[mapserver-dev] 41latitude has an interesting article on googlemaps labeling

[Bob Basques]

Stephen, 

I was thinking the same thing.  I did wonder about the wider outlines around the labels doing the gradual fade to nothing though, which I think is what makes them more readable..  I've never seen or used a setting in MapServer that can do that.  The rest of it seems to be completely rebuildable as far as Mapserver/Mapfiles go. 

bobb 



>>> Stephen Woodbridge <woodbri at swoodbridge.com> wrote:


On 12/15/2010 4:47 AM, thomas bonfort wrote:
> Steve,
>
> Looking at that article, it seems we already have nearly all that is
> needed with simple mapfile syntax and classified data to obtain
> similar results. You do have to ensure that you use label priorities,
> and/or ensure that that your datasource serves out labels by
> increasing order of importance.
>
> based on the different classes of labels, we have:
> * adjustable size of text
> * adjustable color of text
> * adjustable outline width on text
> * absence of clusters by applying a buffer
>
>
>
> the only thing I see missing is the "collar" around major metropolitan
> cities. We have nearly this with label the label buffer, but this will
> also remove city labels from the metro area.

Thomas,

Yes, we are very close. I was thinking about the collar problem and I'm
wondering is that can be achieved with something like MINBUFFER and
MAXBUFFER like:


     city metro     collar
       O---------x------------y

where O is the city, x is the MINBUFFER, and y is the max buffer and we
would allow labels between O and MINBUFFER and exclude labels between
MINBUFFER and MAXBUFFER.

Do you think that this would solve the collar problem?

I'm thinking that this might not be smart enough and we might need
something like:

COLLAR MIN MAX [ATTRIB < num]

Where we apply the collar only to entities where ATTRIB<num which would
all us to filter small cities, but would allow another major
metropolitan area to join the current one. You would obviously have to
have the data ordered by size so that this can be processed in a single
pass. Also I'm not sure what should happen if the 2nd collar excludes
the firsts metro zone or vica versa.

if new point is in any metro zone then its ok
elsif new point is in any collar zone then exclude it.

Anyway, I really liked the article and the analysis they did.

Regards,
   -Steve W

> regards,
> thomas
>
> On Tue, Dec 14, 2010 at 19:30, Stephen Woodbridge
> <woodbri at swoodbridge.com>  wrote:
>> http://www.41latitude.com/post/2072504768/google-maps-label-readability
>>
>> It would be cool if we could do some of these tricks in mapserver.
>>
>> -Steve W
>> _______________________________________________
>> mapserver-dev mailing list
>> mapserver-dev at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/mapserver-dev
>>

_______________________________________________
mapserver-dev mailing list
mapserver-dev at lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/mapserver-dev/attachments/20101215/451e3039/attachment.html