Xref: utzoo comp.sys.mac:15380 comp.windows.misc:496 Path: utzoo!mnetor!uunet!mcvax!ukc!strath-cs!glasgow!gilbert From: gilbert@cs.glasgow.ac.uk (Gilbert Cockton) Newsgroups: comp.sys.mac,comp.windows.misc Subject: Re: 2 button mouse Message-ID: <1013@crete.cs.glasgow.ac.uk> Date: 21 Apr 88 12:52:36 GMT References: <4129@hoptoad.uucp> <283@rhesus.primate.wisc.edu> <1710@ssc-vax.UUCP> <201@glenlivet.hci.hw.ac.uk> <1798@sugar.UUCP> Reply-To: gilbert@cs.glasgow.ac.uk (Gilbert\ Cockton) Organization: Comp Sci, Glasgow Univ, Scotland Lines: 36 Keywords: window human computer interface In article <1798@sugar.UUCP> peter@sugar.UUCP (Peter da Silva) writes: >Why are you always going to run out of mouse-buttons? Excuse my elipsis - if you want to keep a one-to-one mapping between button presses and logical operations (e.g. select, extend-select, move, stretch, pop-up-a-menu, help etc. etc.), then you need a mouse button for every logical operation. If you have less buttons than operations, you will end up with modes and memory problems. The three button mouse workstation I'm using at the moment can get very modey with the meaning of its buttons. So, *for one-to-one mappings* (a good thing *all things being equal*) new operations are bound to exhaust the supply of mouse buttons. Note that no-one with any sense would think that closure is possible with any creatively designed artefact, so we can continue to expect innovation in the fundamental logical operations encountered in graphical interfaces. Again, the advantage of chording with keyboard and mouse is that the supply of 'logical' mouse buttons is increased. The alternative is to have moded interpretation of mouse presses, or to map logical operations onto icons (e.g. zoom box, quit box) or window zones (e.g. move (un)constrained zones, scroll bars). There is a bandwidth problem here too, as eventually windows are going to get cluttered, complicated and intimidating if a growing set of logical operations is forever mapped onto new graphical controls. Less workspace will also become available as the borders of windows grow into complex structures. Although the respondent rarely uses his keyboard, this cannot be the case for programs such as spreadsheets (need the keypad), text processors and programming environments (need the whole keyboard). My feeling is that applications where the keyboard is essential are used for far more person-interaction-hours than applications where a keyboard is not essential. If the chording keys were duplicated down both sides of the keyboard, mouse-keyboard chording could be simplified, indeed you could probably have the necessary keys sticking out of the keyboard garage by the CPU!