Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!mips!apple!amdahl!key!jsp
From: jsp@key.COM (James Preston)
Newsgroups: comp.sys.mac
Subject: Re: Xerox sues Apple!!!
Message-ID: <1328@key.COM>
Date: 20 Dec 89 22:16:04 GMT
References: <33269@ucbvax.BERKELEY.EDU> <10673@encore.Encore.COM> <2676@aecom.yu.edu> <994@biar.UUCP>
Reply-To: jsp@penguin.key.COM (James Preston)
Organization: Key Computer Laboratories, Fremont
Lines: 22

In article <994@biar.UUCP> trebor@biar.UUCP (Robert J Woodhead) writes:
>As for heiroglyphics, lets talk about the UNIX or DOS ones.  "edlin", "mv",
>"subst" "nlsfunc" "chcp" or the options to unix "find".  Icons are easier
>to recognise; numerous perceptual studies have proved this.

Be very, very careful of over-generalizations.  Yes, studies may have shown
that in _many_ cases, icons are easier to recognize, especially when pitted
against typical unix over-abreviations.  But take another look at a typical
Mac paint program.  Tell me that you knew immediately--without reading any
documention and without any help from other mac users and without having
seen any usage of any similar program before--that the little tipping can 
with the stuff pouring out of it was used to fill an area with a pattern.
Tell me that replacing that cute little picture with the words "fill area" 
wouldn't be more straightforward.  (This is probably not even the best
example of "undecipherable icon whose meaning could be better conveyed
with a couple of words", but it's all I can think of right now.)

My point is only that, as with anything, icons can be overdone.  Anytime
you get so focused on ONE way of doing things, you risk overlooking that
in some cases, another way might be better.

--James Preston