You're very welcome, Andrea, and sorry for the lag in replying.

As you say, textual works are problematic in this way, because of the rarity of wholesale embedding; quoting short passages is quite common and acceptred, and in those cases being able to link to the full work is an improvement on print citation styles, but the web provides us with a good way to do this already - my http://mediagora.com/sources.html page links to both full online texts and to print editions (via Amazon).

Stephen Fry (in The Hippopotamus) comments on how poetry vanishes into the public domain as it is repeated.

One method would be to use the subdivision techniques discussed in the paragraphsmall pieces tightly bound - publish a collections of poems both individually with a small price, and as a collection that offers better value.

Defining derivative works is something that keeps courts very busy at the moment - many of the egregious examples of copyright abuse that Lessig cites are in this area. These disputes usually only arise once the derivative has made some money, and often do need to be settled by negotiation or litigation.

If Jim cites Bob's work through mediAgora, then he should be driving sales to Bob for the work that is incorporated - if he writes a short story based on characters in Bob's indivisible novel, then he is expecting his readers to have bought and read Bob's novel first. If they buy Jim's short story, then they have bought Bob's novel too.

A way to resolve disputes over attribution and plagiarism will need to be part of the mediAgora terms of use - perhaps reputation ratings as used successfully by eBay for rating buyers and sellers could come into play.

Bootleg culture at Salon describes exactly the kind of thing I envisage legitimating here.

How does this play in weblogs? Lots of presumably fair-use citation of others' posts. Do blogging tools need to recognize citations?

Edited by author 12-21-2002 05:48 PM
Fair use still applies - mediAgora is a voluntary payment encouragement system, not a replacement for copyright law.
In the case of a weblog post, the original post was presumably free, so citing it would mean you charge your customers the same price, ie nothing.
A hidden presumption in mediAgora is that the works in question are substantial enough to deserve payment of a few cents at least; this is one reason DRM is discouraged. While the basic model scales to arbitrarily small payments, in practice the 'mental transaction costs' that Odlyzko disucsses for the purchasers would likely come into play before the transaction costs of the payment measurement system. A practical implementation would probably impose a minimum price.

I would say, as a customer, that rather than a fair price, I normally look for the lowest price. I can't recall the last time I was at a supermarket and passed up a sale because the price just looked too low, and it wouldn't be fair to the market.

Now you're just being silly. Obviously a price needs to be fair to both parties involved. Absent that, the trade doesn't happen.

Can't believe you don't namecheck Brad Cox. Superdistribution addresses many of the same issues.

OK, heres' the old Superdistribution link:
http://www.wired.com/wired/archive/2.09/superdis_pr.html

Why is this still a non starter 9 years later? Because it upsets customers.

They feel nickeled and dimed to death. The mental transaction costs are too high. They have no incentive to redistribute the goods, becasue it would just nag their friends to death too.

Instead of pay per use, mediAgora advocates pay once for life. A lifetime right to a good quality copy of the media. This is a more attractive proposition, and is valued more by the customer.

Secondly it provides a financial incentive for redistribution when it leads to a sale. This helps align their incentives with the creators'. But they only get this payback if they have bought it in the first place.

How on earth do you do that? How would you troll through Dante and figure out which lines he borrowed and adapted from Virgil? How would you parcel out credit to the blues musicians whose styles went into the Rolling Stones. All art is made from other art. Or does this comment apply only to direct "sampling."

Edited by author 08-07-2003 08:29 AM
Ah- I saw the answer in the comments to the previous paragraph. But that is problematic - copyright law has a tradition of fair use, which defines small-scale quotation as legal. I think fair use should be extended to sampling, instead of using digital technology to trace all the allusions and cause Dante to make micropayments to Virgil's estate.

That is reasonable, Adina. I see such issues being handles on of two ways - fro substantial samples, listing the source work is the correct thing to do; for small fair use ones, small-scale quotations are fine. And both Virgil and Dante are in the public domain now. As I said in the general comments, this is not meant to be a substitute for existing copyright law, but a voluntary license for Creators to adopt.

The source work inclusion at full price is a default model, and can always be overridden by a negotiated agreement to list the author of the sampled work as a co-creator. In addition, Creators could list subsets of works separately with samplers and remixers in mind, much as some artist make separate vocals and backing trecks available to DJs for remixing today.