Humanist Discussion Group, Vol. 36, No. 235.
Department of Digital Humanities, University of Cologne
Hosted by DH-Cologne
www.dhhumanist.org
Submit to: humanist@dhhumanist.org
[1] From: maurizio lana <maurizio.lana@uniupo.it>
Subject: Re: [Humanist] 36.231: the general-purpose machine and unlimited growth (23)
[2] From: Tim Smithers <tim.smithers@cantab.net>
Subject: Re: [Humanist] 36.228: the general-purpose machine and unlimited growth (153)
--[1]------------------------------------------------------------------------
Date: 2022-11-03 10:03:42+00:00
From: maurizio lana <maurizio.lana@uniupo.it>
Subject: Re: [Humanist] 36.231: the general-purpose machine and unlimited growth
Il 03/11/22 07:55, Humanist ha scritto:
> i am an addicted to zotero so i don't say this to imply a critique,
> rather a constatazione
sorry for the italian word which remained!
"i am an addicted to zotero so i don't say this to imply a critique,
rather an observation"
Maurizio
------------------------------------------------------------------------
everybody knows that the war is over
everybody knows that the good guys lost
everybody knows the fight was fixed
the poor stay poor, the rich get rich
l. cohen, everybody knows (1988)
------------------------------------------------------------------------
Maurizio Lana
Università del Piemonte Orientale
Dipartimento di Studi Umanistici
Piazza Roma 36 - 13100 Vercelli
--[2]------------------------------------------------------------------------
Date: 2022-11-03 09:04:09+00:00
From: Tim Smithers <tim.smithers@cantab.net>
Subject: Re: [Humanist] 36.228: the general-purpose machine and unlimited growth
Dear Willard,
In two parts.
Part one: on computers and automation
Bias and prejudice are, I would say, amongst the more recent
things we've automated with computers. And this too seems to
have surprised some people.
Some things we've still not automated, important things, I'd
say, are sound critical thinking and good judgement. These, I
would add, with intended provocation, cannot be computed. If
they could be, perhaps we wouldn't now have the automated bias
and prejudice displayed by plenty of systems built with (so
called) Machine Learning techniques. ['Machine Learning' is a
prejudice laden term.]
But that doesn't stop people trying to do this. See, for
example, something I saw today:
Artificial intelligence is used for predictive policing in
the US and UK – South Africa should embrace it, too
Omowunmi Isafiade, The Conversation, 24 October, 2022
<https://theconversation.com/artificial-intelligence-is-used-for-predictive-
policing-in-the-us-and-uk-south-africa-should-embrace-it-too-191266>
where fair, justified, and explainable, critical judgement of
guilt before a crime is committed is here turned into a
computed probability estimate of a person committing the
crime, using what is called "An Enhanced Naïve Bayes Model for
Crime Prediction."
I wonder what's enhanced here, the naivety of the model, or
the naivety of the authors?
Am I the only one frightened by the people who do this kind of
automation?
Part 2: On what's a junior scholar to do?
First, don't do what your supervisors tell you to do: "publish
whatever you can on whatever you've got, else you'll never
make it". Second, find the researchers and scholars in your
field who demonstrate they do care about others, and, in
particular, about PhDers and young scholars. (Again, this is
intended to be a provocation.)
For those who care, a research publication is built for the
benefit of the intended readers, and not as something needed
on the author's CV, nor as something to increase their
h-index.
A well prepared research publication networks well with the
related and relevant literature, or literatures, it is
supposed to be a part of. At least, this is what I tell the
PhDers I teach.
A reader will then be able to dig deeper, find help to
understand what they don't understand, go back to where the
authors started, read on to where things might go, see where
to find some different and disagreeing alternatives, and so
on, using the networking pointers and links the publication
helpfully integrates, using citations and footnotes. This
further digging, of course, all takes time and effort --
reading the one publication cannot do all this -- and how much
time and effort it needs will depend upon the background,
interests, understanding of each reader, but a good research
publication will make this all possible to do. This usually
needs plenty of well judged and critically selected pointers
and links, so that, as Maurizio urges, quantity is not put in
place of quality. A good research publication should tell
you, the reader, something worth knowing, and be a device with
which to further navigate the relevant, related, and
worthwhile, literature(s), and thus to discover what more is
out there. Sometimes, this is the best thing a research
publication does for you. And that, I think, is how it should
be, for the young and the old.
I'll go and park both my Hobby Horses now.
Best regards,
Tim
> On 2 Nov 2022, at 07:38, Humanist <humanist@dhhumanist.org> wrote:
>
>
> Humanist Discussion Group, Vol. 36, No. 228.
> Department of Digital Humanities, University of Cologne
> Hosted by DH-Cologne
> www.dhhumanist.org
> Submit to: humanist@dhhumanist.org
>
>
>
>
> Date: 2022-11-02 06:32:55+00:00
> From: Willard McCarty <willard.mccarty@mccarty.org.uk>
> Subject: general purpose and unlimited growth
>
> In "Computer science and education" (1969), pioneer computer scientist
> George Forsythe observed that "computing is rapidly invading almost
> every aspect of our intellectual and technological life." He then went
> on to proclaim that, "Indeed, the question 'What can be automated?' is
> one of the most inspiring philosophical and practical questions of
> contemporary civilization." (Information Processing 68, Amsterdam, p.
> 1025) Newspaper evidence suggests that people were still being surprised
> by the idea for some time afterwards. Two aspects of Forsythe's
> statement interest me. First is the irony that as this machine was
> opening up seemingly endless possibilities, it was at the same time
> closing down others as more and more people accepted the notion that
> what couldn't be computed wasn't terribly important and would after a
> time be accommodated by what Herbert Simon called 'satisficing'. The
> second is the delusion of unlimited possibilities itself, of the
> unlimited 'growth' that is now causing us to reach "the limit situation"
> (Lagerkvist 2022).
>
> Some days or weeks ago, Jerome McGann pointed to a limitation of
> Lagerkvist's fine book, that it doesn't say much at all in answer to
> Lenin's useful question, "What is to be done?" John Lanchester's novel
> The Wall (2018) gives us a version of a common reaction to that
> question. But I want to pose a different version of the problem of
> unlimited growth, namely of data--and let me limit this to useful
> scholarly data--and furthermore to the kind one uses by reading it
> rather than applying statistical tests.
>
> With many years as a scholar, a good reputation and respect, one can
> make statements without footnotes, at least in some circumstances, but
> in general what does one do to remain helpful while doing new things? I
> find the number of footnotes growing (e.g. over 100 in a 22-page
> single-spaced paper). These serve two purposes: to reassure my readers
> that there are other respected scholars who would back me up if asked,
> and that here is where you go if you want to find out more. An
> interdisciplinary scholar has an obligation to point the way, no?
>
> Thoughts on this? Responding to the problem by narrowing down ever more
> seems to me a BIG mistake. But what is a junior scholar to do and survive?
>
> Comments?
>
> Yours,
> WM
> --
> Willard McCarty,
> Professor emeritus, King's College London;
> Editor, Interdisciplinary Science Reviews; Humanist
> www.mccarty.org.uk
_______________________________________________
Unsubscribe at: http://dhhumanist.org/Restricted
List posts to: humanist@dhhumanist.org
List info and archives at at: http://dhhumanist.org
Listmember interface at: http://dhhumanist.org/Restricted/
Subscribe at: http://dhhumanist.org/membership_form.php
