icon-slicer 0.3

[Fredrik Höglund]

On Saturday 28 June 2003 22:17, Owen Taylor wrote:
> On Sat, 2003-06-28 at 14:08, Fredrik Höglund wrote:
> > I think that creating a new cursor standard for those cursors
> > would be important, especially to those developers that don't
> > target a specific desktop with their applications, but still need
> > to know that the cursors they use are available in the cursor
> > themes designed for the desktops their applications are
> > being used in.
> >
> > The old X11 cursor font is after all mostly a collection of random
> > images that don't meet the needs of modern applications,
> > and is IMO no longer suitable as a standard for this.
>
> There is recent discussion of updated standard sets of cursors
> from:
>
> http://mail.gnome.org/archives/gtk-devel-list/2003-June/msg00104.html
>
> It's certainly clear that the old X set is both too big and too
> small and needs to be replaced with a set that is chosen on a more
> rational level than "look at this neat image I came up with of a
> space shuttle".

Thanks for the link Owen, I'll comment on parts of that discussion
below.

I think the best way to proceed with this is to assemble a list of
cursors for the spec, based on what's been said in the discussion,
and then take it from there.

I think we should try to make the spec as compatible as possible
with other platforms, since the major tookits are now platform
independant.

For the discussion, this link might also be of some interest:
http://doc.trolltech.com/3.2/qcursors.html#details

If you scroll down past the image you'll find a table that shows
which Xcursor names the current Qt 3.2 beta uses for its cursors.

Regarding the list compiled by Matthias Clasen and the Qt
SplitV/SplitH cursors, those do have X11 equivalents in the form of
the sb_h/v_double_arrow cursors, and they appear to be used for
the same things in both Qt and Gtk+.

In the redglass and whiteglass themes that's not apparent though,
since in those themes they have the same design as the Qt resizing
cursors.

Also, about the resizing cursors, it is true that it's possible to map
the non-projective cursors to projective ones, but it's not possible
to do it the other way around, and that creates a backwards
compatabilty problem for Qt since it only exposes projective
resizing cursors. 

That also needs to be taken into account when deciding which
cursors should be included in the spec.

> When you get to things like tool cursors from drawing programs, then
> I'm not sure you can enumerate all the cursors in advance. For that,
> we'd probably just have to have a registry somewhere so that cursors
> that were the same used the same name. (*)

Yes, that's probably a good idea. There are some tool cursors that
appear to have uses outside of just drawing apps though, like
the zoom in/zoom out cursors.

But like you said, it's difficult to know where to draw the line when it
comes to which cursors should be included.

I think we'll be able to see that more clearly when we have an initial
draft of the spec.

> Regards,
> 					Owen
>
> (*) personally, I think tool cursors probably should be generally
> avoided - we can now do semi-transparent brush-shaped cursors, which
> are vastly more useful.

I think tool cursors have a clear usability value in some cases though.
For example you're not likely to accidently use the fill tool in the Gimp
when you meant to use the airbrush tool.

Regards,
Fredrik