[talk-au] OpenStreetMap in Government

waldo000000 at gmail.com waldo000000 at gmail.com
Thu May 16 22:22:25 UTC 2013


>
> 4wd tracks ? There are simply too many factors at play here for us to
> measure, should we measure the height or spacing of corrugations, the
> 'softness' of sand, the depth of run outs, the narrowness, the slope, the
> wetness of the mud, the effect of weather on the track ?


Well, what information did the mapper gather in order to judge its
suitability? Did the mapper notice the width(/narrowness) of the track? If
so, enter it. Did the mapper notice the maximum depth of run outs? If so,
enter it. Did the mapper notice something else? Enter it. Alternatively,
did the mapper follow a clearly laid out specification on the wiki of what
"suitability" means, in terms of the above factors? If so, follow the
objective procedure in the wiki to enter the "suitability" as a summary
that still has a clear meaning. That is the compromise I'm suggesting.

Even if we could, how could the average map user possibly comprehend the
> data ?
>

If the objective information is directly entered, it is straightforward
(e.g. width=*)! If it is entered in the form of some summary tag (e.g.
suitability=*), it is harder. The user would need to look up what that
means in the wiki. If the wiki description is vague, they have no hope of
comprehending what the tag indicates.

Again, I say, we need to put data in there that is likely to be usable.


Agreed. Usable by applications like renderers, routers, search engines, etc.


> In this case, the user wants advice on should they use the track in
> question.
>

The end user, yes. But the map should not DIRECTLY offer advice, because
advice is a function of the map that we all must share (i.e. representing
the state of the world on the ground) PLUS a user's preferences. There's no
sense muddling the two up when entering tags.

Compare that to the alternative, no information, a map user assumes every
> track shown is suitable for them to drive ! Dangerous indeed.
>

That's a straw man argument. The alternative is entering observable facts,
either directly or in the form of summary tags with objective definitions
in the wiki. I'm really only repeating what has already been said here -
please read it if you haven't yet:
http://wiki.openstreetmap.org/wiki/Verifiability
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-au/attachments/20130517/a274373d/attachment.html>


More information about the Talk-au mailing list