Text 5915, 98 rader
Skriven 2005-07-11 20:44:24 av Ellen K. (1:379/45)
   Kommentar till text 5834 av Rich (1:379/45)
Ärende: Re: Productivity
========================
From: Ellen K. <72322.1016@compuserve.com>

Definitely agree about response time being the way a user measures his or her
computer's speed.  Which is why if I have a database design issue where I can
make the returned information faster vs making the write faster, I'll go with
the former.  The user doesn't know and doesn't care how long it takes the data
to get written if the difference is milliseconds, but they do notice the
milliseconds when they're waiting for their form to populate or whatever.

For example, I made a stored procedure that does a hierarchical search using
various criteria (from more to less likely) to see whether a caller is already
in our database -- the fifth and last rung is soundex first name, soundex last
name, soundex city, and I denormalized the table to put the soundex values in
actual columns AND put an index on the combination..  The result is an
instantaneous answer even when trying all five rungs.  Who cares if it takes a
couple extra milliseconds to do the write!

On Sat, 9 Jul 2005 22:09:13 -0700, "Rich" <@> wrote in message
<42d0ac68$1@w3.nls.net>:

>   They don't have to understand how things work to appreciate the difference.
>
>   My father is similar to the user you remember.  He doesn't understand why
his computer is slow though he knows he wants a faster one.  I have no idea why
but he thinks that more free disk space will help performance.  He switched
from dial-up to DSL because someone told him it would make things faster which
it would if the network was his bottleneck.  In his case the issue is RAM which
he doesn't really understand.
>
>   It's not that his computer is that slow.  It is a little slower than it was
in the past because he is running more software now.  Still, it's running at a
speed with which he used to be satisfied but seeing other faster computers has
planted the seed of desire in his mind.
>
>  Anyway, my point is that my father has no idea why his computer performs the
way it does but he knows that a new computer would perform faster and it would
because new computers have more RAM and faster components then his current
computer.  Specifically, my father's complaints about performance all related
to response time.  It doesn't matter that he composes and reads email or web
pages at the same speed he always has.  What he notices is the time from his
click to completion of the operation whether starting an app or clicking on a
link in a web page.
>
>Rich
>
>
>
>  "Ellen K." <72322.1016@compuserve.com> wrote in message
news:bs91d1tlmak2r8o791s84dcn92bfktv76f@4ax.com...
>  But the average user has no idea what makes "the computer" faster or
>  slower.   I remember a user at a client a few years ago saying he wanted
>  to get a new computer because surfing the net was too slow.  I explained
>  to him that how fast he would see a web page had nothing to do with the
>  "speed" of his computer, but I don't think he believed me.
>
>  On Fri, 8 Jul 2005 01:52:31 -0700, "Rich" <@> wrote in message
>  <42ce3d79@w3.nls.net>:
>
>  >   I think you may be mixing two related areas.  I don't believe people
value time linearly.  The difference between one second and two second response
times can be very noticable and have a significant affect on the user's
perception and satisfaction.  In my earlier example I had the user than take 60
seconds on consuming the results before repeating the operation.  At that level
the difference is 61 seconds vs. 62 seconds.  It may be a valid business
decision that 61 seconds is not worth much more in cost than 62 seconds.  The
user of the faster system though may have measurably higher satisfaction.  Many
users, particularly for personal use, may value satisfaction more highly than
you or maybe your satisfaction is high enough.  My point is that this
satisfaction from responsiveness is distinct from the time (limited or not)
spent on a task.  I believe this is one reason people desire and enjoy ever
faster computers.
>  >
>  >Rich
>  >
>  >  "Don Hills" <dmhills@attglobal.net> wrote in message
news:w3hzCtgaXeAb092yn@attglobal.net...
>  >  In article <42ce0b78@w3.nls.net>, "Rich" <@> wrote:
>  >  >   I can't speak for everyone but I expect ordinary users value a CPU
>  >  >for its peak performance and responsiveness not the steady state and
>  >  >throughput.  When you make a claim that some specific speed is all that
>  >  >they need you are making a claim on the value of that user's time.  When
>  >  >a user clicks a button or similar, they want the computer respond
>  >  >immediately.
>  >
>  >  Agreed. "Transient" response as you defined above was my criteria for
>  >  "fast enough" in my post. One reason I've felt no need to upgrade my
>  >  mail/news machine is because it is responds essentially instantly to my
>  >  input. I can be very productive with it, and I do value my time - I have
a
>  >  limited amount of time in a day to spend on mail and news. The faster I
can
>  >  do it, the more value I get from it.
>  >
>  >  --
>  >  Don Hills    (dmhills at attglobaldotnet)     Wellington, New Zealand

--- BBBS/NT v4.01 Flag-5
 * Origin: Barktopia BBS Site http://HarborWebs.com:8081 (1:379/45)