[Humanist] 23.101 programming: the fear of it

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Tue Jun 23 07:14:51 CEST 2009


                 Humanist Discussion Group, Vol. 23, No. 101.
         Centre for Computing in the Humanities, King's College London
                       www.digitalhumanities.org/humanist
                Submit to: humanist at lists.digitalhumanities.org

  [1]   From:    James Rovira <jamesrovira at gmail.com>                      (25)
        Subject: Re: [Humanist] 23.99 programming: the fear of it?

  [2]   From:    Joris van Zundert <joris.van.zundert at gmail.com>           (54)
        Subject: Re: [Humanist] 23.99 programming: the fear of it?


--[1]------------------------------------------------------------------------
        Date: Mon, 22 Jun 2009 02:00:28 -0400
        From: James Rovira <jamesrovira at gmail.com>
        Subject: Re: [Humanist] 23.99 programming: the fear of it?
        In-Reply-To: <20090622055146.356D721FA4 at woodward.joyent.us>

I was an almost 4.0 undergrad student, Willard -- got As in math (up
to first year calculus), the sciences, English, statistics, etc.
Received one grade of B+ in four years of college.  I'm not afraid of
programming.  It's just drudge work.  Mental digging.  There's some
gratification to seeing an end product working well, but the sheer
boredom leading up to that end product isn't worth it.

I understand that those who love writing lines of code can see
"poetic" qualities (broadly defined -- really, I suspect all that's
really being referred to here is simplicity) in writing code, but this
is all hidden to everyone but programmers checking the code.  All that
matters, really, is the functionality of the end product.  An
awkwardly written code supporting a stable system is more important
than a poetically written code that crashes.

Perhaps that's the problem with PCs.  They have poets writing their code.

So I wouldn't assume "fear" on the part of anyone uninterested in
programming.  Some of us feel we have more rewarding work to do.

Jim R

> I'd like to get us to pay more attention to the question of why students of
> the humanities often find programming less than attractive as a subject for
> study -- why the fear of doing it.  I think that the beginnings of a useful
> response require us to recognise that fear is real psychologically and
> useful biologically but that our task is to investigate what it points to.
> How is the translation of programming into the interpretation of poetry
> involved?
>



--[2]------------------------------------------------------------------------
        Date: Mon, 22 Jun 2009 13:35:23 +0200
        From: Joris van Zundert <joris.van.zundert at gmail.com>
        Subject: Re: [Humanist] 23.99 programming: the fear of it?
        In-Reply-To: <20090622055146.356D721FA4 at woodward.joyent.us>


Dear Willard,

But as a
> humanist I want to know about the 1.4% that cannot be handled. For me the
> appeal has to be that a particular skill in capturing things will help me
> get closer, sense more keenly the fleeting transcendence of these things.
>

Totally agree.

But I'd say formalization is not about just doing the homework of
distinguishing between the 98.6% that's normal or regular or non significant
and the 1.4% that's supposedly beyond formal comprehension.

Formalization, I think, is a method (not more and not less) to get a focused
grip on the properties of the object of study. Formalization allows one to
build models; either by algorithm, statistics or if need be pen and paper -
it's the formalization that's key, not specifically that it might be
computable or not. The nice thing of a *formalized* model is that it's
testably descriptive. So if I have a model that I think describes adequately
a phenomenon of text, the formal part of it means that I can repetitively
test the model by letting it be a description of another text that according
to human reader judgment shows the same phenomenon. If the model holds (i.e.
judges the same as the human reader) the model is still adequate. If not, we
can refine the model (or ultimately design a new one) to obtain a model that
apparently is a better descriptor of the phenomenon.

That's all just basic model theory and Turing etc. Now, I'm certainly not
suggesting that modeling should supplant our existing methods, not at all. I
just think formalization and modeling are nice additional instruments that
usher in repeated scientific testing, thereby allowing us to build
descriptive models that we can iterate into adequate theories of textual
phenomena. The nice thing being that the theory would be formally
falsifiable as a general principle. That general principle then allows for
evolving the model further until the point that it might even be descriptive
of parts of that 1.4%.

I think it would actually result in intellectual poverty if we just stopped
at sort of 'grasping' the 1.4% with our intellects, just pondering on it in
our research papers. I'd rather try to venture beyond: do we have methods
that allow us to really explain what's happening within that 1.4%? I don't
feel that's claustrophobic intellectually, I find it challenging.

For the moment though we're not much further than some grammar parsing, some
stylistic statistics and a faint hint of semantics of course. There's
actually a huge intellectual challenge to be found here. Yes, it might be a
different mode of intellectual challenge, specifically not one to everyone's
liking. But an intellectual challenge nonetheless.

Kind regards
-- Joris



More information about the Humanist mailing list