Nick.Whitelegg at solent.ac.uk
Sat Jan 21 14:16:42 GMT 2012
Just to clarify this: on my site, the client does indeed only request data for the bounding box currently being displayed, see
The features.php script expects a bbox (comma-separated west,south,east,north bounding box) parameter.
All the freemap code is available at
I haven't actually worked on simplifying the feed at low zoom levels just yet, though my approach would be to filter based on feature type (e.g. only cities and large towns).
-----Frans Thamura <frans at meruvian.org> wrote: -----
To: Graham Jones <grahamjones139 at gmail.com>
From: Frans Thamura <frans at meruvian.org>
Date: 21/01/2012 10:16AM
Cc: Nick Whitelegg <Nick.Whitelegg at solent.ac.uk>, OSM-talk Openstreetmap <talk at openstreetmap.org>
Subject: Re: [OSM-talk] geojson
i post another question related to this, in the OSM API question
we will see, :) we will use your JSON model :)
and we can see what issue will raise up, and we have started the
threada about this scalability. that will become my homework to solve
that, when it happen
On Sat, Jan 21, 2012 at 5:08 PM, Graham Jones <grahamjones139 at gmail.com> wrote:
> Hi Frans,
> I haven't checked Nick's code, but I think that it works by the client
> requesting data for the specific area that is being displayed on the map.
> The BrewMap code just uses a single data file to cover the entire area (in
> this case the uk), so even if you zoom in very close, it still uses data for
> the whole uk. This is fine if you only have a few hundred data points, but
> will become very slow if you have tens of thousands.
> Now for low zoom levels (large displayed areas) this amounts to the same
> thing, so there would be no benefit, unless you do some filtering on the
> server side to simplify the data (eg group POIs if they are going to be
> plotted on top of each other). I have not done this yet, but Nick, and the
> folks working on KothicJS are likely to have thought about it and may have
> solutions, which is why I think it is worth investigating how they do it.
> Whether it is worth putting effort into simplifying the data on the server
> depends on how much data you are dealing with.
> Graham Jones
> Hartlepool, UK.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the talk