[Humanist] 24.112 inadequacies of markup

Humanist Discussion Group willard.mccarty at mccarty.org.uk
Tue Jun 15 07:37:00 CEST 2010


                 Humanist Discussion Group, Vol. 24, No. 112.
         Centre for Computing in the Humanities, King's College London
                       www.digitalhumanities.org/humanist
                Submit to: humanist at lists.digitalhumanities.org

  [1]   From:    Richard Lewis <richard.lewis at gold.ac.uk>                  (67)
        Subject: Re: [Humanist] 23.794 inadequacies of markup

  [2]   From:    Richard Lewis <richard.lewis at gold.ac.uk>                  (60)
        Subject: Re: [Humanist] 23.795 inadequacies of markup


--[1]------------------------------------------------------------------------
        Date: Mon, 14 Jun 2010 16:51:13 +0100
        From: Richard Lewis <richard.lewis at gold.ac.uk>
        Subject: Re: [Humanist] 23.794 inadequacies of markup
        In-Reply-To: <20100504055823.E625B53618 at woodward.joyent.us>

At Tue,  4 May 2010 05:58:23 +0000 (GMT),
Humanist Discussion Group wrote:
> 
> 
> --[1]------------------------------------------------------------------------
>         Date: Mon, 3 May 2010 18:02:50 +1000
>         From: Desmond Schmidt <desmond.schmidt at qut.edu.au>
>         Subject: RE: [Humanist] 23.792 inadequacies of markup
>         In-Reply-To: <20100503051446.4628855EC4 at woodward.joyent.us>
> 
>   [2]   From:    Patrick Durusau <patrick at durusau.net>                    
>         Subject: Re: [Humanist] 23.789 inadequacies of markup
> 
> 
>   [3]   From:    maurizio lana <m.lana at lett.unipmn.it>                     (54)
>         Subject: Re: [Humanist] 23.789 inadequacies of markup
> 
> >very good, software the tool allowing to make encoded judgments
> >"dynamically and systematically manipulable"? that is: i get a list
> >of the markup elements which are in a given text, and i choose
> >which ones to see/visualize and which ones to ignore (e.g. i keep
> >the non-interpretative ones and only some of the interpretative
> >ones); or even which ones to delete (e.g. i delete all the
> >interpretative ones).
> 
> Firstly, what kind of manipulations can one make with XSLT? -
> mathematical ones. What if the manipulations I want to make are
> subjective, interpretative, or if I want to replace one set of
> markup with another one, or mix two sets, say a base tag set
> describing text structure with an interpretative set by someone
> else?
> 
> Secondly, XSLT is a programming language. Because it's expressed in
> XML it's also a rather cumbersome one. My concern is why should a
> humanist have to write a stylesheet to manipulate his
> judgements. Well, maybe he/she should learn, but it's not happening
> in any significant numbers, and I don't see how it ever will. The
> digital humanists can do it - probably. But the ordinary humanists
> don't even want to get involved at that level, and why should they?
> Is it the role of 'very good, software' to force the user to write a
> program? In my view it's the goal of very good software to make the
> interface disappear, so that each task of the user is performed by
> the machine automatically with the least effort.
> 
Apologies for jumping on this thread so late. Just wanted to make a
remark about "good software". The criteria by which software is judged
vary considerably. From the point of view of programmers it's the
elegance of well designed code. For sysadmins it's easy installation
and good documentation. For UNIX users, it's correct use of input and
output streams, and meaningful option names. And for novice users it's
intuitive user interfaces. But for everyone (I assume), functionality
is probably the most important criteria. Does the software do what you
want it to?
-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Richard Lewis
ISMS, Computing
Goldsmiths, University of London
Tel: +44 (0)20 7078 5134
Skype: richardjlewis
JID: ironchicken at jabber.earth.li
http://www.richardlewis.me.uk/


--[2]------------------------------------------------------------------------
        Date: Mon, 14 Jun 2010 17:23:45 +0100
        From: Richard Lewis <richard.lewis at gold.ac.uk>
        Subject: Re: [Humanist] 23.795 inadequacies of markup
        In-Reply-To: <20100505052753.A342956D01 at woodward.joyent.us>

At Wed,  5 May 2010 05:27:53 +0000 (GMT),
Humanist Discussion Group wrote:
> 
> 
> --[1]------------------------------------------------------------------------
>         Date: Tue, 4 May 2010 06:24:57 -0400
>         From: "John A. Walsh" <jawalsh at indiana.edu>
>         Subject: Re: [Humanist] 23.794 inadequacies of markup
>         In-Reply-To: <20100504055823.E625B53618 at woodward.joyent.us>
> 
> Another response to Desmond Schmidt (Date: Mon, 3 May 2010 18:02:50 +1000
>        From: Desmond Schmidt <desmond.schmidt at qut.edu.au>
>        Subject: RE: [Humanist] 23.792 inadequacies of markup)
> 
> > Firstly, what kind of manipulations can one make with XSLT?
> > - mathematical ones.
> 
> We are dealing with computers and digital objects. At some level,
> every manipulation done in these environments and with these
> objects is a mathematical one. And if we move beyond that fundamental level,
> I would say that most XSLT manipulations are not, on the surface, mathematical.
> Sure we can add, subtract, divide, and multiply and compare numeric values,
> but more often we are comparing and matching strings and nodes, the
> content and structural elements of the text. Or we are comparing the
> interpretive values we have added to the text.

Again, apologies for being late. I think the idea that computers are
fundamentally mathematical isn't quite true. Turing's aim was not to
build tools for doing mathematics, but to answer the question of what
is an effectively computable number. His subjects of analysis were
"computers", those who carry out computations, and who Andrew Wells
(in Rethinking Cognitive Computation, Palgrave 2005) calls
"computants" (to distinguish them from the modern understanding of
"computer"). But what Turing actually studied was the procedures that
computants carry out and asked how those procedures may be formalised
into irreducible (from the point of view of the computant) steps;
marking a symbol on a tape, erasing a symbol from a tape, moving a
tape to the left or right. It happened to be simple arithmetic
operations that Turing formalised, but there's no particular reason
why mathematics should be privileged in formalised procedure. And much
of real interest in mathematics is often very hard to model with
computers, just as much of interest in other domains of knowledge is
hard to model.

So could it be the case that we could formalise the procedures of
effective text analysis? Try and imagine an answer that doesn't
involve a digital computer. Is *that* humanities computing?
-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Richard Lewis
ISMS, Computing
Goldsmiths, University of London
Tel: +44 (0)20 7078 5134
Skype: richardjlewis
JID: ironchicken at jabber.earth.li
http://www.richardlewis.me.uk/



More information about the Humanist mailing list