web use/response time
Gill, Kathy
site:
Guidelines for Multimedia on the Web
http://www.w3j.com/5/s3.nielsen.html
Response Time
Many multimedia elements are big and take a long time to download with the
horribly low bandwidth available to most users. It is recommended that the
file format and size be indicated in parentheses after the link whenever you
point to a file that would take more than 15 seconds to download with the
bandwidth available to most users.... [snip]
The 15-second guideline in the previous paragraph was derived from the basic
set of response time values that have been known since around 1968. System
response needs to happen within about 10 seconds to keep the user's
attention, so users should be warned before slower operations. On the Web,
current users have been trained to endure so much suffering that it may be
acceptable to increase the limit value to 15 seconds. If we ever want the
general population to
start treating the Web as more than a novelty, we will have to provide
response times within the acceptable ranges, though.
<end snippet>
That said, I think most WWW users will wait more than 10 seconds if
*something* is happening [ie, they can see a download occurring].
I still subscribe to the "hold your breath" test, however. ;-)
Kathy
Kathy E. Gill
AT&T Wireless, NBO Deployment
Redmond Town Center, Bldg 4, 3155B
425.580.6459
mailto:kathy...@attws.com
You can discover more about a person in an hour of play than in a year of
conversation. -- Plato
JQ Johnson
> The 15-second guideline in the previous paragraph was derived from the
> basic set of response time values that have been known since around 1968.
> System response needs to happen within about 10 seconds to keep the user's
> attention
I did some human factors work in the mid 70s and looked closely at that
research at the time (it was still fairly new then!). The results were all
over the map even then. One big issue was the size of a transaction that
was being responded to. 10s is wildly overstating most of the results --
with a typical form-oriented transaction (user fills out a screen on a 3270
terminal and presses send) one could find that 10s was acceptable, but most
of the studies found that there was severe degradation in attention and
user satisfaction at as little as 3s. The studies also found that
predictablilty (variance in response time) was much more important to user
satisfaction than actual response time -- people were willing to wait if
all transactions took 10s, but if half of them took 1s then any long
transaction was very frustrating.
For other kinds of response, e.g. keyboard echo time in a full duplex
typing environment, the acceptable response time was much lower.
What would you think of a guideline that required that web pages load in
3s? Not very practical given current techology. Or how about a tool that
delayed all loading to a minimum of 10s to reduce variance?
I'm not convinced that we can learn anything useful to the web from these
old studies. Although one should respect Nielsen's intuitions on this, I
think that's all they are. His practical recommendations for UI design are
very well thought out, but one should not accept a magic number here, be it
3s, 15s, or 30s.
What the acceptable response time is for web sites remains an open and
interesting question.
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
> From: JQ Johnson[SMTP:j...@DARKWING.UOREGON.EDU]
>
> Kathy Gill quotes Nielsen:
>
> > The 15-second guideline in the previous paragraph was derived from the
> > basic set of response time values that have been known since around
> 1968.
> > System response needs to happen within about 10 seconds to keep the
> user's
> > attention
>
> I did some human factors work in the mid 70s and looked closely at that
> research at the time (it was still fairly new then!). The results were
> all
> over the map even then.
>
I hope I didn't imply that the paper was definitive -- it was the only one I
found doing a quick search.
I think the more important issue is of Feedback and predictability (your
point). The feedback mechanism is now built into Windows/Mac interface with
the "clock" that ticks away while you wait for something to happen. We don't
have that on the web (yet).
> What would you think of a guideline that required that web pages load in
> 3s? Not very practical given current techology.
>
I agree that's not practical -- didn't I adhere to my "hold your breath"
test in my note. <smile>
> What the acceptable response time is for web sites remains an open and
> interesting question.
>
>
I asked Keith Instone, who maintains the Usable Web site
(http://www.usableweb.com/) if he knew of any studies off the top of his
head that looked at this issue. He said 'no' (he's also list mom for the ACM
CHI-Web list).
He does have a "speed" section, http://usableweb.com/items/speed.html .
Primarily reports opinions/articles not research, though, except for this
one from CHI 97:
Internet Delay Effects: How Users Perceive
Quality, Organization, and Ease of Use of Information
http://www.acm.org/sigchi/chi97/proceedings/short-talk/als2.htm
It shows a definite correlation between "negative perception" and long wait
times [mean: 6750 msec] for pages to load compared to short wait times
[mean: 575 msec].
Note, this was 1997.
And AnchorDesk (also 1997)
Avoid Number 1 Web Site Sin: Slow Loading Pages
http://www.zdnet.com/anchordesk/story/story_1244.html
Kathy