[Merkaartor] translation of data primitives

Richard Lyons richard at the-place.net
Tue Jan 4 18:16:52 GMT 2011


On Tue, Jan 04, 2011 at 11:42:40AM +0100, Martin Feuersänger wrote:

> Chris Browet wrote:
> > On Mon, Jan 3, 2011 at 00:27, Richard Lyons <richard at the-place.net>
> > wrote:
> >
> >> On Sun, Jan 02, 2011 at 11:46:04AM -0700, James Ewen wrote:
> >>
> >>> On Sun, Jan 2, 2011 at 7:04 AM, Christian Kraemer
> >>> <christian.kraemer at gmail.com> wrote:
> >>>
> >>>> in german translation "Knoten" is used for "node". Despite very
> >>>> common unfortunately "Knoten" in german is "knot" and I doubt that a
> >>>> newbie will know, that "Knoten" means something like a dot or point.
> >>>>  Next phrase: The OSM "way" term is "Weg" in german translation". I
> >>>>
> >> would
> >>>> prefer "Linie" (means line) to not have to explain why a stream is
> >>>> mapped with a "way" data primitive. (As far as I know this was
> >>>> valid only once in history :-) What's the situation in other
> >>>> languages? Did you use the OSM-terms or more newbie-friendly
> >>>> expressions?
> 
> As I'm not a GIS expert, I don't know what is used there as there "right"
> terminology for these objects.
> But for me, with an engineering background, "node" and "way" resp.
> "Knoten" and "Weg" make sense.
> 
> Those terms align both in German and English with graph theory.
> And the data in OSM is nothing but a description of such graphs in all
> their varieties.
> Check http://de.wikipedia.org/wiki/Graphentheorie or
> http://en.wikipedia.org/wiki/Graph_theory
> 
> In the end this is probably owed to the fact that OSM was initiated by
> computer scientist and not by GIS experts.

Yes, that is all very well, but the point is that drawing or using maps is
not really graph theory.  

When looking at a map, a "way" is usually a "carriageway" or a "right of
way".  The line outlining an urbanized area or the perimeter of a body of
water a national park or whatever are not "ways" in the common sense of the
word.  Using this term is confusing, particularly as there may indeed be a
way along part of such a perimeter, but not along other parts.  In addition,
tools exist to handle the drawing of the double lines used to indicate
actual "ways" (roads, etc), that are not needed when drawing a perimeter.

In contrast, "node" is a normal word for the end of a line or the meeting
point of two or more lines.  "Node" is a perfectly natural and unambiguous
term for this use, and is in everyday use.  I cannot comment on the german
word, as I have no idea how widely it is in use.

However, I think we are merely "grinding water" in this discussion, since
the decision, however perverse, seems to have been made long ago. 

richard



More information about the Merkaartor mailing list