> 2. is it useful to be able to work with tags in a relative
> way (eg, viewnext(), etc) as well as an absolute way.
| I was under the impression that the consens is "No".
In all my postings to the list, the subtext is that I'm "thinking aloud" about
if there are any better ways to work, and hoping that someone else may
come up with better ideas than I've got (regardless of mainstream codebase suitability).
| Tags can be
| ordered. However, what we view are subsets of the set of all tags. There
| are (too) many ways to order such subsets. The thing is, none of these
| orderings is outstandingly suitable.
That's the most general mathematical view of tags. What I'm wondering
about is there a way of using tagging that makes sense for a human user
of a window manager that makes what they want to do easier that also
partially "de-absolutises" tagging. (And for
someone who really wishes he was still a mathematician, that's quite
a statement :-) ) As I mentioned, I'm getting quite fond of the `push to temporary tag, pop
when finished' way of working. Indeed, one of the points of view I
tried to put across back when tags were first raised for wmii as
a replacement for workspaces was that with workspaces you `transfer
windows between workspaces', which means if you shift a window
somewhere temporarily it has lost any connection with where it was
originally, so if you want to go back to how things were you have to remember
where it came from and move it yourself. Whereas tagging would enable you to quickly put a windows
into a temporary `context' where you could work with them, then go
back to how things were automatically. The intuitive notion was
a sort of `narrowing and widening focus of attention' (and of course
screen space).
What's annoying about the way things are done in my patch is that
the `highest viewed tag+1'-based rule makes it difficult to be working on two things
in this way in parallel (ie, switching between the two). You can do it, but
you have to manually choose your
two start points in the tags list which are separated numerically enough
that they won't collide as you push levels of temporary tagging onto
the numerically lower one.
Note I haven't got any good answers for these isssues, I just don't
think it's obviously a totally unworkable area.
cheers, dave tweed
Received on Thu Aug 31 2006 - 17:17:03 UTC
This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 14:30:48 UTC