On Tuesday, May 25, 2004 at 18:05:36, Milan Zamazal wrote: > >>>>> "RW" == Robert Widhopf-Fenk <[EMAIL PROTECTED]> writes: > > RW> Well it is unemacsish, but inconvenient and confusing? > > If it is unemacsish then it is inconvenient and confusing ;-).
Well Emacs is not intuitive for a lot of newbies and probably I am not a representative emacs user, although I am using and hacking emacs for nearly 10 years now ... > RW> We could argue a lot on that, actually I added them as they > RW> are convenient and intuitive, at least to me. > > Seriously, a typical Emacs user doesn't expect arrow keys performing > non-movement actions. It depends on your perspective of MOVEMENT, by now it is moving to the next logical level. Seriously it is not withing the same buffer, i.e. it is changing the content. But how often do you want to copy a region from the *tla* buffer. IMHO only when you want to invoke tla by hand, but that means xtla lacks a feature. > It's OK if you feel those bindings convenient, but make > them optional, bound only when explicitly requested. Or make it optional to unbind them. We should honor the majority of users. This would require a majority to vote, but by now we are just a group of unrepresentative hackers. I get the feeling this is a discussion like the thread on empty commits I started on gnu-ach.users ... > An important Emacs feature is that it tries to be "safe" > and nonaggressive to a user by default. IMHO the term AGGRESSIVE would fit an destructive operation, I cannot see it here, nothing is destroyed. Probably we are using xtla in totally different ways. Would be cool to see both of use performing the same tasks and see the different approaches. > Advanced, possibly > confusing features, are always optional. > > RW> Then ALSO up/down should be unbound, as they are not moving > RW> up/down a line but to the next item? > > No, I can't see any serious problem with it, since they are movement > commands <beeing-picky-here> How do you select the name of an archive? The cursor will be placed on the location line when using up/down, this is not what an typical emacs user will expect, he will expect to move one line up/down. All other modes I know do not modify cursor movement. </beeing-picky-here> > and standard things like dired or the completion buffer > work in a similar way. In calendar-mode it is not that way! And also not in completion buffers, rather than moving for a single char the selection moves to the next logical item! > RW> IMHO would be more convenient to allow a binding copying the > RW> THING at point to the kill-ring, no need for you to set the > RW> mark, move around and add it to the kill-ring. > > It might be convenient and I'd like to have it, but anyway a user > should be allowed to perform killing (and other common commands) in > the same way as in any other read only buffer. Killing is not modified, just cursor movement. > Another nice Emacs feature is that you always work with > buffers and can operate on all of them in the same way. > User interface unification is a very important Emacs > feature. Well you mean most of the time, but not always. My mood on xtla is, that it should make everyday life with tla easier, i.e. the fewer keys I need for doing daily work and the more intuitive bindings are the better it will be. The despotic answer is, Steve said he did not like it, so I will make it optional. Cheers Robert
