On Fri, Oct 15, 2004 at 12:44:11PM +0200, Ilya Konstantinov wrote: > There's no curve to learning it as 'rules of thumb'. Our software's > behavior, though, should have as litttle mystery around it as possible.
The MS solution has its own mysteries. See my reply to (4) below. > If we can implement a feature in a non-mysterious way (that is - staying > in the user's metaphors domain), even if it makes our implementation > more complicated, we should stick to the non-mysterious way. > > What *can* you find in the user's metaphor domain, you ask? > 1. All characters are printed (for techies: have non-zero width). Except paragraph break, forced line-break, tab, space and others that can be exposed in "display hidden characters". > 2. Characters are for passing textual information, not for formatting > control. > 3. BiDi direction is a part of "formatting", not a part of the textual > information (he perceives it similar to paragraph alignment). Should the bidi information be saved on plain-text expoert? > 4. "When I switch to Hebrew keyboard, I type right-to-left" However to make that metaphore complete you have to expose the language property of characters. Otherwise the user will eventually be stuck with typing numbers etc. in the wrong language. Typing numbers in the wrong language has caused many strange misformats in the past. -- Tzafrir Cohen +---------------------------+ http://www.technion.ac.il/~tzafrir/ |vim is a mutt's best friend| mailto:[EMAIL PROTECTED] +---------------------------+ ============================================================== To unsubscribe, send mail to [EMAIL PROTECTED] with "unsubscribe hebrew" in the message body.
