[Humanist] 30.803 hands on

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Sat Mar 4 07:50:13 CET 2017

                 Humanist Discussion Group, Vol. 30, No. 803.
            Department of Digital Humanities, King's College London
                Submit to: humanist at lists.digitalhumanities.org

  [1]   From:    David Zeitlyn <david.zeitlyn at anthro.ox.ac.uk>             (20)
        Subject: Re 30.801 hands on! & Beauty

  [2]   From:    Stéfan_Sinclair <sgsinclair at gmail.com>                   (15)
        Subject: Re: 30.799 hands on [Re: Humanist Digest, Vol 102, Issue 3]

        Date: Fri, 3 Mar 2017 15:04:34 +0000
        From: David Zeitlyn <david.zeitlyn at anthro.ox.ac.uk>
        Subject: Re 30.801 hands on! & Beauty
        In-Reply-To: <7e941812-908d-4173-9c32-04f04353e9a1 at HUB02.ad.oak.ox.ac.uk>

Dear all
As a riff on Gabriel Egan's evocation of Rube Goldberg / Heath Robinson 
let me commend to you the video of two Swiss installation artists:

Peter Fischli, and David Weiss, 'Der lauf der dinge' (The life of things 1987) and the role of causality in it. It is a film showing an elaborate causal chain of events. (In 2003 a television advertisement featured a similar chain of events using Honda car parts.) Fischli and Weiss invest domestic artefacts with apparent lives of their own, each object setting in motion the next. The film shows a literal chain reaction. For example, a flame is lit that heats a kettle until it boils; the steam from its spout propels it down a railway track until it meets the next object and sets it moving.

[See https://vimeo.com/175928976]

A strange and and uncanny beauty, a fabricated one at that
best wishes


David Zeitlyn,
Professor of Social Anthropology (research). ORCID: 0000-0001-5853-7351
Institute of Social and Cultural Anthropology, School of Anthropology and Museum Ethnography
University of Oxford
51 Banbury Road, Oxford, OX2 6PF, UK.
http://www.mambila.info/ The Virtual Institute of Mambila Studies

Oct 2015 open access paper 'Looking Forward, Looking Back'

Vestiges: Traces of Record http://www.vestiges-journal.info/ Open access journal

        Date: Fri, 3 Mar 2017 10:56:21 -0500
        From: Stéfan_Sinclair <sgsinclair at gmail.com>
        Subject: Re: 30.799 hands on [Re: Humanist Digest, Vol 102, Issue 3]
        In-Reply-To: <mailman.3.1488538804.2045.humanist at lists.digitalhumanities.org>

Dear all,

I've been hesitating to write, in part because I'm incorrigible lurker, and in part because this thread has wandered in various directions, though I've never known Willard to object to fruitful wanderings. This has grown to be a bit longer of a post, my apologies in advance, and no hard feelings if you scroll quickly :).

I've been especially interested in parts of the discussion that explore code and beauty, in part because I do think that somewhere there is a sometimes useful distinction between some of what happens in, say, CS, and some of the development work that happens in DH (please forgive the awkward attempts to avoid wholesale characterization of either discipline).

I'm not a computer scientist (though often enough I seem to be mistaken for one), but my impression based on considerable interactions is that code structure and form do matter a lot in CS pedagogy and research. Principles of code design can be not only about current best practices but can also be about a sense of aesthetics developed over time and that's often partly language-specific. Of course, having code that can achieve its intended purpose is usually important too.

But my sense of coding in DH is much closer to what Gabriel Egan describes, which I take to have a much stronger *pragmatic* aspect. The Rube Goldberg is a wonderful example. In DH I often think of how Mark Olsen used to describe the wonderful ARTFL resource: it's awful spaghetti code, but it works! And for the user it most certainly did, it was blindingly fast and powerful and enabled all kinds of new research (past tense because I was more familiar with it in the 90s). Arguably, as a DH project, that's what mattered most. Having said that, I do recognize there are good reasons to write clear, sustainable code in larger projects.

However, when I'm teaching coding I tend to ask more "does it do what you wanted" (which is rarely the end of the conversation); it's *not* about clarity, brevity, the use of certain functions, structures, libraries, etc.). I've found that introducing programming theory (or at least my grasp on the relevant parts) is usually counterproductive, at least at first. In fact, I usually don't even structure a programming course on programming concepts (strings, arrays, conditionals, etc.), but around DH tasks (cleaning, tokenizing, counting, etc.), and programming techniques come up on a need to know basis. We haven't fully succeeded in this and it's in need of an update, but if you're interested in a more concrete example of what I mean see http://nbviewer.jupyter.org/github/sgsinclair/alta/blob/master/ipynb/ArtOfLiteraryTextAnalysis.ipynb

I find one of the biggest struggles in teaching project-oriented courses that are heavy on programming is to encourage students to first seek out someone else's code that could be used or adapted; the impulse for originality is deeply inculcated in humanities students and even with their familiar citation practices it can be hard for them to learn to reuse code as much as possible. There's also a catch-22: I don't want to teach programming concepts in the abstract, but it can be difficult to look for other code if you don't understand much of what you see. In fact, I usually spend much time working with students to practice and refine search engine queries to find useful tips and solutions - quite possibly the most useful thing students get from the course.

I realize I'm already rambling on too much, but one final thing if I may: I do a lot of programming as a DHer and I wouldn't want to undevalorise that coding work. I take great personal pride and satisfaction in my craft, not because I write anything like beautiful code, but because some of what I make is useful to others and contributes in an original and scholarly way what might be called the state of the art of digital resource creation in DH (we're very close here to issues of evaluating digital works for hiring, tenure and promotion, needless to say). Coding for me is both creative and scholarly. Sometimes the ends are very modest: a one-liner to pre-process a text in a way that is incompatible with existing tools (or finding and learning the tools would be more time and hassle than it's worth). Sometimes the payoff is not in utility but in the pleasure of impressing a single colleague. And sometime it really is in the satisfaction in knowing other people have found enough merits in your resource to use it. In my mind it's all DH.



Prof. Stéfan Sinclair, Digital Humanities, McGill University
Department of Languages, Literatures & Cultures
Office 341, 688 Sherbrooke St. W, Montreal, Quebec, Canada H3A 3R1
Tel. (1) 514-398-4400 x094950
@sgsinclair http://stefansinclair.name/

More information about the Humanist mailing list