Transclusion and Google Print
Published by Jeremy Douglass November 14th, 2005 in search, News, Text Art, Social.For those of you who haven’t typed your favorite topic words and phrases into Google Print, I highly recommend it. Some search terms I tried included:
Some results are, of course, better than others. “Electronic literature” is a bit hit-and-miss compared to eliterature, and “chatbot” results are quite thin. Still, when looking at the results, I experience two simultaneous emotions:
- strong satisfaction at seeing all the recognized texts come up
- intense curiosity at the ones I don’t recognize
It really is remarkable how wrong the Association of American Publishers has been in its lawsuit against Google, particularly their implication that Google Print is a service that will replace the original text and thus depress sales. Quite the contrary, looking at a page of results makes me immediately order books that before I never new existed. This point has been made by many people elsewhere, but perhaps the best response to the AAP lawsuit so far has been Nick Schulz’s article “Don’t Fear Google” (Forbes. 2005-11-03):
“…my finding that snippet and using it for this article is not a copyright violation. I didn’t ask Schroeder or her publisher for permission to use the quote in her book. Indeed, there’s an entire industry, book reviewing, predicated on the ability of people to do something similar to what I’ve just done.”
Book review excerpts are also similar to fair-use quotation as it is used in academic scholarship. Neither practice is strictly commercial or non-commercial, likewise neither exclusively devoted to nor opposed to the free spread of knowledge. What interests me more generally about this ongoing debate is the way that the language of “copying,” always the sticking point of so-called intellectual property law in the online space, is reaching a crisis-point in the unified textual databases of corporations like Amazon and Google.
The problem for copyright holders is not that Google is, Grokster-like, engaging in the dissemination of copies. The danger is that Google and others are *not* copying. In hypertext and eliterature terms, what Google Print search results do is transclusion, Ted Nelson’s term for the ‘live’ inclusion of a digital document section into another digital document. The HTML web developed largely without transclusion infrastructure, as there was no unified frame to guarantee transclusion operations - even if I could transclude a section of the Forbes article above, I have no motivation to do so, as I can’t trust Forbes not to archive or hide it, and in doing so damage the sense of my article. Instead of transcluding, I (legally) copy an excerpt. Then there are two copies of Nick Schulz’s text - one on the Forbes server, and one (excerpt) on mine.
Yet within an enormous closed database architecture such as Amazon or Google’s, transclusion is king - in fact, in most paradigms of database design actively discouraged copying. One piece of text is never stored in two locations simultaneously. Instead, database design focuses on uniqueness and avoiding duplication. One article. One instance of the text. That instance, and only that instance, will be consulted for every excerpt - it is in essence a single file on a digital shelf in the world’s biggest library… until visitors begin loading page views, at which point duplication occurs as part of the fundamental operation whereby information travels through an electronic system (such as the internet).
But to return to the point: Many texts, one unified database, one master record per text. These are the preconditions for a massive transclusion machine. Although Google’s aims are in some ways quite different from those of Nelson’s Xanadu, it is interesting to think what else one could do with such a setup….
7 Responses to “Transclusion and Google Print”
- 1 Trackback on Nov 14th, 2005 at 1:31 pm
Jeremy,
I’d recommend: “conversational agent” or another term for your “chatbot” search. That’s probably more accurate for the broader technology. I know I use chatbot because of the connotations, though I realize these have more to do with the historical object than the developing technologies.
Another gigantic transclusion machine (at a more atomic level) is Google Base, just officially out. In fact, the atomic unit of Google Base, the “item,” may not only appear in search results, but “may also be included in the main Google search index and other Google products like Froogle and Google Local.”
Google Base uses an extremely flexible concept of an “item” - broken down by a flexible typology that is user editable, with a huge number of possible field types. To this extent, it doesn’t make sense to discuss it in terms of classic database table or XML design, which focuses on rigid similarity of required values between all members. Methods of customizeable and extensible fields (WRT’s current blog software, WordPress, does a fair bit of this itself) push at the edge of the ‘table’ paradigm, allowing for easy collection of significantly heterogeneous content.
Sure enough, conversational agent gives much more satisfying results.
Could you expand on why you prefer the term chatbot - is it identifying with particular technologies and communities of authors, or is it coming out of some particular history…?
Thus, electro-telepathy is progressively expanding its territory and external effects.
For one mind to query any other mind in the past, present, or presumably the digitized future, the web must be free, spontaneous, floating luxuriously above the petty strife of human hubris, corporate greed, lawyerly debate on intellectual “property” (what a whopping misnomer par excellance). Mix one part oxymoron with two parts hydromoron, shake vigorously, voila: the quick crucifixation of the poor thought morsel.
A peering into the pyramid reveals that the theatre of colliding ideas will play itself out in the screens of computers filling both earth and outer space.
I mean, to duplicate a slice, a thin slice, of another text, and put in either within another text, or alongside it, ala Derrida and Dada poetics, this is just the circus tent, and not the cotton candy, right?
We will dominate or submit. I choose to tease language and force odd concepts to co-exist in formerly ordinary mind-machines.
Cataclysmic Whisper Transmission Zone: the Blogosphere 4.0
Also connected to this idea of the unitary (rather than multiple) database text is a discussion of metadata and the networked book over at if:book. Interesting, this penumbra of metadata, which seems to coalesce around some singular thing….
I like the invocation of Nelson’s ‘transclusion’ as a way to thwart the idea of ‘copying’. In my cross-media research I try and explain the difference between transmedia and repurposing or versioning through the notion a single source with multiple windows. So, this notion has resonnance for me. I like it. Thanks Jeremy!