14.0837 methodological response: hypertext

From: by way of Willard McCarty (willard@lists.village.Virginia.EDU)
Date: Fri May 04 2001 - 01:56:26 EDT

  • Next message: by way of Willard McCarty: "14.0835 disappearing messages not censored!"

                   Humanist Discussion Group, Vol. 14, No. 837.
           Centre for Computing in the Humanities, King's College London

             Date: Fri, 04 May 2001 06:53:33 +0100
             From: Patrick Durusau <pdurusau@emory.edu>
             Subject: Re: 14.0827 methodological response: hypertext


    I seem to have started a rabbit chase with my reference to the mental
    processes of a reader who encounters a traditional footnote or a
    hyperlink. Perhaps the analogy was poorly choosen or inadequately

    The original point in question (as I understood Willard's earlier
    posting) was whether or not hyperlinks are something more than
    traditional methods of referencing other materials. I suggested that
    hyperlinks, like footnotes, refer the reader to other material but both
    pale by comparison to the actual associations formed by a reader while
    reading a text. Perhaps I should have just avoided the mental
    associations imagery and emphasized the mechanical equivalence without
    further comment. (I am aware of the convenience of hyperlinking to make
    material more accessible. That does not answer the question of whether
    hyperlinks are fundamentally different from traditional references.)

    In terms of specific responses:


    > Date: Sun, 29 Apr 2001 07:35:22 +0100
    > From: "Fotis Jannidis" <fotis.jannidis@lrz.uni-muenchen.de>
    > >
    > > From: Patrick Durusau <pdurusau@emory.edu>
    > > Hypertext links are no more than an unremarkable, seamless,
    > > inconsequential retooling of traditional reference mechanisms. (To
    > > appropriate some of Willard's language.) Perhaps a little less
    > > contentiously and more accurately, hypertext links are expressions of
    > > the same mental processes that are evidenced by traditional reference
    > > mechanisms, albeit easier to consult than traditional references.
    > It seems to me you mention the main points of difference but I can't
    > follow your argument how to evaluate them.
    > Maybe it is useful to apply the distinction between material text,
    > that is some marks on paper or some structured bits, and mental
    > text, that is the text as it is represented in the mind of an author or
    > a reader. If we use this distinction to look at a hypertext link, it
    > becomes obvious that the mental text is quite similar to that of a
    > footnote or a similar reference whose target is readily available, but
    > the material text is quite different, because the reference
    > mechanism has to be coded in a way which is not only
    > understandable to a human but also to a machine.

    I am contending there is no meaningful difference between footnote and
    hypertext links. Willard has asserted the contrary to be the case but I
    remain unconvinced. Hypertext links may be more accessible (assuming no
    "404 - not found" errors) but that does not strike me as a fundamental
    distinction between footnotes and links.

    > Another concept can be used to refine this picture: the amount of
    > work to resolve a link. That this is an important factor becomes
    > obvious if you don't look at one link but many. Even in the case of
    > footnotes or endnotes the almost unnoticeable amount of work to
    > move your eyes to the footnote and back to the main text amounts
    > after a while to something noticeable. These becomes even more
    > obvious in the case of links to very different texts. The work to
    > resolve a conventional reference becomes part of your mental map
    > (all people working with books are full of stories about the
    > difficulties to get some of them) and the main consequence is that
    > you only follow references which look promising enough to
    > undergo this work. This really changes with hypertext links. The
    > work to resolve the link is almost null, so only the time to read the
    > target text remains to be taken into consideration. So it is not only
    > a little bit easier, but from this difference a new praxis of reading
    > results.

    Interesting. I assume you would contend the praxis of reading changed
    with the transition from scrolls to codex? At least in the current
    environment we have the opportunity to study reading practices as they
    develop for hypertext. But it would be from such studies, not simply
    contending it to be the case, that evidence of "a new praxis of reading"
    would emerge. Even if such a demonstration exists, it does not address
    the question of a qualitative difference between hypertext links and

    At the risk of starting another false trail, consider the difference
    between the quoting of a work under discussion (examples abound in the
    Church Fathers) versus a reference to such other work in a footnote. The
    latter presumes the rise of collections of works with some method of
    reference that can be resolved by the reader (more or less) but serves
    the same function as the earlier practice. Much like the footnote versus
    the hypertext link, I don't see a qualitative difference between the two
    practices, although the demands upon the reader are greater in the
    latter case than the former.

    > Maybe the idea that
    > hypertexts are working as our minds work is misleading.

    Yes, a poorly chosen example.



    I must confess I have not read all 200 pages of your essays cited in
    your post prior to typing this reply. While I think you advance a number
    of interesting points, I am not sure any of them address the issue of
    any essential difference between a hypertext link and a footnote. To
    illustrate, the matter of "gutter (the space between panels in comics)"
    is a question of presentation, and I certainly concede that changing a
    presentation may alter a reader's perception of a work. That does not
    address the issue of hypertext links versus footnotes.



    > Somewhere in his voluminous writings (someone please tell me where) the art
    > historian Ananda K Coomaraswamy speaks with sharp tongue about what he
    > calls the "nothing-more-ist" response to visionary insights, which it
    > construes as unsupportable claims, "nothing more than X", where X is
    > something unremarkable. Being Coomaraswamy, he gives a precise Sanskrit
    > term (and likely any Patristic equivalents also) for this sort of attiitude
    > to bury it as deep as intensely learned anathema can. But all this is
    > really itself no more than a semi-learned flourish of entertainment to
    take up
    > once again my argument with Patrick Durusau (with whom I also partially
    > agree). In Humanist 14.0822 he responds in the nothing-more-ist fashion to
    > hypertext:

    First, I don't consider footnotes to be "unremarkable." Perhaps
    familiarity has breed contempt for the footnote. Has it fallen to the
    state of: "It is a footnote, it is not read." ? (Apologies to Victor

    Second, simply proclaiming something a "visionary insight," even if such
    claims are quite in vogue, does not make an insight visionary.

    > >Hypertext links are no more than an unremarkable, seamless,
    > >inconsequential retooling of traditional reference mechanisms. (To
    > >appropriate some of Willard's language.) Perhaps a little less
    > >contentiously and more accurately, hypertext links are expressions of
    > >the same mental processes that are evidenced by traditional reference
    > >mechanisms, albeit easier to consult than traditional references.
    > .....
    > >Despite all the "now I am awake" rhetoric from the W3C crowd, the fact
    > >remains that hypertext (even assuming robust implementaitons of
    > >XLink/XPointer/XPath) is an impoverished expression of the associations
    > >that a skilled reader forms while reading a text. That is not an
    > >argument against hypertext, but against the notion that it is
    > >qualitatively different from traditional reference practices. Neither
    > >can fully reflect the associations made by a reader.
    > >
    > >Hypertext technology should be supported/promoted for its enormous
    > >potential for accessibility, research, and collaboration. It has enough
    > >promise in those areas to not need the more questionable "now for
    > >something completely different" claim.
    > Perhaps I am missing something essential, but it does seem to me on this
    > overcast Sunday morning that one cannot have the matter both ways:
    > hypertext cannot be both "no more than an unremarkable, seamless,
    > inconsequential retooling of traditional reference mechanisms" and "an
    > impoverished expression of the associations that a skilled reader forms
    > while reading a text". We are remarking and will continue to; we feel the
    > bumps right now; and the consequences of that impoverishment are enormous!
    > Which is to say, I certainly agree that *from the perspective of the
    > natural-language devices one has for making references* hypertext of any
    > kind I have experienced or read about or can imagine is indeed a profoundly
    > impoverished metalinguistic set of mechanisms for referentiality. Just as
    > all forms of applied computing, indeed all results of modelling phenomena,
    > are impoverished from the get-go. From a research point of view, bringing
    > such impoverishment into focus is precisely the point of the exercise. Long
    > live such impoverishment: it shows us where the riches are, gets us to look
    > at and try to understand them better. (Yes, from the engineering
    > perspective, which I also treasure, such impoverishment keeps one awake at
    > night and busy through the day.)

    Note that in my post I said that both hypertext links and footnotes were
    pale imitations of the associations a reader forms while reading a text.
    I could have inserted hypertext links at several points in this post but
    I doubt that I have the skill to insert (either due to missing resources
    or the difficulty of catching every association I have made while
    writing this reply) enough hyperlinks to duplicate. The point I was
    attempting to make was that all reference mechanisms fall short of fully
    representing the associations readers form while reading a text.

    > I also have trouble with the statement that "hypertext links are
    > expressions of the same mental processes that are evidenced by traditional
    > reference mechanisms". Again, wake me if I am asleep, but if you find me
    > awake tell me please how we know what our mental processes are? Is not any
    > statement about these processes a model of something we cannot know
    > directly -- until that day when we see face to face (or after the
    > revolution, as you prefer)? I don't think that people like Steven Pinker,
    > who in books such as Words and Rules make bold to tell us exactly how the
    > mind works, have been vouchsafed a peek; I think they've simply forgotten
    > that all schemes are tentative constructs.

    Yes, a poorly choosen example. I have noted when (other) authors not
    only solve the problem of other minds but appear to know what they were
    thinking while writing replies in scholarly disputes. It was meant
    simply to convey a richness of associations that exceeds that possible
    to express with footnotes or hypertext links.

    > No, I am not claiming that now our intellectual sun is dawning on a day
    > never before seen. Hypertextual links are in some respects the same, in
    > some respects different in what I can guess are the mental processes
    > involved -- at least the ones I consciously work with when making
    > reference. Indeed the differences are hard to identify when you get down to
    > looking at the matter carefully. (See, for example, the historically
    > informed arguments of Carla Hesse, James O'Donnell, Paul Duguid, Geoffrey
    > Nunberg et al. in *The Future of the Book*, Univ of California Press,
    > 1996.) My interest is in stimulating that nitty-gritty work of careful
    > analysis, against the techno-evangelists, who really should turn to a
    > religion with some future and leave us to get on with the work. In other
    > words, I think (with Frank Tompa and Darrell Raymond, among others) that we
    > need to do very careful artefactual analyses of the referential works we
    > have inherited, from the perspective of computational hypertext, always
    > asking where the differences lie. Also, as Adrian Miles points out, think
    > up new and interesting ideas (a.k.a. theories) of how to look at
    > referentiality.

    I did not read your original post as "stimulating that nitty-gritty work
    of careful analysis" since it appeared to presume the existence of such
    differences. I should have indicated at the outset that it was that
    assumption that I found troubling. There may be such differences but I
    have yet to see any convincing analysis to support that claim. My
    personal suspicion is that physical texts and reference systems are far
    more complex than our current fascination with hypertext media is
    willing to investigate.


    Patrick Durusau
    Director of Research and Development
    Society of Biblical Literature

    This archive was generated by hypermail 2b30 : Fri May 04 2001 - 02:02:05 EDT