Frankenstein and Power Laws
Apr. 24th, 2004 02:29 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
imomusThere are two kinds of people in the world: those who think Frankenstein is a monster, and those who know that he is Victor Frankenstein, a scientist who creates a monster. Now, the second group might be 'right'; that's how Mary Shelley wrote the book. But the first group is far, far bigger, and might makes right, at least if 'usage' and 'sociology' count for anything. So Frankenstein has come, in Twix advertisements and mass culture generally, to mean the monster.
The Frankenstein syndrome was in action last Thursday, when Yahoo Japan News -- a stitched-together monster of a news service -- ran a story it picked up from CD Journal about my song Corkscrew King. I'd written, in my LiveJournal, that the song 'runs together Ken Shimura's Bakatono and Henna Ojisan roles and mixes the resulting Frankenstein's monster with the poet W.B. Yeats'. The Yahoo story, however, reported that 'Momus throws us a curveball by also offering the explanation that the song, "mixes Shimura Ken, Frankenstein, and W.B. Yeats". Average listeners are bound to be left scratching their heads.' Of course, that's not at all what I said. I just used 'Frankenstein's monster' as a metaphor for the transposition of two of Shimura's characters with elements from the biography of W.B. Yeats. Neither Dr Frankenstein nor his monster have any part in the dramatis personae of my song.
I can see why this kind of thing happens, though. A journalistic text is a kind of Frankenstein's monster too. The writer has to patch together a lot of assumptions and myths and understandings and try to make a coherent piece that's also interesting to read. This journalist was trying to reconcile what 'the eccentric British songwriter' meant with what 'the average listener' might be able to understand. In the end, he travestied both of us, concluding that average listeners should give the song some attention despite its quirks. For me the song was really only worth writing when I added Yeats and the theme of impotence; for the journalist, these elements almost ruined the song by making it confusing and obscure. The journalist lives in a world where, if enough people say Frankenstein is the monster, that's how you report it. I live in a world where, if Mary Shelley created Dr Frankenstein as a scientist, that's how he should remain. (Some philosopher somewhere has called these different views of the world the 'contract model' and the 'status model': contract people just care about agreement, status people are pedants who insist on some kind of 'truth'. We're always playing to the gallery.)
Occasionally these two worlds -- the public and the cogniscenti, contract and status -- meet. This is what happened on Thursday. My Corkscrew song was a classic novelty hit. It had a TV tie-in element. The reason bestsellers in record and bookshops tend to be TV tie-ins is clear: the public watches a lot of TV. Now, TV is a don't-care, don't-need-to-know kind of medium. Low cost, low commitment. Whatever's on. It passes the time. Whereas music and books need to be sifted and bought and committed to. The internet contains both modes. Although it's a pull medium, most people are pulling the same five websites each day.
Clay Shirky's Power Laws, Blogs, and Inequality is a fascinating exploration of this. He shows how equality of opportunity and equality of result are totally different things, and that even when there seems to be a level playing field -- when everyone can have a blog for free, for instance, accessible to the whole world -- Power Laws dictate that there will soon be a classic comet-shaped ratings curve, with a big spike at the left and a long tapering tail to the right. Attention is not distributed equally. Bloggers tend to settle into the same pattern of 'stars' and 'the rest' that we see in TV and other media. My dictum that 'in the future everyone will be famous for fifteen people' was not wrong, but only a partial truth: it needs to be amended to 'In the future everyone will be famous for fifteen people -- and there'll be a handful of stars who are world famous'. It's and-and, not either-or.
Have a look at the 'comet' shape you get when you chart LiveJournal bloggers by the number of Friends they list. The big spike represents the stars, and the long tail represents more democratic 'family and friends' kind of chatter:
What charts can't so easily show is the quality of attention those star bloggers -- the apparent attention hoggers -- are getting. One indication, though, might be duration stats. Sure, a lot of people might be hitting your page, but how long are they spending there? How deep are they going?
Here's an interesting thing I discovered after my big spike on Thursday. Although my website got more hits than it has ever had, the duration chart shows exactly the opposite shape: page view time dropped to the lowest recorded levels. People with a low level of commitment, web flaneurs, were hitting my pages (and sure, language also played a part: these were Japanese people confronted with an English-language site, although I like to think there's some interesting visual material there too) and leaving within seconds. They also weren't paying: I received a total of ten euros in donations from people with Japanese names. Yes, Paypal is not widely used in Japan. But it's also this: why pay for a song about Ken Shimura when your whole life you've been watching the real thing on TV for free?
My friend Craig Robinson found that the attention of a mass audience is a mixed blessing when a mention of his Minipops pixel popstar drawings on BBC Radio 1 brought hundreds of thousands of people to his website, Flip Flop Flyin. (Craig made the sleeve for my 2003 compilation Forbidden Software Timemachine.) Craig experienced the same sequence that I did: delight when someone told him about the mention, dismay to find the website basically undergoing a 'denial of service' attack, panic as we scrambled to save our servers and find alternative accommodation, more panic as we worked on getting something saleable into the window before the chance passed, and finally a return to normal. Wham bam thank you ma'am!
I'm ambivalent about this. I'm actually really happy distributing mp3s to a small, intelligent, generous and committed community of connoisseurs who share my values. Corkscrew King was never intended to be a 'hit' or to reach a wide public. Its 'success' has been as useless to me as the radio hit I had with The Hairstyle of the Devil back in... oh, never mind. But making content available on the web does always bring the possibility of this kind of swarming and with it the illusion of instant success.
An article in today's Guardian cautions that Apple's mainstream consumer success with the iPod may be the thing that kills the company. By neglecting their core market -- computer snobs and aesthetes like me -- and making their money from portable music hard drives, Apple may be chasing a fad. As the article points out, 'it is hard not to imagine that within a decade most people will be listening to music on the move via a player on their mobile phone.' Nobody will need a hard drive for music any more. And in the meantime, Apple will have lost some of its identity as a serious computer firm, and possibly slipped to even less than its current 1.7% of the US market. When your business depends on optimising your marginality (which means concentrating on 'the good difference'), nothing kills like success.
The Frankenstein syndrome was in action last Thursday, when Yahoo Japan News -- a stitched-together monster of a news service -- ran a story it picked up from CD Journal about my song Corkscrew King. I'd written, in my LiveJournal, that the song 'runs together Ken Shimura's Bakatono and Henna Ojisan roles and mixes the resulting Frankenstein's monster with the poet W.B. Yeats'. The Yahoo story, however, reported that 'Momus throws us a curveball by also offering the explanation that the song, "mixes Shimura Ken, Frankenstein, and W.B. Yeats". Average listeners are bound to be left scratching their heads.' Of course, that's not at all what I said. I just used 'Frankenstein's monster' as a metaphor for the transposition of two of Shimura's characters with elements from the biography of W.B. Yeats. Neither Dr Frankenstein nor his monster have any part in the dramatis personae of my song.
I can see why this kind of thing happens, though. A journalistic text is a kind of Frankenstein's monster too. The writer has to patch together a lot of assumptions and myths and understandings and try to make a coherent piece that's also interesting to read. This journalist was trying to reconcile what 'the eccentric British songwriter' meant with what 'the average listener' might be able to understand. In the end, he travestied both of us, concluding that average listeners should give the song some attention despite its quirks. For me the song was really only worth writing when I added Yeats and the theme of impotence; for the journalist, these elements almost ruined the song by making it confusing and obscure. The journalist lives in a world where, if enough people say Frankenstein is the monster, that's how you report it. I live in a world where, if Mary Shelley created Dr Frankenstein as a scientist, that's how he should remain. (Some philosopher somewhere has called these different views of the world the 'contract model' and the 'status model': contract people just care about agreement, status people are pedants who insist on some kind of 'truth'. We're always playing to the gallery.)
Occasionally these two worlds -- the public and the cogniscenti, contract and status -- meet. This is what happened on Thursday. My Corkscrew song was a classic novelty hit. It had a TV tie-in element. The reason bestsellers in record and bookshops tend to be TV tie-ins is clear: the public watches a lot of TV. Now, TV is a don't-care, don't-need-to-know kind of medium. Low cost, low commitment. Whatever's on. It passes the time. Whereas music and books need to be sifted and bought and committed to. The internet contains both modes. Although it's a pull medium, most people are pulling the same five websites each day.
Clay Shirky's Power Laws, Blogs, and Inequality is a fascinating exploration of this. He shows how equality of opportunity and equality of result are totally different things, and that even when there seems to be a level playing field -- when everyone can have a blog for free, for instance, accessible to the whole world -- Power Laws dictate that there will soon be a classic comet-shaped ratings curve, with a big spike at the left and a long tapering tail to the right. Attention is not distributed equally. Bloggers tend to settle into the same pattern of 'stars' and 'the rest' that we see in TV and other media. My dictum that 'in the future everyone will be famous for fifteen people' was not wrong, but only a partial truth: it needs to be amended to 'In the future everyone will be famous for fifteen people -- and there'll be a handful of stars who are world famous'. It's and-and, not either-or.
Have a look at the 'comet' shape you get when you chart LiveJournal bloggers by the number of Friends they list. The big spike represents the stars, and the long tail represents more democratic 'family and friends' kind of chatter:
What charts can't so easily show is the quality of attention those star bloggers -- the apparent attention hoggers -- are getting. One indication, though, might be duration stats. Sure, a lot of people might be hitting your page, but how long are they spending there? How deep are they going?
Here's an interesting thing I discovered after my big spike on Thursday. Although my website got more hits than it has ever had, the duration chart shows exactly the opposite shape: page view time dropped to the lowest recorded levels. People with a low level of commitment, web flaneurs, were hitting my pages (and sure, language also played a part: these were Japanese people confronted with an English-language site, although I like to think there's some interesting visual material there too) and leaving within seconds. They also weren't paying: I received a total of ten euros in donations from people with Japanese names. Yes, Paypal is not widely used in Japan. But it's also this: why pay for a song about Ken Shimura when your whole life you've been watching the real thing on TV for free?
My friend Craig Robinson found that the attention of a mass audience is a mixed blessing when a mention of his Minipops pixel popstar drawings on BBC Radio 1 brought hundreds of thousands of people to his website, Flip Flop Flyin. (Craig made the sleeve for my 2003 compilation Forbidden Software Timemachine.) Craig experienced the same sequence that I did: delight when someone told him about the mention, dismay to find the website basically undergoing a 'denial of service' attack, panic as we scrambled to save our servers and find alternative accommodation, more panic as we worked on getting something saleable into the window before the chance passed, and finally a return to normal. Wham bam thank you ma'am!
I'm ambivalent about this. I'm actually really happy distributing mp3s to a small, intelligent, generous and committed community of connoisseurs who share my values. Corkscrew King was never intended to be a 'hit' or to reach a wide public. Its 'success' has been as useless to me as the radio hit I had with The Hairstyle of the Devil back in... oh, never mind. But making content available on the web does always bring the possibility of this kind of swarming and with it the illusion of instant success.
An article in today's Guardian cautions that Apple's mainstream consumer success with the iPod may be the thing that kills the company. By neglecting their core market -- computer snobs and aesthetes like me -- and making their money from portable music hard drives, Apple may be chasing a fad. As the article points out, 'it is hard not to imagine that within a decade most people will be listening to music on the move via a player on their mobile phone.' Nobody will need a hard drive for music any more. And in the meantime, Apple will have lost some of its identity as a serious computer firm, and possibly slipped to even less than its current 1.7% of the US market. When your business depends on optimising your marginality (which means concentrating on 'the good difference'), nothing kills like success.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
For You, Momus...
Date: 2004-08-12 07:24 am (UTC)(Haven't heard any of your music yet, so I don't know whether I'd like that as well. But I'm sure I'll be trying that sooner or later.)
The idea of quantity versus quality of attention is very interesting, and seems to be generally true... *sigh*
Web flaneurs, I like that term! Great.
For the records...
That Guardian Article is a bit stupid, in my opinion.
Where do you think the music people will get streamed to their mobiles will be coming from??? If Apple doesn't seriosuly screw up something, it will exactly be coming from their servers...
And for the stats...
1=It.
2=The Alternative.
3=The others.
4=The one in place of my favourite number.
All my best wishes
-K.K.