[OSM-talk] New "Highways" view in OSM Inspector

[Anthony]

On Sat, Jan 9, 2010 at 8:55 AM, Stephen Hope <slhope at gmail.com> wrote:

> But if you want to edit a small area in the middle, your editor won't
> download the huge outer areas required to find that such a line
> exists.
>
> This was done by design, because it makes processing an API request
> much easier, and cheaper on server resources.


It makes the programming much easier, anyway.


> It may well be that
> your ideal of two points only is the best way, but unless the API is
> changed to work that way it is irrelevant. It doesn't work.
>
> The whole great circle issue is a red herring - it doesn't matter as
> long as the API doesn't change.  Don't expect the API to change in a
> hurry.
>

Personally, I don't, but then again, I don't really care.  I'm not going to
jump through hoops to facilitate lazy design.  Were it not for "the whole
great circle issue" I'd say use as few points as possible, and let the
designers of the API be embarrassed every time that causes problems.

But the fact of the matter is that the only lines in OSM which are going to
be perfectly straight and accurate are the ones which are defined
algorithmically with reference to latitudes and longitudes, not the ones
that are built out of atoms.  As they are defined with reference to
latitudes and longitudes, I figure they are straight with reference to
latitudes and longitudes as well.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20100109/0bd68625/attachment.html>