[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gzz] Generalizing views - Collecting everything together
From: |
Tuomas Lukka |
Subject: |
Re: [Gzz] Generalizing views - Collecting everything together |
Date: |
Fri, 28 Feb 2003 15:52:27 +0200 |
User-agent: |
Mutt/1.4i |
On Fri, Feb 28, 2003 at 02:48:39PM +0200, Asko Soukka wrote:
> Fri, 28 Feb 2003, Tuomas Lukka kirjoitti:
>
> > On Fri, Feb 28, 2003 at 02:04:28PM +0200, Asko Soukka wrote:
> > > Fri, 28 Feb 2003, Tuomas Lukka wrote:
> > > > I disagree. I think that we *can* have automatic interpolation most of
> > > > the time.
> > > > One important thing on the way is to *show* that the same node is shown
> > > > twice:
> > > >
> > > > > A-C-B
> > > > > | /
> > > > > B -
> > > >
> > > > A ghost connection (beam, or such).
> > >
> > > I discussed with Tuukka about this yesterday. The interpolation in Loom
> > > look ugly and should probably be replaced by a ghost connection, but how
> > > to make it look nice in AWT?
> >
> > Could we first think about how we want it to look and only *then* how to
> > approximate the ideal situation in AWT? Constraining ourselves to AWT's
> > limitations this early in the design phase is not good.
>
> On GL side we can use unique backgrounds.
That's a good idea; this would be an ideal application for them, since they
emphasize the identity.
> Probably also fillet-like(?)
> ghost (slightly transparent?) connections,
Yes.
> but then the screen could get
> messy if there are a dozen of cross-connections going across the wheel.
Something we did once was emphasize those connections that were close to the
mouse.
> In our simple views all connections are straight lines. Will this change,
> or could we use some kind of bezier line connecting the different
> occurrences of the same node.
You could use *ANYTHING* you can draw for the connection. Bezier, subdivision
surfaces, lightnings, ...
> Should the connection go across the wheel or
> try to circle along the edges? Color coding for connections? Dashed lines?
> ...
This would depend on how many nodes we expect to see, and how many connections
among them.
Tuomas