Hi,
Knock out criterion?
I am not sure that I understand why you would want a level of detail
functionality to reside in the rendering at the browser. This should
probably be handled on the server side not the client side. Why download
several levels of detail and tie up scarce bandwidth?
If level of detail is important to your application you could use a
spatial database for the maps on the server side. You can then handle lod in
a couple of ways.
1. Query using a simplify function filter. This would require some
javascript to identify a lod threshold as well as the aoi. Then the svg view
object would be changed to a new view object based on the results of your
spatial query.
2. Keep a detail level as a column in the spatial table and limiting
the query to the level appropriate to the viewbox. Again you would use
javascript to monitor the onzoom events. For example level 1 includes just
highways level 2 includes roads as well as level 1, etc If scale
discrepancies exist you would probably not be able to use a clause like "and
level<=3" Another approach is to keep a min and max visibility threshold for
each object in the table then the query would add a clause like "and
(visibility>min and visibility<=max)" This approach allows you to handle
scale discrepancy issues by setting the min and max thresholds to eliminate
coarse scale when moving to finer scale.
If you don't have a spatial database you could simulate one by dividing the
view into the various level of detail (one svgz for each) and calling them
with javascript from the onzoom event to detect when a specific level of
detail threshold has been reached. This could get complicated in a hurry
though and does not save much download time at the higher levels of detail
unless you go further and come up with a tiling mechanism too. In which case
as you zoom down, the lod svgz files are divided into horizontal tiles and
only the required tiles are merged to replace your svg view object src.
By the way there is lod built into the renderer since objects have to be
mapped to screen pixels which automatically reduces detail intricacy. I
assume that is not what you want though.
Trying to include lod in the renderer doesn't seem like a good idea.
>this question is related to my "pseudo LOD" question a couple of
days ago:
Hi Pierre.
Thank you for supplying me the correct technical term. I have done
some searches with SVG and LOD and found several recent proposals to
include LOD support to SVG.
Odd that this capability has not been added so far. The lack of lod-
support is almost a knock-out criterium in my eyes. I mean what's the
big deal about having the graphic rendered on the client side if you
cannot easily some into and out of details?
It certainly makes SVG much less usefull for mapping applications.
[Non-text portions of this message have been removed]
-----
To unsubscribe send a message to: [EMAIL PROTECTED]
-or-
visit http://groups.yahoo.com/group/svg-developers and click "edit my
membership"
----
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/svg-developers/
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
