web user & download time
Laurie Hathman
will wait for a page to download before giving up and going somewhere else.
I have checked Search Engine Watch and done searches in periodical indexes
but I cannot find this exact info. If anyone has a lead on this info, I
would appreciate a response.
Thanks,
Laurie
Laurie Hathman
Head of Public Services
Greenlease Library
Rockhurst College
1100 Rockhurst Road
hat...@vax1.rockhurst.edu
816-501-4655 (voice)
816-501-4666 (fax)
Gerald E. Boyd
>I am looking for statistics or other information about how long a web
user
>will wait for a page to download before giving up and going somewhere
else.
> I have checked Search Engine Watch and done searches in periodical
indexes
>but I cannot find this exact info. If anyone has a lead on this info, I
>would appreciate a response.
10-15 seconds max.
Art and the Zen of web sites (under 15 seconds)
http://www.tlc-systems.com/webtips.shtml
See "A Tip" just under halfway down the web page
Improving Web Site Usability and Appeal (10 seconds)
http://msdn.microsoft.com/workshop/management/planning/improvingsiteusa.
asp
See Wait-time feeback halfway down the page
Improving HTTP Latency (causes and effects)
http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/DDay/mogul/HTTPLatency.html
AN EXPLORATION OF DYNAMIC DOCUMENTS (12 seconds or so)
http://home.netscape.com/assist/net_sites/pushpull.html
--
Gerry Boyd -- gb...@netcom.com
Tery Griffin
>I am looking for statistics or other information about how long a web user
>will wait for a page to download before giving up and going somewhere else.
> I have checked Search Engine Watch and done searches in periodical indexes
>but I cannot find this exact info. If anyone has a lead on this info, I
>would appreciate a response.
Child, D. A. (1998). The Effects of System Response Time on User Behavior in a Hypermedia Environment. Journal of Educational Multimedia and Hypermedia, 8(1), 65-87.
---Tery
___________________________________________________________
Tery Griffin
E-mail: tag...@is.nyu.edu
Web: http://pages.nyu.edu/~tag211/
Doctoral Candidate in Educational Communication and Technology
Research Assistant and Instructor,
Higher Education Opportunities Program (internal mail code 4755)
New York University
____________________________________________________________
Walt Howe
>At 10:06 AM 5/3/99 +0100, Laurie Hathman wrote the following:
>
user
>will wait for a page to download before giving up and going somewhere
else.
> I have checked Search Engine Watch and done searches in periodical
indexes
>but I cannot find this exact info. If anyone has a lead on this info, I
>would appreciate a response.
For another look at this question, see my "How to Lose Your Web Viewers!" at
http://www.delphi.com/pubweb/gg2.html
Walt Howe
walt...@delphi.com
http://www.delphi.com/navnet/
http://www.delphi.com/pubweb/
http://people.delphi.com/walthowe/
JQ Johnson
that there is none. The general consensus is that willingness to wait is
probably an exponential distribution, but that the parameters depend on a
myriad of things. Anecdotally, I know I'm much more willing to wait when
(a) I'm looking for specific information on a page rather than just
browsing, (b) I'm fairly sure that this site has the information I want,
(c) I know that I'm on a slow link or have other reasons to understand why
the page is taking a long time to load, (d) I get some feedback such as a
partial page, even though I may have to wait for the rest of the page to
download, (e) other pages that are relevant (either that contain similar
info or that I've recently surfed) are also taking a long time, (f) etc.
Note in particular that any study that quotes you a single number like 10s
is grossly oversimplifying the issue. Is that 10s before half of your
potential customers give up, or 10%, or what?
Since this list is mostly librarians interested in information-rich pages
rather than marketing people, the information seeking behavior of surfers
is particularly relevant to us, and produces some surprising results. For
example, take a look at the research done by UIE (Jared Spool's company)
<http://www.uie.com>. They get all sorts of counterintuitive results, e.g.
that links at the end of a long page have greater salience than links at
the beginning. Their paradigmatic task sets Ss to seek for particular
information on a web site, so it's the kind of task that you may find
especially relevant.
JQ Johnson office: 115F Knight Library
Academic Education Coordinator email: j...@darkwing.uoregon.edu
1299 University of Oregon phone: 1-541-346-1746 -3485 fax
Eugene, OR 97403-1299 http://darkwing.uoregon.edu/~jqj/
Gill, Kathy
>
> >I am looking for statistics or other information about how long a web
> user will wait for a page to download before giving up and going somewhere
> else.
>
AOL users seem to have infinite patience. People used to surfing with Lynx,
almost none. ;-)
The rule of thumb I use is the 'hold your breath' test .. will the page
download on a 28.8 modem in the length of time you can hold your breath.
That said -- the It Depends depends a LOT on how valuable the information is
perceived to be, how accessible said information is from other sites ...
lots of intangibles.
General consensus among designers is to minimize download time as much as
possible. This means, clean code, optimized graphics - and graphics used
ONLY to communicate something (not just cause they are pretty).
You might be interested in the CHI-WEB list -- computer human
interaction/web --
http://www.acm.org/sigchi/web/chi-web.html
And here are some usability links:
http://www.usableweb.com/
http://www.dotparagon.com/resources/HCI.html
Kathy