CHI-WEB Archives

ACM SIGCHI WWW Human Factors (Open Discussion)

CHI-WEB@LISTSERV.ACM.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Coneti GiriMohan <[log in to unmask]>
Reply To:
Coneti GiriMohan <[log in to unmask]>
Date:
Mon, 21 Jun 1999 10:39:53 -0400
Content-Type:
multipart/mixed
Parts/Attachments:
I fully second Don's thoughts!! I would like to add that metaphors restrict the
represenatation of additional functionalities/features which the product claims
to provide. The fax interface that i have in my home pc, has remote resemblance
to any fax machine. They on the contrary use the common GUI control elements but
have strongly structured the interaction flow which eases the use.
Following a particular metaphor lead to over-expectation wherein the user starts
to believe that the interface can do everything that the real life system would
do.
Sonic's MP3 layer has moved away from the thght of using a pre-conceived
metaphor attached to any sound device. What has resulted from this is the
ability for sonic to incorporate more features outside the metaphor world. But
they have failed to establish visual cues and aids that augment the initial
learning process of an interface that is stunningly appealing. When a new visual
represenattion is introduced , it is neccesary to focus on the initial learning
experience- the first few miniutes are when the user is going thru emotions that
are a mix of excitement, curiosity, trepidation, and anticipated elation. Under
circumstances like these it would be awesome to introduce a sample file of the
process that is more or less a visual walkthru of how the interface works which
could be kicked off the first time the user opens the application, ofcourse an
announcement to this effect needs to be displayed wherein the user is given the
ability to skip this demo process of the sample file.


Don Norman wrote:

> Peter Merholz's comments on "the Invisible Computer" and my argument against
> metaphors are nice, but I obviously need to clarify my position for this
> audience.
>
> I am strongly opposed to metaphors because they lock you in to a false
> representation.  You use a metaphor to get you started because you think it
> aids the uninformed user.  Well, the user is uninformed for only the first
> few minutes and after that, you are stuck with the excess baggage.
>
> I just critiqued a web MP3 player for this reason: they used the metaphor of
> a tape player so they had to draw the slot for the audio tape, wasting
> valuable screen real estate, cluttering up (the already over-cluttered
> graphic), and adding zero useful value expect possible confusion.  And then
> when they wanted to add new functions, they were stuck-- how do we add this
> to the existing player, they asked (which is why they were consulting me).
> I recommended they start over again (I recommended they hire a good design
> firm, and I recommended three to them.)
>
> I am a strong believer in conceptual models - present a clear conceptual
> model of how the system or web page works and make sure you are completely
> consistent with that model forever.
> --
> George Lakoff's work on metaphor is important, but has little to do with how
> to use metaphor in design -- it does not contradict these statement (I have
> discussed this with him).
> --
> The choice of control for loudness, pitch, channel settings, etc are not
> metaphorical - these are representational issues.  These are what I call
> mappings: mapping control of volume or pitch or station to a control device.
>
> Volume and brightness (and color saturation) should always be controlled
> vertically. These are all intensive dimensions. We represent more by higher,
> and more sound or light is a natural representation.  Pitch and station and
> color (hue) are substitutive dimensions. It does not make sense to add more
> hue or more pitch. One station is not higher than another, even if the
> numeral that represents the station represents a larger value.  Here, the
> station number is an ordinal dimension -- it is an identifier, not a value.
>
> Knobs have a peculiar history.  By tradition, clockwise is more, but
> clockwise is a completely arbitrary direction (chosen because of the
> direction a shadow moves around a sundial in the northern hemisphere).  So
> clockwise means more, unless you are dealing with water, where clockwise
> tightens the screw, so closes the faucet, so lessens the flow. Speak of
> complicated conceptual models.  (Unless you have a plumber who is an amateur
> psychologist and thinks faucets should be mirror images, so the left faucet
> is clockwise and the right one counterclockwise, meaning that when you are
> in the shower with soapy eyes, you risk scalding or freezing each time you
> touch a faucet).  (The rationale is that the body is mirror-image
> symmetrical.  Such plumbers should be forced to endure scalding showers in
> hell.)
>
> (I now live in a rented house which took all the bad examples from the
> Psychology of Everyday Things and installed them. Light switches, faucets,
> doorknobs, and even the refrigerator from hell (well, from GE).)  To top it
> off, I bought the new Siemens wireless phone system for $1,000 but had to
> throw it away several days later when my family was reduced to tears and
> aggression because of its complete incompetence ad user interface.)
>
> Someone must be trying to get even with me.
>
> But I digress
>
> So on a remote control saturation, brightness, and loudness ought to be up
> and down.  Station, pitch and hue ought to be something else -- such as
> horizontal, or anything else.  Most remotes get it backwards.
>
> In a map or diagram, color (hue) should never be used to represent amount --
> it is an unnatural mapping.  Brightness or saturation should be used.  Most
> maps get this wrong.
>
> All this is spelled out in numerous good books on representation, including
> my own: Things that make us smart.
>
> Don


ATOM RSS1 RSS2