1. 5

  2. 2

    An inescapable truth is that publishers and aggregators both need to generate revenue to continue their operations.

    This is sufficiently wrong to kill my gumption for reading the rest of the document in earnest good faith. For example, here’s a paradigm which is totally absent, because of the folly of the above line:

    Digital Libraries. A digital library is a curated collection of selected hypertexts and other interactive media. The digital library includes both large literal corpora and also many indices of hyperlinks and other cross-referencing.

    User Experience: Customizeable. While all of the original data is still present for the “optimal” experience, users are also able to directly manipulate hypertext as they like, and create custom presentations of data.

    User Privacy: Proactive. Unlike the “optimal” privacy experience where user metadata is sent to publishers, a digital library does not remit its users’ private information to the publishers, and even better, the library can also hide the identity of authors. When the library’s indices are themselves decentralized, then a user’s privacy footprint is limited to the peers that they directly talk to. (This is still not yet optimal; there’s theoretical ways to run this sort of library without any metadata leaks, at the cost of sending much more data overall.)

    Content Preview: Supported. Index generation can include preview generation, so library indices can include content previews.

    Preloading: Supported. Both in the way that the article imagines, and also in the sense that digital libraries can allow users to consume unbounded amounts of content.

    Publisher Level of Engagement: Reasonable. Publishers traditionally don’t like libraries, but they don’t have to actually do any positive action in order to be included in a library, other than make good art.

    Revenue Model: None. Publishers, users, and the curators of the digital library are uncompensated. However, their combined efforts are too cheap to meter, and so it doesn’t make sense to try to monetize the system.

    1. 3

      Not following your point. As I understand it, the quote is as mundane as stating that publishers sometimes require capital to compensate creators, and for operational overhead, and aggregators certainly also have operational overhead and development costs. We can meaningfully compare different revenue models, but to simply host a popular web service in the first place does cost money.

      Digital libraries take a significant amount of work to maintain. Digital archivists do meaningful work for which they need to be paid either directly or indirectly. Sysadmins and other tech workers must put in effort to build out and maintain those user experiences you mention; even moreso the more complex and customizable it becomes.

      Revenue Model: None. Publishers, users, and the curators of the digital library are uncompensated. However, their combined efforts are too cheap to meter, and so it doesn’t make sense to try to monetize the system.

      Maybe we should compare to Wikipedia, which may be the most successful system similar to the one you’ve proposed. Wikipedia very much requires money to operate: https://digital.hbs.edu/platform-rctom/submission/wikipedia-here-for-now/.

      1. 1

        Publication costs can be so cheap that the artist is effectively self-published as soon as they share their art. Similarly, aggregation can be done at the moment of publication. A simple model embodying this is the bulletin board. The capital cost of the bulletin boards in my local parks, as well as the maintenance cost, are included in the cost of the entire park and are negligible; they don’t justify nor require dedicated employees just to maintain bulletin boards. The board is spatially local and provides aggregation via local publication.

        The folly in the original sentence stems from a belief that publication and aggregation need to somehow be dedicated tasks, complementary to but never overlapping with the creative effort, and therefore that publishing houses and media aggregation companies therefore deserve to exist. However, no, the only reason that we value art is for the art itself, and we only value publishers of art for what they publish; publishers and aggregators are middlemen who stand between creators and viewers and seek rent.

        Yes, to “simply host a popular web service” is not free. We can conclude that the Web is not a suitable structure for sharing data at scale, and indeed the Web is a series of hacks deployed to keep pace with the exponential growth of the software industry. Right now, if we bypass the Web, then we can amortize the amount of bandwidth required to publish, and perhaps eventually something like NDN will replace the Web as the main content delivery protocol.

        I created content for Wikipedia. I wasn’t paid for it. You seem to think that I should have been paid. Who would have paid me, and for what? I didn’t generate that much value on my own. Moreover, as soon as there are financial incentives, then the art suffers; concretely, paid Wikipedia editing is almost always too biased to survive Wikipedia’s neutral-point-of-view requirements. Don’t worry about Wikipedia; it’s got deep pockets and it’s already been paid for. They’ll have to figure out the long-term investments, but they have plenty of cash and cash flow. Rather, quoting from your link:

        These contributors are the true value-creators of Wikipedia and losing significant numbers of contributors would be the death knell for the platform.

        To try to restate my point: We are currently stuck in a hellish society with works-for-hire, copyrights after death, and other abominations. We trap users in our applications, and then demand that they use our designs. But this doesn’t have to be how we live together. We could simply give users the raw data for whatever art they are trying to consume; rather than us trying to justify our large salaries by extending more corporate control, we could leave the users alone and let them design their own user experiences.

        Or, to be extremely pointed: It’s cheaper to operate arXiv than Elsevier, about a thousand times over.