessible) content to their context.  They can just provide metadata that Tomahawk can see, and use, to go out and automatically fetch the streams in background and play back to you (through their websites).   


Not only does this benefit the consumers and the curators, but also the music subscription providers.  Why should they be incuring the royalty obligations to stream me a song that I already own?  As I have worked 3 music subscription services myself, I can - without hesitation - tell you that every penny saved is another penny closer to a having a sustainable business model.  With Tomahawk's cascading content resolving framework, if you already have a copy of the song it will play that one instead of unnecessarily streaming it.  When it comes to the next song in queue, it independently finds the best source for that one and seamlessly, and invisibly, handles the transition between the two.  As a consumer, you no longer have to know or care where the music is - you get to just focus on what you want to listen to.  For those that have heard me on this soapbox before, they know that I like to call this "taking the work out of play".


There are a lot of previously impossible user experiences that this content resolution approach enables.  I have tried to touch on some of them, and hopefully I haven't confused the hell out of you.  As with many innovative technologies it is sometimes hard to convey the value until you try it yourself.   But, in the meantime maybe these screencasts can help:




The good news is that Tomahawk is totally free and open-source.  So not only can you use it, you can directly participate in the realization of the value the platform can deliver.  I hope you enjoy it, I know I sure am.






Related articles
Enhanced by Zemanta
Posted by at 22 comments:
Labels: ,

Tuesday, December 07, 2010

My Favorite Albums of 2010 (Updated)

There are still a couple of weeks to go, but taking a sneak peak at my favorites albums of 2010 (via @mark_reeder's hack).

Filterable Music Charts

The problem with always listening via shuffle all is that I don't agree with the order of my list, I may need to editorially override some of my implicit play behavior (as represented by this list below). Alas, no one seems to have reliable release date - making these hacks a bit hit-or-miss at times - but Mark has enabled you to manually set the release date on the albums it doesn't have data for (by setting the slider to start a Year 0).

What albums did I totally sleep on this year?




There were also a number of albums that came out last year that I missed. So in comparison, here are my most listened to albums over the past year that includes those one was released at some point in 2009.


Posted by at No comments:
Labels:

Friday, March 05, 2010

Poll: Does your ideal music service take the (primary) form of an application or a website?

Posted by at 4 comments:

Saturday, December 19, 2009

What I Did at Music Hackday...

Music Hackday BostonImage by teamstickergiant via Flickr

A few weeks ago I spent the weekend up in Boston at the first stateside "Music Hackday". It was a fantastic event all-around... the organization, the facilities, the people, the panels (even though I was on the panel about "The Future of Music") and the hacks.

For me, the highlight was really just spending time hanging out with other music tech geeks (metabrew, toby, jwheare, __lucas, lucas_gonze, dankantor, plamere, fascinated, theineke, bwhitman and many more)... bouncing ideas off of each other, and collaboratively building interesting things.

Why wife, somewhat famously, referred to it as like "a dungeons and dragons conference from music geeks".

Being in the same room with all of these other folks meant that I could not-so-subtly plant ideas for other people's hacks that would support some use cases that I am interested in. Toby Padilla appeased me (thanks Toby!) by adding some support to his great Playgrub app to support scraping artist and track metadata from Twitter so that the content could be resolved using Playdar (or more specifically by the content resolver plug-ins for Playdar that I have installed).

Not sure what I am talking about? Maybe it's easier just to watch. Want to try it yourself?

Just tweet:

★ artist name ♫ track name #playtapus (or any other hashtag/playlist name)


Then use the Playgrub bookmarklet and let Playdar to do the rest.

In short, we are just dealing with extracting and moving song metadata around and handing it off to Playdar (which runs on your local machine). Playdar takes it from there and figures out the best place to source the content from and fulfills the link/play action.


Definitely check out Playgrub (and all of the other sites it supports), as well as the other excellent Playdar-powered web apps like Playlick (by James Wheare), PlaydarTunes (by Dan Kantor) and iTunesAirBridge (by Lucas Hrabrovsky).

Read more about it...

http://musicmachinery.com/2009/12/01/boston-music-hack-day-is-in-the-can/

http://blog.hypem.com/2009/11/music-hack-day-boston-wrap-up/

http://notes.variogr.am/post/255966005/music-hack-day-boston

http://developer.yahoo.net/blog/archives/2009/11/music_hack_day_boston.html

http://www.grantcerny.com/blog/2009/11/22/music-hackday-boston/

http://savetherobot.wordpress.com/2009/11/25/boston-music-hackday-links-round-up/

Last, but not least, big thanks to Dave Haynes, Jon Pierce and Paul Lamere for making the whole thing a reality. When's the next one in the U.S.?! :-)



Related articles by Zemanta







Reblog this post [with Zemanta]
Posted by at No comments:
Labels: , , , ,
Subscribe to: Posts (Atom)