[Openstreetmap] deagging interface

[Graham Wall]

What is the lifetime of the cached tiles by the way?

On 19/10/05, SteveC <steve at asklater.com> wrote:
> * @ 19/10/05 02:46:35 PM lars at aronsson.se wrote:
> >
> > > > > > > http://www.openstreetmap.org/edit/view-map2.html
> > > > > > wow!
> >
> > This new viewer is very nice, but there are some issues.
>
> It's still beta, thanks for the bug reports :-)
>
> > 1. When I recenter the map by double-clicking, the next zoom moves
> > to a different center point.
>
> Double clicking is a known bug. It doesn't work in IE at the moment.
>
> > 2. It is very slow.  Aren't tiles cached on the server?  The
> > zooming in and out seems to result in fixed levels of scale, e.g.
> > 0.703125, 0.17578125, 0.0439453125, 0.010986328125,
> > 0.00274658203125, 0.0006866455078125.  But I can also set (in the
> > URL) any arbitrary level of scale (e.g. 0.01000, 0.01001) and zoom
> > in and out from there, resulting in larger or smaller rounding
> > errors.  If tiles are cached at every arbitrary scaling level,
> > there would be no end to the storage need.
>
> scale maps to a integer zoom level that (should) round off errors but it
> might be mapping to a non-integer zoom level which would be bad.
>
> Any WMS image requested is cached, simply.
>
> As for speed, you should only see tiles taking time when they're being
> rendered for the first time. Why is this slow? Because of three things,
> in order of my estimate:
>
> 1) the NASA Landsat server can be very, very, very slow when the
> americans wake up and start using it. And we backend to that for sat
> imagery.
>
> 2) MapServer is a piece of junk which I've almost replaced.
>
> 3) SQL calls for loading some of the street data isn't totally optimised
>
> have fun,
>
> SteveC steve at asklater.com http://www.asklater.com/steve/
>
> _______________________________________________
> Openstreetmap mailing list
> Openstreetmap at vr.ucl.ac.uk
> http://bat.vr.ucl.ac.uk/cgi-bin/mailman/listinfo/openstreetmap
>