Re: Anthropomorphism is bad because...

Subject: Re: Anthropomorphism is bad because...
From: Janice Gelb <janice -dot- gelb -at- oracle -dot- com>
To: techwr-l -at- lists -dot- techwr-l -dot- com
Date: Wed, 23 Jun 2010 15:59:12 +1000

On 23/06/10 02:06 PM, Lauren wrote:
> On 6/22/2010 3:42 PM, Janice Gelb wrote:
>> I have no problem with a system "waiting." I have a
>> problem with implying that the system *knows* it's
>> waiting ("expects," "anticipates"), tapping its
>> foot impatiently, no doubt :->
> I think of the terms "expect" and "anticipate" as implying that the
> system may perform any number of processes in anticipation, er, of
> waiting with requirements, of data. My email program "expects" me to
> enter some text or characters in the email message window and then it
> expects me to click send or close the window and abort the message.
> While my program waits, it saves a draft of my message, which it
> overwrites during subsequent saves. If I click "send" and I have not
> entered text or characters in the message window, my email program will
> display a message window that I need to enter something in the message
> area or click, "OK," to send the message, anyway. There is an
> expectation here.

There is no expectation here on the part of your
email program. Your email program does not think.
It does not expect anything. When you click Compose,
a message window appears. Nothing else will happen
until you respond in some way. Developers have
programmed various responses to your behavior. If
you do a certain thing, the program will respond
a certain way. It doesn't "expect" anything. It
does not consider behavior from you that results
in sending a message as meeting an expectation or
behavior from you that does not result in sending
a message (trying to send a blank message, for
example) as not meeting an expectation. It merely
responds to what you provide to it.

> There is also an expectation in more complex systems that will perform
> more complicated processes in the event of bad data. Many systems
> include complex data handling programs that check or sometimes repair
> data for compliance with requirements. To say that a system "requires"
> data does not encompass systems that perform processes without user or
> administrator intervention. Requirements are a part of these systems,
> but what about the automated processes that a failure to meet
> requirements will trigger?

Again, the program is only reactive. It
responds in certain ways to certain events.

> Strict avoidance of anthropomorphism tends to produce a language that
> treats non-humans as things devoid of functions independent of human or
> instinctive control. If I say that my dog expects a treat when I ask my
> dog to perform a trick, then my dog *does* expect a treat. The
> anti-anthropomorphist may say that the dog has been conditioned to
> accept a treat after performing a trick. Now it is easier to say that
> my computer, rather than my dog, does not "expect" certain data of a
> certain type because my computer requires human control in order to
> operate.

Your dog is cognitive. It can reason. Computers can't.

> I think that, for
> lack of computer-specific terms, we need to use terms that convey more
> than what is conveyed by static anthropomorphistic terms. There are no
> middle-ground, computer system-specific terms that are easily understood
> by average computer users, so the terms "expect" and "anticipate" should
> probably take on new meaning in the field of computer systems.

I think that using anthropomorphic language for
systems is more confusing than helpful most of
the time, especially for inexperienced users.
It reinforces some of the crazier help-line
questions that you often see, where people are
convinced that the computer is deliberately
provoking them or not doing what they want on
purpose, even though they haven't clearly indicated
what they want in a way that the system can understand.
I think that the clearer we can be about how systems
and programs actually work, the more users might start
to understand how to interact most beneficially with them.

-- Janice

Janice Gelb | The only connection Oracle has with
janice -dot- gelb -at- oracle -dot- com | this message is the return address

Gain access to everything you need to create and publish information
through multiple channels. Your choice of authoring (and import)
formats with virtually any output. Try Doc-To-Help free for 30-days.

You are currently subscribed to TECHWR-L as archive -at- web -dot- techwr-l -dot- com -dot-

To unsubscribe send a blank email to
techwr-l-unsubscribe -at- lists -dot- techwr-l -dot- com
or visit

To subscribe, send a blank email to techwr-l-join -at- lists -dot- techwr-l -dot- com

Send administrative questions to admin -at- techwr-l -dot- com -dot- Visit for more resources and info.

Please move off-topic discussions to the Chat list, at:


RE: Anthropomorphism is bad because...: From: Chris Despopoulos
Re: Anthropomorphism is bad because...: From: Janice Gelb
Re: Anthropomorphism is bad because...: From: Lauren

Previous by Author: Re: Anthropomorphism is bad because...
Next by Author: Re: Anthropomorphism is bad because...
Previous by Thread: Re: Anthropomorphism is bad because...
Next by Thread: Re: Anthropomorphism is bad because...

What this post helpful? Share it with friends and colleagues:

Sponsored Ads