>I recall seeing usability research somewhere to the effect that, as long as
>a page "appears to be doing something", then users perceive it to be
>loading faster and are more likely to wait for it than for a page where
>nothing much happens until every element is loaded. Even if the two
>pages in question in fact take the same time to load. Makes sense I
>guess. I seem to remember seeing a Javascript routine that displays the
>status of the pre-loading process as it happens ("Now loading image1...
>image2..." etc); I guess this is the best approach, eh?
believe it..
there's a recorded case at Microsoft of user focus groups complaining that
Excel "took to long to start up". it was the number one gripe, and people
were unhappy enough about it that boot timing was identified as a major
target for improvements in the next version upgrade.
the problem is that Excel has a huge feature set.. if they wanted to offer
*more* services, they'd have to bundle it with an intern.. and it takes
time to load all the support software. users are rarely willing to give
up features, and even if they are, the only practical way to get a feature
set reduction through marketing is to engrave it on the side of a
high-velocity bullet. the only realistic (sic) option was to improve the
boot technology.
that involved a major redesign of the core software. all the services had
to be rebuilt as thin boot modules which could load additional resources at
need. it took a hell of a lot of work, and the users were even less
satisfied with the results. not only were they still complaining about
the launch taking too long, now the program itself was 'sluggish'.
according to the mythology, the issue was resolved when a programmer forgot
to remove a diagnostic screen from the test code before handing it off to
the users. all the screen did was list the names of the modules as they
were loaded, so the programmers could see which ones were taking the most
time. item by item, the load times were pretty good, but the collection
was still so big that the overall time was considered unacceptable.
unacceptable, at least, until the users saw the names shooting past in the
dialog. then, even though that version took just as many seconds to boot
as an earlier version, the users thought this one booted 'quickly'.
further investigation found that the real problem was bad communication on
the part of the user-test department. the users really just wanted to
know that the machine was actually doing something, and hadn't locked up.
unfortunately, there was nothing like that on the Objective Response Form.
the closest people could get was 'too slow at launch'.
the double-barreled morals are:
1. make sure you actually know the problem before trying to solve it.
2. giving the user feedback is a Good Thing.
mike stone <[EMAIL PROTECTED]> 'net geek..
been there, done that, have network, will travel.
____________________________________________________________________
--------------------------------------------------------------------
Join The Web Consultants Association : Register on our web site Now
Web Consultants Web Site : http://just4u.com/webconsultants
If you lose the instructions All subscription/unsubscribing can be done
directly from our website for all our lists.
---------------------------------------------------------------------
