20.334 speed matters

               Humanist Discussion Group, Vol. 20, No. 334.
       Centre for Computing in the Humanities, King's College London
  www.kcl.ac.uk/schools/humanities/cch/research/publications/humanist.html
                        www.princeton.edu/humanist/
                     Submit to: humanist_at_princeton.edu

         Date: Fri, 01 Dec 2006 08:03:10 +0000
         From: Geoffrey Rockwell <georock_at_mcmaster.ca>
         Subject: Re: 20.330 speed matters, but who says?

Dear Willard,

Jakob Nielsen talks about response times for web sites in _Designing
Web Usability_ (New Riders: 2000, pages 42-51. He references the
following study:

Miller, Robert B. (1968). Response Time in Man- Computer
Conversational Transactions. Proceedings Fall Joint Computer
Conference 33(part 1). AFIPS Press, 1968. 267-277.

Neilsen has an excerpt from _Usability Engineering_ on "Response
Times" at http://www.useit.com/papers/responsetime.html

Searching for this in the ACM Portal I came across the following
paper by T. W. Butler that suggests that Miller's guidelines were
based on anecdotal evidence:

Butler, T. W. 1983. Computer response time and user performance. In
Proceedings of the SIGCHI Conference on Human Factors in Computing
Systems (Boston, Massachusetts, United States, December 12 - 15,
1983). A. Janda, Ed. CHI '83. ACM Press, New York, NY, 58-62. DOI=
http://doi.acm.org/10.1145/800045.801581

Abstract: " Nearly everyone agrees that computer response time is
very important to the users of interactive systems. Many papers have
been written describing the bad effects of computer response times
that are too long or too short, and many sets of "guidelines" for
appropriate human-engineered computer response times in human-machine
systems have been published, as well. Nearly all these sets of
guidelines are direct descendants of the set published by Robert
Miller (1968) about 15 years ago. When Miller wrote his guidelines,
he was quite open in describing them as based only on his experience,
and he called for experimental data that would allow for the
formulation of better, empirically-based rules for setting computer
response time for optimal human performance. About fifteen years
later, these studies are still missing, for the most part. Aside for
the problem-solving studies of Grossberg, et al. (1976), Goodman and
Spence (1981), Bergrnan, et al. (1981), and others, the literature is
sadly lacking in empirical data to support the simplest assertions
about how computer response time affects computer users. Though there
is only the sparsest data to support them, several assertions about
computer response time and user performance have become accepted as
common knowledge."

Ben Shneiderman has a survey article from 1984 that might also be
what you were looking for:

Shneiderman, B. 1984. Response time and display rate in human
performance with computers. ACM Comput. Surv. 16, 3 (Sep. 1984),
265-285. DOI= http://doi.acm.org/10.1145/2514.2517

This can be read off the ACM Portal if you have access as can the
Butler article.

Yours,

Geoffrey R.

On 29-Nov-06, at 1:34 AM, Humanist Discussion Group (by way of
Willard McCarty <willard.mccarty_at_kcl.ac.uk>) wrote:

> Humanist Discussion Group, Vol. 20, No. 330.
> Centre for Computing in the Humanities, King's College London
> www.kcl.ac.uk/schools/humanities/cch/research/publications/ humanist.html
> www.princeton.edu/humanist/
> Submit to: humanist_at_princeton.edu
>
>
>
> Date: Tue, 28 Nov 2006 08:14:28 +0000
> From: Willard McCarty <willard.mccarty_at_kcl.ac.uk>
> >
>Dear colleagues,
>
>Years ago I read a psychometric study of how people use computing
>systems as a function of the time between a user-initiated action and
>the system's response. The conclusion of the study was that if more
>than ca 1.5 seconds elapsed, the delay would begin to affect what the
>user then did. For example, this study suggested, if I put a query to
>a system and more than the threshold delay occurs, then I will begin
>to censor subsequent queries based on some assessment or other that I
>make of the probability of a useful response. The conclusion seems
>quite plausible. Consider, for example, unavoidable delays in
>conversation between saying something and getting a response.
>Spontaneity goes.
>
>My question is, can anyone here give me a reference to this or any
>similar study?
>
>Many thanks.
>
>Yours,
>WM
>
>
>Dr Willard McCarty | Reader in Humanities Computing | Centre for
>Computing in the Humanities | King's College London | Kay House, 7
>Arundel Street | London WC2R 3DX | U.K. | +44 (0)20 7848-2784 fax:
>-2980 || willard.mccarty_at_kcl.ac.uk www.kcl.ac.uk/humanities/cch/wlm/
Received on Fri Dec 01 2006 - 04:06:43 EST