18.731 nomenclature

From: Humanist Discussion Group (by way of Willard McCarty willard.mccarty_at_kcl.ac.uk>
Date: Sat, 23 Apr 2005 09:04:47 +0100

               Humanist Discussion Group, Vol. 18, No. 731.
       Centre for Computing in the Humanities, King's College London
                     Submit to: humanist_at_princeton.edu

         Date: Sat, 23 Apr 2005 08:22:48 +0100
         From: Norman Gray <norman_at_astro.gla.ac.uk>
         Subject: Re: 18.720 nomenclature


> From: Joris van Zundert <joris.van.zundert_at_gmail.com>
>Might it be that we feel we need a *defining* name
>because in most cases we're not yet accepted as an integral and standard
>part of the instrumentation in the field of Humanities?

The situation isn't necessarily qualitatively different in the physical
sciences Joris mentions. Though using machines and writing software is
pervasive, anyone who gets too interested in computing is perceived as
flirting with grave professional -- and moral -- danger.

>A point I'd like to contribute on this issue is
>that I can hardly think of other disciplines concatenating 'Computing' to
>it's field name, although computing might be heavily involved. There's (to
>my knowledge) no such thing as "Physics Computing", "Digital Resource for
>Chemistry" or "Computational Archaeology".

There are quite a few `Computational X' sub-displines in the physical
sciences. I can think of Computational Chemistry (smells without the mess,
where you do your chemistry by direct simulation of molecular bonds using
quantum mechanics), Computational Fluid Dynamics (CFD: modelling air or
fluid flow without plasticene and wind tunnels), and Computational
Cosmology (working round the irritatingly single status of our universe,
and seeing what other alternatives might possibly exist). In a separate
category, I think, are Lattice QCD (particle theory done by direct
simulation rather than analytical approximation) and possibly Solar System
Dynamics (the production and use of ephemerides which, unlike the other
ones, doesn't require spectacular machines).

The three Computational-X disciplines can be linked, since in each case the
computing aspect is not the centrally interesting bit, and though the
techniques involved are specialised and sophisticated, the technique is no
more than that, a technique for researching chemistry, fluid mechanics or
cosmology, done by a particular type of chemist, engineer or cosmologist.

For Lattice QCD and for some aspects of the ephemeris work, on the other
hand, the calculational technique is itself the object of the
research. For example, I was once subjected to a set of lectures on
Lattice QCD, which was deemed Good For Me even though I was unlikely ever
to do such a calculation for myself; I do not, however, have the first idea
how to simulate a molecule from first principles, and nor does anyone think
I ought to have, as part of my general education.

Joris mentioned Bioinformatics -- this is different again. Unlike each of
the topics above, bioinformatics is of interest to computer scientists:
they're pleased to have the right sort of problem to tax their shiny new
ideas, and the geneticists are reluctantly coming round to the idea that
someone may have a better approach to their problem than they have come up
with themselves.

Melissa Terras listed a few possible terms, with interesting shades of
distinction between them. But here's another one: what would Computational
Humanities look like? It's humanistic research that you do when you happen
to have been given a week of supercomputer time.

An example of that might, for example, be the task of generating some
monster concordance of the entirety of Google's corpus (presuming this were
somehow useful). That would throw up all sorts of technical problems, from
normalisation and tagging of the sources, down to the boring complications
of cramming the work into only a terabyte of physical memory. These would
probably be fascinating to the pale souls doing the actual work, but be
neither the point not seen as much worth reporting. This is possibly
closest to Melissa's `Computing in the Humanities', with the vaguely
pejorative connotations she detects in `Digital Humanities'.

`Cultural and Heritage Informatics', in Melissa's construction, is the one
that sounds most like Bioinformatics to me, inasmuch as it is an example of
some mutual backscratching. the Bio- and the -Informatics people are happy
to have found each other, and are having a great time helping each other
along the road, but it's not a long term thing and I myself doubt that a
new, independent, discipline is emerging here (hmmm, Bioinformatics as
Brief Encounter; see also <http://dx.doi.org/10.1002/cfg.442>).

Last week (from 18.699) Willard's exam question prompted a number of
responses. These were very various, but they were fairly uniformly not
(just) about theories of critical reading, and they weren't (just) about
{SG,X}ML, but seemed instead to be responses about reading texts in general
informed by a particular intellectual background, and particular
obsessions. They're also interesting and productive outside the field of
`Humanities Computing'.

None of this goes against the grain of what others have said here before on
many occasions. But to return to Joris's original point, it seems to me
that what is important is not whether `the heavy use of computation and
computing in these disciplines is simply [...] accepted as part of the
discipline's standard instrumentation' -- since the Humanities' reliance on
word processors is arguably different from this only in degree. Lattice
QCD is something rather distinct from the Computational-X fields, and is a
case where we don't simply see computers shovelled in, facilitating a
pre-existing set of research questions, but instead see a new set of
generally-interesting ideas emerge. Significantly, I think, in the term
`Lattice QCD' there is _no_ explicit mention of computation, even though
novel mechanical and mathematical calculational techniques are at the very
heart of what it is about.

So I have no suggestions for a better term for `Humanities Computing', or
improvements on any of Melissa's terms, but I can predict that if a new
room does have to be built in the academy, it won't have `comput-' or
`info-' anywhere on the nameplate.

Best wishes,


Norman Gray  :  Physics & Astronomy, Glasgow University, UK
http://www.astro.gla.ac.uk/users/norman/  :  www.starlink.ac.uk
Received on Sat Apr 23 2005 - 04:32:55 EDT

This archive was generated by hypermail 2.2.0 : Sat Apr 23 2005 - 04:32:55 EDT