28 Mar 2012

Too Big To Know - David Weinberger (Review)

The first time I came across David Weinberger was when I read his paper Tagging and Why it Matters* a few years ago, so before I even opened the book I figured the central premise of Too Big To Know would be closely linked to concepts like information overload, social filters and metadata. Whilst this is certainly true, Weinberger's analysis is thankfully not that simplistic and one-sided.

He begins his argument with the idea that information overload is not a new problem 'created' by the internet, echoing a similar point made by James Gleick in last year's excellent The Information. Ever since Gutenberg's printing press individuals have felt overwhelmed by the idea that there is more information out there than one can possibility assimilate in a lifetime (so there goes my dream of being a Renaissance man polymath...). The internet is simply the latest incarnation of a longstanding problem, but as Weinberger argues it is not so much a 'problem', but rather a fact. In print form, ideas were theoretically connected but often physically separate; in the online world knowledge is communicated, linked and shared within the context of a network (or multiple networks) - and it is the network which becomes most valuable. Expertise is no longer something which can be narrowly defined. Instead it is now collaborative and multidisciplinary, and the opportunity to connect knowledge and ideas in this way stems from the concept of the network.

As information becomes increasingly open, we see post-filtering (i.e. a publish then filter model) rather than pre-filtering (the traditional gatekeeping role played by publishers). This ensures that false negatives (good content which should be published but isn't) are minimised, even if large volumes of seemingly valueless content are also published. In the same way that Clay Johnson suggests responsibility lies with the individual user to consume a healthy 'information diet', Weinberger invokes Clay Shirky's "filter failure" - the idea that we need to develop better and more efficient means of filtering information. Logical stopping points like the authoritative print reference sources of old are harder to see online, but are there nonetheless. Indeed Weinberger's pragmatism brings an awareness that in most cases we utilise information like any other product - as a means to an end - in short, "knowledge is not a library but a playlist tuned to our present interests."

Weinberger also offers an interesting critique of The Shallows by Nicholas Carr and the premise that the internet is negatively influencing our information behaviour by encouraging scanning - ultimately making us lose sight of our higher order thinking skills. Conversely, printed books promote engagement with the longer thought and stimulate creativity and critical thinking. Weinberger respectfully questions this notion however, arguing instead that books are narrow sources of information limited by their form and design - the need for a beginning, middle and an end, all of which must generally avoid digression due to physical constraints. These attributes are not required with electronic publishing, and surely the opportunities presented by richer multi-layered arguments, collaboration and instant feedback are a good thing? Furthermore, accepting the new open ecology of information as a given automatically places a greater emphasis on the need for critical thinking and appraisal, including a responsibility on individuals to fully reflect on and assess the information they find. I would certainly like to think this is true and will be borne out over time, however whether it is over-optimistic remains to be seen. Myself and Mr Weinberger have our fingers crossed.

*Weinberger, D. (2005) Tagging and Why it Matters. Retrieved 27/03/2012 from http://cyber.law.harvard.edu/sites/cyber.law.harvard.edu/files/07-WhyTaggingMatters.pdf


  1. I'm sure Gleick is probably correct to say that, historically speaking, information overload is not a new phenomenon, but surely what we are experiencing now is of a different order of magnitude altogether? Not only are we more aware of all of the knowledge and/or information out there, it's also so much more accessible.

    I like the idea of post-filtering - for me twitter fulfills that function. It is impossible to keep up with all the interesting blogs/ websites out there, but I think by following people with similar interests you can make some sense of the cacophony.

    Not sure I'm really convinced by Weinberger's idea that the book's format is inherently limited as a source of information - surely it's the quality of what's being said that matters ultimately, and that holds true for content published electronically as well.

  2. Thanks for the comment John - interesting points! I know where you are coming from, but perhaps it is all relative? I mean the arrival of the Penguin popular paperbacks made books far more accessible in a short space of time, when books had been traditionally seen as objects exclusively for the learned and elite classes. That was a pretty big change at the time I am sure, just like today when we are going from a state of print information being widespread and relatively accessible to digital publishing - difficult to compare / contrast them without having experienced both personally I suppose!

    I guess Weinberger is coming from the perspective that the printed book, as a static, fixed and finite form, is limited as regards the world of scholarly research, where ideas are typically shaped and refined over time through collaboration and information sharing. However, I am sure many would argue that Dickens and Roth have managed ok with it so far :)

  3. First, Michelle, thank you for the thoughtful and positive review.

    John, I do agree with you that info overload now is different than it has been. The book does indeed acknowledge that the magnitude of info now is so so so much greater than in the fast. I also point to the difference in the nature of filters now and before. In the past, filters physically removed that which was separated out. In the digital world, however, filters only "filter forward" (i.e., reduce the number of clicks it takes to get to an item), while leaving all the stuff that didn't make it through the filter still available. So, yes, I do agree with you.

  4. Thanks for the clarification David. I do think that ultimately the the potential overload factor of a 'filterer forward' process is worth the 'cost' in the long run, as making all data and information available (not just in a filtered form) is potentially valuable. Thanks a lot for the comment David and good luck with the book.