Aucb.157 fa.editor-p utzoo!decvax!ucbvax!C70:editor-people Thu Dec 10 12:57:42 1981 Re: Cross-Product Languages >From Moran@PARC-MAXC Thu Dec 10 12:55:12 1981 DLW: If a user really can really push Control and Meta together, ie, as a simple chord, then I think that this action can be approximated as a single keystroke. So, Control-Meta-D would be two keystrokes -- for a user who has developed the skill to do this chording, which is not part of normal typing skill. Of course, typing n Meta-Control-F's in a row takes n+1 keystrokes, since the Meta-Control only has to be pushed once (or n+2 for a user who cannot chord). JQJ: We are talking only about speed, not ease, here. And we are talking only about skilled users. Also, we are only considering the time to execute physical actions (key presses) and ignoring any mental time. A skilled keystroke sequence will probably require mental preparation operations, which we call M operations in our model. Since these are about 1.5 sec a crack (ie, equivalent to about 5-8 keystrokes each), it is important to consider whether a keystroke sequence requires them (clipping a keystroke or two at the expense of incurring an M operation is not a good tradeoff). Check the Keystroke-Level Model paper for details about the heuristics for determining where M operations go in a command sequence. The notion of "ease" is, I would assert, more bound up with these mental operations than with the physical keystroke operations. But ease is not an operationalized concept and seems to refer to several underlying psychological factors. For example, I intuitively feel that a single touch-typing keystroke command is qualitatively "easier" than a two-keystroke command, far beyond the extra keystroke time (and there is usually no extra mental operation required for a two-keystroke command). GS: (1) Your points about EMACS are interesting. The notion of consistency of commands is quite subtle, and we don't have a good theory. I know of only a couple of papers proposing psychologically-oriented grammars that might capture some of the notion. The grammatical notion of consistency usually only deals with commands as content-free symbol (for action) sequences. What your Transpose-Word example points out that there is also an aspect of a command sequence being "consistent" with the task it is supposed (assumed) to perform. The tradeoffs between the various aspects of consistency are not understood. (2) In skilled typing, the fastest keystrokes occur between hands (about 25% faster than than keystrokes within hands). Typing the same key twice is basically no faster than typing any other within-hand keystroke pair. Tom Moran ----------------------------------------------------------------- gopher://quux.org/ conversion by John Goerzen of http://communication.ucsd.edu/A-News/ This Usenet Oldnews Archive article may be copied and distributed freely, provided: 1. There is no money collected for the text(s) of the articles. 2. The following notice remains appended to each copy: The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996 Bruce Jones, Henry Spencer, David Wiseman.