Too Fast Too Facile: The Rise Of Online Annotations

In 2014, technocrats and open source crusaders from around the world gathered at an annual conference in California to ruminate over the possibilities of palliating an information-saturated internet with the use of online annotations. Conspicuous among the attendees were representatives from Genius, formerly Rap Genius, which has been provisioned with millions of dollars of VC funding since its inception in 2009. The thrust of the conference was the creation of a universal online annotation system that would not only critique and question the veracity of online content but also network it by hyperlinking and minimizing the degrees of separation between reams of webpages which might otherwise be insulated from each other.

At the conference, Nick Stenning, a developer with, made the most compelling case for online annotation. “…the web will be a vast, varied assembly of sources of information. Annotation provides us with the way of navigating that information…without requiring that the publishers provide it themselves.” The crux of his argument lies in the phrase without requiring that the publishers provide it themselves. As it happens, it is often the web publisher—by having sole discretion over inserting hyperlinks to sources, related webpages—that lays down the route a seeker of information must take in navigating the web. Discounting his comments, which are— besides being at the end of the page and hence inconspicuous and relatively decontextualized—vulnerable to deletion, the user has no recourse to link a webpage to another. Conversely, annotations, by virtue of being inline, function as an incisive, line-specific commentary that let users paste hyperlinks to related webpages without requiring the publisher’s imprimatur.

And indeed that is how Genius, with its melioristic mandate—Annotate The World, could affect a change. Of all the emergent annotation platforms, Genius seems poised to break new ground not least because it is buoyant with venture capital but also because it is being shepherded by Marc Andreessen, the co-founder of the now defunct Mosaic, one of the earliest web browsers which also introduced online annotations to the nascent internet community of the early 90s. Having failed in creating an annotatable web on first attempt, Andreessen, with Genius, hopes to reinvigorate online annotation and this time for good. But the concept of online annotations predates Mosaic. Even though it can be argued that the idea harks all the way back to Vannevar Bush’s Memex Machine, it was Ted Nelson, who, with his seminal Project Xanadu, first began to think critically of the possibilities of creating an annotatable web.

Despite having similar intentions, Nelson’s and Andreessen’s visions were fundamentally dissimilar. Much before Hypertext, which would link countless blogs and primitive webpages that hitherto existed in isolation, Nelson, who actually coined the term, was busy ideating a radically different but vastly superior version of the internet as we know it today. An exposition of how Project Xanadu differs from the contemporary World Wide Web would require another paper but a brief excursus into its fundaments is crucial to drawing lessons for online annotations.

To Nelson, the World Wide Web is an aborted and slipshod version of what he had in mind for Xanadu. “[Xanadu] has always been much more ambitious…where documents may be closely compared side by side and closely annotated; where it is possible to see the origins of every quotation; The Web trivialized this original Xanadu model, vastly but incorrectly simplifying these problems…Fonts and glitz, rather than content connective structure, prevail.”

It is Nelson’s emphasis on connective structure that makes WWW pale in comparison to Xanadu; two-way links, as he labels them, allow the user to view a document that either borrows, references or derives content in simultaneity with the source document from which it borrows, references or derives. In fact, by using beams to connect content to its source document, it visualizes not just connections between documents but between lines and paragraphs scattered across documents.

A screen shot of Xanadu’s working deliverable, courtesy:

Stated laconically, Xanadu traces not just the genealogy of documents but functions as a kind of an omniscient library system, mapping the web of interconnections in the accumulation of human knowledge.

But one could argue that the web, as it exists today, allows for comparing a source document and a derivative document by displaying them in different tabs or windows; but, it does not provide for two-way links. For example, a news report about discrepancies in a company’s financial statement would hyperlink itself to the said financial statement released on the company’s website. But, despite the availability of myriad backlink softwares that notify a webmaster every time another webpage links to their website, it is unlikely that the website would reciprocate the action by linking its financial statement to the news report. As mentioned earlier, this is because hyperlinking to another webpage is at the discretion of the web publisher. Xanadu’s provision for two-way linking ensures that no document can exist in isolation which effectively means that it displays not just links (or beams, as illustrated above) to a source document but links from a source document to all documents that source from it.

In the ground-breaking Death Of The Author, literary theorist Roland Barthes wrote:
“The text is a tissue of citations, resulting from the thousand sources of culture”

Arguably, Barthes’ aphorism is a more elegant summary of the more banal dictum: All Knowledge Is Derivative. With two-way links, Xanadu imitates it by tracing every quotation or idea to its very source such that there is no document or text that exists without being foregrounded in the scholarship that precedes and influences it and that which proceeds and is influenced by it.

But how does this inform annotations?

Although Xanadu never got off its feat and is considered more or less vaporware, online annotations can be an effective tool in mitigating some of the damage that the shoddy implementation of hypertext has wrought on the web. And this is where Genius can be of service.

Any user, after signing up for a free Genius account, can annotate the web. Genius is different from other annotation platforms which tend to be browser plugins; rather, it looms over the web which is to say that it precedes the URL—for example, past Genius annotations on the LA Times website can be viewed by Genius users and new annotations can be made by going to If a Genius user finds a story on Seattle Times that is related to another story on LA Times, he can make an annotation on and insert a hyperlink to This is, however, still a one-way link. But, with the right backlink software, Genius will be notified that a user has linked to and can use bots to display the in-bound link to Seattle Times ( on LA Times (

What makes this prospect of two-way linking irresistible is that Genius can do this without needing the permission of either newspaper. The whole mechanism may seem tedious even undoable. But, with the weight of influential investors and millions of dollars behind them, Genius is perfectly positioned to delve into two-way linking and channelizing funds into conceiving new ways of accomplishing it.

Two-way linking is essential not only for more transparency and navigability of information as the two examples illustrated but also for creating a highly interlinked web. More connections and reciprocal connections would create an infinitely networked and heuristic World Wide Web where information would be more accessible and one where users can amble from one website to another without solely relying on search engines and a list of favored websites as their gateways to the web. It would pave the way for a more equitable internet—one where information would be scattered across multitudes of websites—and where a few media organizations would not hold a monopoly over privileged information and take editorial calls over the publishing of sensitive content.

Nelson’s Project Xanadu was a spectacular failure; but it prognosticated problems that have only come to the surface since big internet companies started implementing Hypertext with little foresight and content began pullulating the internet in the last two decades.

Online annotation, which is yet to come to fruition, can, to some degree, bring us closer to a Xanadu-like internet. But, Genius, with its emphasis on the ‘Worse Is Better’ model of business, seems to be prioritizing scaling up over and above other imperatives. In fact, the founders of Rap Genius are taking comfort in the fact that the introduction of Hypertext was met with similar consternation which eventually fizzled out. In doing so, it is evincing the same haste and impatience that the internet behemoths demonstrated in their road to El Dorado.

Nelson wouldn’t be surprised.


RapGenius Rebrands With $40M, Aims to ‘Annotate the World’, Lora Kolodny, Wall Street Journal
Perpectives on Annotation, W3 TPAC Conference, Oct 2014
Why Andreessen Horowitz Is Investing in Rap Genius
Toward an ecology of hypertext notation, Catherine C Marshall, Xerox Palo Alto Research Center
Pioneering hypertext project Xanadu released after 54 years,
The Death Of The Author, Roland Barthes
Xanalogical Structure, Needed Now More than Ever:
Parallel Documents, Deep Links to Content, Deep Versioning and Deep Re-Use, Project Xanadu

The curse of Xanadu, Gary Wolf, Wired
Genius Idea, Reeves Wiedeman, New York Magazine

2 Replies to “Too Fast Too Facile: The Rise Of Online Annotations”

  1. [This comment is not part of my review:
    Nitant, I loved your essay. I thoought it was fresh, interesting (yes interesting) and above all well-written. You’re a fantastic writer. The review I’m leaving below is what I would send you if I was reviewing your work for an academic journal. In other words, while it’s an honest review, it’s way more critical than a 1500 word essay due the week before the magazine project deserves.
    Thanks for a great read! ]

    “Too Fast Too Facile: the Rise of Online Annotations” explores how Genius, an online annotation service, could impact web-based reading. There is a tension in this essay between two related concepts: that of social annotation and that of two-way linking, both of which are made possible by Genius. This essay would be better served by choosing on one of these two elements and elucidating the possibilities it entails for publishing and the web in general.

    One of Genius’ strengths lies in the possibility for social annotation. The author does a good job of contextualizing the importance of this tool: “Annotations, by virtue of being inline, function as an incisive, line-specific commentary that let users paste hyperlinks to related webpages without requiring the publisher’s imprimatur.” The concept of social annotation is not new: Wattpad, CommentPress and Kobo all make this possible. However, as the author astutely points out, Genius’ “genius” is in divorcing this tool from the publisher’s platform and making it available on every website. The author could push this topic further by delving deeper into how, in his words, “it would pave the way for a more equitable internet—one where information would be scattered across multitudes of websites—and where a few media organizations would not hold a monopoly over privileged information and take editorial calls over the publishing of sensitive content.” To develop this topic further, the author would need to bring up the changes brought about by social reading, as well as the need to filter comments (in Genius’ case, through its IQ ranking system).

    A more interesting aspect of Genius, and one that takes priority in this essay, is the possibility it opens for two-way linking. The author does an excellent job of situating this feature within its historical context, and the paragraphs on Xanadu are well-placed and surprising. The quotation from Roland Barthes grounds Xanadu’s connection to publishing. It hints at the possibilities for a two-way system: as the author puts it, “Xanadu traces not just the genealogy of documents but functions as a kind of an omniscient library system, mapping the web of interconnections in the accumulation of human knowledge.” The author could have delved further into the possibilities evoked by this fascinating system, especially as it relates to publishers. At its most basic level, Xanadu’s architecture solves some of the biggest problems with the World Wide Web: the difficulty of tracing links backwards to determine influence (which publishers must now rely on Google and other services to do with spiders) and the impermanence of links, which are broken when content moves to a different URL (which publishers solve with DOI, among other things). But the possibilities don’t stop there: see Jaron Lanier’s Who Owns the Future? for a far-out account of how two-way links could revolutionize copyright and IP laws. Genius adds another layer to this as it relies on a social element where only links posted by outside users would be two-way, but not the links within the text itself. In this respect, the comparison between Genius and Xanadu feels rushed, and the author could have developed further why he believes Genius is failing to live up to his potential.

    Overall, this is an excellent, well-written essay. I appreciate that the author chose a subject that is fresh and unusual, especially as it relates to publishing. Since not much has been written about this subject before, it is hard to decide which aspects of it to write about, and the temptation is to include all. As I have argued, the author will need to decide which aspect of Genius to prioritize–its social annotation features or the possibility for two-way links–and delve further into its implications for publishing through that lens.

  2. This is essay offers a novel take on a novel subject. It takes the energy and enthusiasm that annotations bring to the web, and links it back to a time in the early history of the web where there were boundless possibilities. By providing that historical view, the author opens us up to considering any and all possibilities that a project like Genius can offer us and succeeds in getting us to want Genius to be everything and more.

    Moreover, the introduction of project Xanadu and of two ways links makes a compelling case for the need, even if the the explanation is a little clumsy at times. In this sense, I felt the essay could have been better organized to first convince us of the need and advantages of two way links, show us its deep historical roots, and then show us how annotations are the renewed opportunity to make two-way links a reality.

    Overall, however, a great essay, one that invites us to dream of the possibilities that have yet to be realized on the web.

Comments are closed.