Re: [Gzz] Re: [Gzz-commits] gzz/doc/pegboard 1008/PEG_1008.rst 1009/PEG_100...

[Tuomas Lukka]

On Mon, Oct 07, 2002 at 07:50:45PM +0200, Benja Fallenstein wrote:
> Tuomas Lukka wrote:
> 
> >On Mon, Oct 07, 2002 at 07:22:37PM +0200, Benja Fallenstein wrote:
> > 
> >
> >>Ok, since I have explained my opinion (have I?) and you have made up 
> >>your mind, I guess I should just step back and leave this to you, 
> >>because there is no reason in endlessly repeating the same arguments. 
> >>Right?
> >>   
> >>
> >
> >I'd really like you to understand the way I propose to take care of this;
> >it's technically quite nice.
> >
> 
> Well, I don't think you can convince me of this. All the details I have 
> heard (scaling undefined in the general case, line widths aren't scaled, 
> width and height are supposed to be determined by scale...) make me feel 
> that no matter what's at the core, there are so many layers of 
> inconsistency layered on top of it that the API will be conceptually 
> tangled and because of that, difficult to use. 

Ok, maybe the problem is that you're now thinking of scaling the coordinate
system as scaling the image. This not the point here. The point is that
coordinate systems give everything coordinates but the rendering code
decides what to do with them.

> That's just my opinion, 
> but that perception is why I think it's better if I simply opt out of this.

I'll feel really bad if I can't convince you otherwise. 

The system needs to be made somehow clearer.

        Tuomas