At 12:00 04/01/2003 +0000, you wrote:
> >> I think there shouldn't be any "minor edits" flag at all; we have the
> >> ability to have the software determine if an edit is "minor"; why not
> >> let the software flag a change as minor in the html if only a few bytes
> >> have changed?
> >
> >Richard Wagner was an anti-Semite
> > ->
> >Richard Wagner was not an anti-Semite
> >
> >4 bytes. Minor edit?
>
>The current system also fails to catch that one, since a person making a
>change like that would probably mark it as minor. My proposal to let
>the software do the work is no worse than the current system, and is a
>win from a usability point of view.
Your proposal is, I think, worse than the current system. Unreliable users
still wouldn't be trusted and would still have every edit checked, reliable
users still would be trusted and have every edit ignored by default. But
with your proposal, large edits (in terms of sheer size), which might
actually be very minor (lots of spelling fixes, table formatting, restoring
vandalised articles, etc), would not, in fact, be marked minor. And
automatically marking small edits is minor is surely a bad idea - it would
be worse than the current system because people who are honest about such
small (in terms of size) changes actually being major would have their
edits automatically marked minor as well.
It might well be "a win from a usability point of view" inasmuch as one
doesn't have to click on a pesky check box if one's edit is minor, but this
is surely outweighed by the fact that those "M"s on Recent Changes would
all become unreliable (instead of just some of them being unreliable).
LP (camembert)