POSITION AUTO for point labels shouldn't choose from UC and LC first

[cnieman]

All,

I think it is cartographically incorrect to have UC and LC as the first two options for labeling points. As I mentioned on the dev list, from the cartography classes I have taken, and from some research I did after realizing that UC and LC are now the first options, I believe the order should be:

UR, UL, LR, LL, CR, CL, UC, LC

or

UR, LR, UL, LL, CR, RL, UC, LC

[sdlime]

CC'ing @tbonfort.. This is a recent change. Used to be (in 5.6 at least) UL, LR, UR, LL, CR, CL, UC, LC but Thomas had a good argument for changing... Ideally this would be tunable.

Steve

[tbonfort]

There was a discussion on the mailing list for this after several requests where made on the tracker to stop using UL, LR, UR, LL as the first positions (e.g. #4242 #4249). There's clearly no solution that will fit everyone until someone funds or implements configurable positions for AUTO positionning as proposed in #4242. Until then you can patch your install as it is a trivial fix.

[havatv]

@tbonfort I support @cnieman here. Mapserver should stick to what is recommended in the cartographical literature. I have sent an email to the dev-list to get a new discussion on this. While the literature suggests that there are several options, the current (and past) Mapserver way of doing things is very much in conflict with the literature (and practice).

[sdlime]

The past setup followed ESRI ArcPlot conventions and actually matched what @cniemanhttps://github.com/cnieman proposed….

Steve

From: Håvard Tveite [mailto:notifications@github.com]
Sent: Friday, September 21, 2012 9:10 AM
To: mapserver/mapserver
Cc: Lime, Steve D (DNR)
Subject: Re: [mapserver] POSITION AUTO for point labels shouldn't choose from UC and LC first (#4402)

@tbonforthttps://github.com/tbonfort I support @cniemanhttps://github.com/cnieman here. Mapserver should stick to what is recommended in the cartographical literature. I have sent an email to the dev-list to get a new discussion on this. While the literature suggests that there are several options, the current (and past) Mapserver way of doing things is very much in conflict with the literature (and practice).

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/4402#issuecomment-8765760.