[OSM-talk] Coordinates in OSM. Really annoying

Dave F davefoxfac63 at btinternet.com
Mon Apr 24 13:42:23 UTC 2017


We're *all* volunteers giving our valuable time to OSM. It could be more 
efficiently spent if fixable variations such as this didn't occur. When 
I add data I'm, quite rightly, expected to do it to a certain standard 
of quality.

DaveF

On 24/04/2017 13:46, Nick Whitelegg wrote:
>
>
> Could one of the reasons be that open source developers are doing this 
> in their own time and have a million other responsibilities in their 
> lives?
>
>
> TBH I think it's a case of live with it and read the documentation. 
> There are more important things to worry about. ��
>
>
> Nick
>
> <http://www.solent.ac.uk/disclaimer/disclaimer.aspx>
> ------------------------------------------------------------------------
> *From:* Dave F <davefoxfac63 at btinternet.com>
> *Sent:* 24 April 2017 11:31:24
> *To:* OSM Talk
> *Subject:* Re: [OSM-talk] Coordinates in OSM. Really annoying
> Hi
>
> I'm unsure why go on about anyone being "thrown out".
>
> I'm unsure why doing it inconsistently in the past in 'legacy' code is
> any reason for not trying to sort it out for the future.
>
> Programmers can reformat to any standard* of their desire /within/ their
> own program. It can't be hard to do; it is, after all, pure ASCII text.
> What's irritating is the responsibility oft the syntax has been passed
> on to the end user.
>
> What are the reasons the authors of the programs listed can't coordinate
> with each other to simplify it for users within OSM community?
>
> * Are they really adhering to a 'standard' or just doing it one way
> because a competitor did it another?
>
> DaveF
>
> On 23/04/2017 08:29, Roland Olbricht wrote:
> > Hi,
> >
> >> Is there /really/ any need for *six* coordinate formats? It's hard
> >> enough to learn a new process without basics like this tripping you up.
> >
> > There is nobody who is trusted enough to set an universal standard:
> > https://xkcd.com/927/
> >
> > Basically, there is an ISO standard
> > https://en.wikipedia.org/wiki/ISO_6709
> > to have latitude before longitude. Leaflet complies, OpenLayers does 
> not.
> >
> > This is for historical reasons. When multiple projections were
> > commonplace as exchange formats, then they often used x and y as names
> > for the two numbers, and x often decoded to something loosely or
> > tightly related to longitude.
> >
> > However, OpenLayers is too useful to be thrown out just for having the
> > wrong coordinate order. The same applies to a lot of other tools with
> > legacy coordinate order.
> >
> > To have a gentle pressure towards the ISO standard, the advertised
> > interface is latitude-longitude. There are some precautions for inert
> > legacy tools:
> > http://dev.overpass-api.de/blog/bounding_boxes.html#lonlat_bbox
> >
> > As Lester has pointed out, XML requires explicit parameter names. By
> > the way, I am not aware of anybody actively using the XML syntax. You
> > can safely ignore that.
> >
> > For the delimiter question: There are programming languages with a
> > combined market share of almost 90% that agree to have to semanticy in
> > whitespace. The sole widespread-used exception is Python. Once again:
> > Are you seriously asking the OSM community for a crusade to throw out
> > Python for minor syntactic infrigement?
> >
> > Beside Python, the delimiters are always commas and semi-colons. As
> > commas tend to be used to delimit parameters, they are for the numbers
> > of the bounding box the delimiters of choice.
> >
> > Cheers,
> >
> > Roland
> >
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20170424/4a521eac/attachment-0001.html>


More information about the talk mailing list