New model for music playlists

We were talking in #joiito about wanting to play music for each other, but not being able to just post urls to our mp3's for fear of vexatious prosecution.

So here's my initial straw man suggestion for how to do this.

1. create a new playist format that refers to songs by a canonical naming scheme - MusicBrainz has a good starting point. These get put in URI's that look something like:

songid:66fea472-0093-4764-a574-7ef87ade4433?artistid=d1601842-8052-4ac1-81ef-67e8259250dd&artist=Shannon%20Campbell%20and%20Scott%20Andrew%20LePera&title=Nothing%20New

How does one resolve this URI?

Implement an app that tries several alternatives:
1. Is it in the local music collection?
2. Is it available on a label or artist website/bittorrent?
3. Is a promotional extract up on Amazon or iTunes store?
4. allow further plugins

In each case the name/title are used for an initial match, and the sing identification process defined by MusicBrainz used to be sure we have a real match.

DRM 'Industry' hates freedom

Bill Rosenblatt :

The DRM industry should hope that the ISPs lose. If they do, then that means that ISPs do not have a free pass around liability for their treatment of content, and therefore that they should be proactively offering DRM services, both to protect themselves legally and as a source of incremental revenue from copyright owners. If the ISPs continue to win these cases, then a major segment of the digital media value chain that could support DRM technologies and services may disappear.

Remarkably honest reporting by DRM Watch here. The DRM 'Industry' has a vested interest in destroying the free transmission of information over the public internet in favour of closed, restricted networks.

Deeper pixels needed

Glenn Reynolds, assorted Slashdotters and Peter Lewis at Fortune are speculating on whether Ansel Adams would have used digital cameras.

As I worked on the first official release of Ansel Adams images in digital form - the Ansel Adams screensaver - I feel somewhat qualified to pontificate on this.

What everyone is missing is that it is not a question of resolution, but of dynamic range.
With film the issue is blurred by questions of grain, but with digital the problem is which image format to use.

TIFF supports more than 8-bits per channel, but JPEG does not, nor do most computer displays. Most digital cameras only generate 8-bits per channel of dynamic range, and are still competing on resolution.
On the other hand, even the $30 scanners do 12-bits per channel these days, and 14-bits per channel (42-bit) are only about $50.

I expect that digital cameras will pick up on this soon, but we'll need a file-format change (and possibly a better transient storage device than flash memory) and new low-cost software to make it mainstream. Ironically, all the sophisticated exposure calculation that makes many digital cameras too slow to use for action snapshots could be reduced if they upped the dynamic range a few stops.

If you want to see what kinds of dynamic range digital photography is really capable of, have a look at the Hubble images, where every photon is carefully counted.

Tasty links

Down on the right is my del.icio.us linkblog. Its a bookmark replacement, and a place to find great links, as you can see other people's links and categories too.

HTTP diffs

Steve Jensen Jenson:

I came up with an idea on the train recently that might be useful here: If you're given the right bytes and told where they should go, you can construct the right document. HTTP does have semantics for giving you bytes and where they should go.

Steve, there is an existing HTML RFC for sending diffs that is perfectly suited to this - RFC3229. I mentioned this when it came out in Jan 2002.

The technical argument against DRM

Although the economic argument is more powerful - that DRM destroys value for customers and hence will be shunned by them - the technical argument is strong too.
This rests on one of the fundamental pillars of Computer Science - the Church Turing Thesis that states that any computer can emulate any other. When this is combined with the continual improvement in computing power available, it means we will always be able to run old software, or indeed protected software, by emulating the environment it runs within.
Simson Garfinkel describes how emulation saved the BBC Domesday Project, the authors of which I worked with at the BBC and the MMC.

"But that wasn't DRM" I hear the cry, "just obsolete hardware and data formats".

How about a systematic program that defeats the hardware protection for pay per use interactive experiences that works in a general enough way to encompass 25 years worth of hardware design?

It's called MAME and it has just been ported to the Nokia N-Gage cellphone/game gadget. It has emulators for various CPUs (and graphics and sound chips) to run the code directly from the original game ROMs - they look and feel just like the real thing
If Nokia are smart they will license this and the games and use it to promote the gadget - this company has licensed Atari ROMs for sale. After all, those 80s games are smaller than most MMS photos that get sent, and they're lots more fun than ringtones.

I hope Ed Felten and maybe can explain this to the assembled lawyers at the Berkman conference today. Most of them seem to like on compulsory licensing schemes.
I wish I had been able to take the chance offered to join them and present mediAgora to them. I look forward to reading the blogging of the event.

Here's a cartoon I made with the wonderfully silly Bayeux Tapestry Construction Kit

99% perspiration

Tim Oren takes me to task for saying Dave invented outlining. Now I could go on about the long history of simultaneous or convergent invention in history, but I won't, I'll just say this.

Tim - you're in VC. You should know the difference between a demo and a product.

Commodities are a good thing

Doc:

The stupidest conceit of the software business is that commodities are bad.

If it weren't for commodities, we wouldn't have civilization. Or food.

There's plenty of money to be made in - and on (or choose any other preposition) - commodities. You just have to think smart about the stupid stuff. Is it that hard?

Commodities are great - to paraphrase something Clayton Christensen said - once your business has become commoditized it is simple enough that you can hire some MBAs to run it for you.

Jessica Litman's music idea

Comment on Sharing and Stealing - Jessica Litman

It is a great essay until this paragraph::

The key to the opt-out mechanism I propose is the selection of a single digital file format or family of formats capable of conveying copyright management information as defined in section 1202 of the copyright act. The format will probably incorporate digital rights management capability because the people who will be using it will desire that feature, but there's no need for any copy-protection to be hack-proof, or even exceptionally durable. It should also be compatible with the current generation of digital playback devices, including CD players. I'll call the format *.drm for short.

This aspiration: "It should also be compatible with the current generation of digital playback devices, including CD players." is impossible. CD players play unencrypted, uncompressed digital audio. A drm'd format would require new players.

Hierarchies, webs and emergence

Last week Dave Sifry and I met up with Dave Winer and Steve Gillmor at Technorati to share ideas. We talked about the public resource that Dave W created in weblogs.com, about Technorati, and about Dave's new idea to help people categorize blog postings and the things they link to.
Dave W said: I feel we're at a turning point in the weblog world, either we're going to be like every other hierarchy that's ever been, with secret deals, lots of impediments to progress, eventual stagnation; or we're going to overcome that.

Dave thinks in hierarchies; whether this is because he invented outlining, or why he invented outlining I'm not sure. Along the way he added links into the picture, so his hierarchies can link to other nodes, or other hierarchies to get as complex as you like.

The conventional wisdom is that links beat out hierarchies - Google's link-centric approach beat out Yahoo's hierarchy-centric approach (the HO in Yahoo stood for Hierarchically Oriented).

However, another way of looking at it is top-down versus bottom-up - central design versus emergence.

Dave W wants to build a bottom-up emergent taxonomy, using open debate and open standards.

Steve Gillmor is saying something similar about how we can grow new things.

I have a couple of ideas that I need to write up as spec proposals to try to start such discussions - one about 'vote links', one a new bit of metadata for feeds saying whether they are complete or not.

Now that's data mining

Mining company publishes geology data and offers prizes for good spots to dig

What am I missing?

One fascinating thing about Technorati is how many blogs are showing up in languages other than English. One of the highest on the the Top 100 is an Iranian blog, which is a great thing.

However, I look at blogs like this and feel like Ginger in Gary Larson's classic What Dogs Hear.
'squiggle squiggle squiggle Blog squiggle squiggle squiggle Permalink'

iTunes Security blather

Tracey Meyers wins the most errors in the shortest amount of text award this week:
"So sue me" stated young Norwegian hacker Jon Lech Johansen last week after he posted a program to crack iTunes air-tight security.

Johansen posted the program called "QTFairUse", with the previously mentioned quote, on his own website. The free software has the uncanny ability to sidestep iTunes anti-copying software, MPEG-4 Advanced Audio Coding that once installed illegally views protected music files in QuickTime without paying a fee or royalty.

1) 'air-tight' security is remarkably meaningless. In fact, oscillating air is the primary security 'hole' in iTunes, followed closely by the built-in CD-burning code.
2) 'uncanny' is an odd thing to say about source code. It's pretty clear what he is doing if you can read C, and if you can't why not talk to someone who can first?
3) MPEG-4 Advanced Audio Coding is not anti-copying software, it is a compression format. The anti-copying stuff is called FairPlay.
4) That last sentence manages to call AAC illegal.
5) You can already listen to (not view, unless you count the tiny spectrum display) the protected music files in QT Player, as long as you know the username/password.
6) You have to have already paid for the music for this to work. All it does it replace the 'burn to CD' option with a less convenient way to extract the audio (It only works on one song at a time as it is played).
7) It's highly arguable that this is illegal, given that transferring music to other forms is explicitly legal - indeed were it not, the CD to MP3/AAC part of iTunes would be illegal. There may be a DMCA case on circumvention grounds here, but it isn't a good one, given the fact that iTunes will let you make CDs from the same file. In any case the DMCA is not Norwegian law.

Still, I'm surprised it took this long for someone to start the arms race in this instance. I hope my old friends in Apple don't get dragged into it.

Donate to Rupert - the True Survivor

Rosie and Andrew were shocked by the betrayal on Survivor last week when Rupert was thrown off that they asked me to set up a website for him. Here it is:
Donate to Rupert - the True Survivor
Andrew drew the picture.
He made a t-shirt too.

Co-opting the Future By�John C. Mahler

Trolls, are all the rage in some quarters. We're told that trolls will evolve into a unique source of information and are sure to become the future of journalism. Well, hardly. Two things are happening to prevent such a future: The first is wholesale abandonment of troll sites, and the second is the casual co-opting of the troll universe by Big Media.

Let's start with abandoned trolls. Thanks to busy debunkers trolling has got harder.

The most obvious reason for abandonment is simple boredom. Writing is tiresome. Why anyone would do it voluntarily on a troll mystifies a lot of professional writers. This is compounded by a lack of feedback, positive or otherwise. Perseus thinks that most trolls have an audience of about 12 readers. Leaflets posted on the corkboard at Albertsons attract a larger readership than many trolls. Some people must feel the futility.

The problem is further compounded by professional writers who promote trolling, with the thought that they are increasing their own readership. It's no coincidence that the most-read trolls are created by professional writers. They have essentially suckered thousands of newbies, mavens, and just plain folk into trolling, solely to get return links in the form of the blogrolls and citations. This is, in fact, a remarkably slick grassroots marketing scheme that is in many ways awesome, albeit insincere.

Unfortunately, at some point, people will realize they've been used. This will happen sooner rather than later, since many mainstream publishers now see the opportunity for exploitation. Thus you find professionally written and edited faux trolls appearing on MSNBC's site, the Washington Post site, and elsewhere. This seems to be where trolling is headed�Big Media. So much for the independent thinking and reporting that are supposed to earmark blog journalism.


So now we have the emergence of the professional troll working for large media conglomerates and spewing the same measured news and opinions we've always had�except for fake edginess, which suggests some sort of independent, counterculture, free-thinking observers. But who signs the checks? The faux troll will replace the old personality columns that were once the rage in newspaperdom. Can you spell retro? These are not the hard-hitting independent voices we were promised. They are just a new breed of columnist with a gimmick and a stern corporate editor.


By� John C. Mahler

Previously by this author: Deconstructing the Troll