That's really nice, and I imagine it'd work well. Indeed you more want to
see _something_ fast, I feel like it's ok if the labels load a bit more
slowly.
I wonder if there's some way we could make that set up easier for users.
Like make a special layer type in OpenLayers, and some way to ease
configuration on the geoserver. Like if we had like WMS options for 'no
labels' and 'only labels'. And then just did that at the renderer level,
instead of making people create separate SLD's.
And on the openlayers side we could make a layer type that can take two
URLs, the cache and the dynamic label WMS.
On Fri, Mar 18, 2011 at 10:39 AM, Ivan Grcic <ivan.gr...@geo-solutions.it>wrote:
> Hi all,
>
> some time ago i had project where labeling was also pretty important
> (parcels) so I came to one solution that perfectly suited my needs.
>
> As I used OpenLayers for client, i had tiled and untiled layers at my
> disposal. If you are only bound to tiled client, then using large
> metatiling would be the best choice imho - as Chris allready said.
>
> First, let me explain why i wasn't satisfied with existing choices:
> 1) metatiling with big tiling factor is usable only if you pre-seed
> layers, as it could take longer time to generate this big image and
> chop it to tiles (sometimes that can be really long process though,
> and if your data tends to change frequently it just wont do) You will
> still have 2/4 labels in small place at metatile intersection, which
> can be annoying if you have parcel data.
>
> 2) fixed label position - if you have polygon data then position is
> usually set somewhere close to the polygon center. this way many
> labels don't show up if your map view covers only part of the polygon,
> so you have to pan the map to the center of the polygon to see the
> label. Kinda funny seeing big polygon that covers most of your map,
> and the label doesn't show up :)
>
> So I came up with an idea to create complex OL layer, that internally
> creates two layers: first one tiled/cached geometry layer, and the
> second one untiled layer for labels.
> Each layer is configured with different settings, like different URLs
> (to tilecache&GS), different style for label layer, image type and
> resolutions.
>
> Layer with labels had to be carefully prepared to keep the images
> returned by server small (things like scale filtering on sld, custom
> created pallete etc)
>
> This way I had fast cached layers for geometries, and somehow less
> faster layer for labels - but usually you dont notice its slow because
> of the fast loading tiled layer beneath it :)
>
> Hope it helps!
>
> regards,
> Ivan
>
> On Fri, Mar 18, 2011 at 2:40 PM, Andrea Aime
> <andrea.a...@geo-solutions.it> wrote:
> > On Fri, Mar 18, 2011 at 2:07 PM, Chris Holmes <chol...@opengeo.org>
> wrote:
> >>
> >>
> >> On Wed, Mar 16, 2011 at 11:17 AM, Andrea Aime <
> andrea.a...@geo-solutions.it>
> >> wrote:
> >>>
> >>> On Wed, Mar 16, 2011 at 4:01 PM, Steve Way
> >>> <steve....@infotech-enterprises.com> wrote:
> >>> > Hi All,
> >>> >
> >>> >
> >>> >
> >>> > Just a curious question, but how do we avoid our labelling from
> looking
> >>> > a
> >>> > mess within openlayers or some other tiling client when trying to
> >>> > configure
> >>> > the overlaps etc which might be intersected by the edge of a tile?
> >>>
> >>> The short answer is, don't use tiling with labelling.
> >>> The longer answer might be, reduce the mess using a tile cache and
> setting
> >>> up meta tiling
> >>>
> >>
> >> Note that you can use really large meta tiling. I think ESRI's default
> is
> >> 4096x4096, so 16x16 256 pixel tiles. Though at that size it's much
> better
> >> to pre-seed.
> >
> > Even at 16x16 meta tiling you'll get those unlucky polygons falling at
> metatile
> > system intersections labelled four times in a small space.
> > It's just going to happen less
> >
> > Cheers
> > Andrea
> >
> > --
> > -------------------------------------------------------
> > Ing. Andrea Aime
> > GeoSolutions S.A.S.
> > Tech lead
> >
> > Via Poggio alle Viti 1187
> > 55054 Massarosa (LU)
> > Italy
> >
> > phone: +39 0584 962313
> > fax: +39 0584 962313
> > mob: +39 333 8128928
> >
> > http://www.geo-solutions.it
> > http://geo-solutions.blogspot.com/
> > http://www.youtube.com/user/GeoSolutionsIT
> > http://www.linkedin.com/in/andreaaime
> > http://twitter.com/geowolf
> >
> > -------------------------------------------------------
> >
> >
> ------------------------------------------------------------------------------
> > Colocation vs. Managed Hosting
> > A question and answer guide to determining the best fit
> > for your organization - today and in the future.
> > http://p.sf.net/sfu/internap-sfd2d
> > _______________________________________________
> > Geoserver-users mailing list
> > Geoserver-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/geoserver-users
> >
>
>
>
> --
> -------------------------------------------------------
> Ivan Grcic
> GeoSolutions S.A.S.
> Software Engineer
>
>
> Via Poggio alle Viti 1187
> 55054 Massarosa (LU)
> Italy
>
> phone: +39 0584 962313
> fax: +39 0584 962313
> mob: +39 328 0559267
>
>
> http://www.geo-solutions.it
> http://geo-solutions.blogspot.com/
> http://twitter.com/geosolutions_it
> http://www.youtube.com/user/GeoSolutionsIT
>
>
> -------------------------------------------------------
>
>
> ------------------------------------------------------------------------------
> Colocation vs. Managed Hosting
> A question and answer guide to determining the best fit
> for your organization - today and in the future.
> http://p.sf.net/sfu/internap-sfd2d
> _______________________________________________
> Geoserver-users mailing list
> Geoserver-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/geoserver-users
>
------------------------------------------------------------------------------
Colocation vs. Managed Hosting
A question and answer guide to determining the best fit
for your organization - today and in the future.
http://p.sf.net/sfu/internap-sfd2d
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
