- From: Norman Gray <norman@astro.gla.ac.uk>
- Date: Thu, 23 Nov 2006 15:54:37 +0000
- To: Dan Connolly <connolly@w3.org>
- Cc: semantic-web@w3.org
Greetings. On 2006 Nov 21 , at 15.51, Dan Connolly wrote: > Wikipedia cites a NIMA tech report... eek... it's only good > thru 2010?!? I wonder if the process for the next version is > rolling yet. It'll be good for a good long while after that, for all practical purposes (given that your personal practical purposes don't include dropping cruise missiles on folk -- guess why the various WGSxx specifications were originally associated with the US DoD and GPS). The earth neither changes shape, nor nutates, quickly enough for most of us to notice datum errors. As a more general point, it does seem slightly eccentric for <http:// www.w3.org/2003/01/geo/> to privilege WGS84 over all other possible datums. WGS84 is what the GPS uses, but no-one writing a geographic mashup is likely to have to care about that. The issue isn't just one of obsolescence (how many internet years ago is 1984?), but parochialism. Galileo has been launched, and is due to come on-stream around 2010: it uses its own datam (as far as I can tell), which is distinct from, but based on, the ITRS (International Terrestrial Reference System). Thus if you were, at some point in the future, to add to your FOAF file a position taken from a navigation system tuned in to Galileo (without making the Galileo- >GPS corrections), it would be formally incorrect to call them a wgs84_pos (and you could be as much as a metre adrift of your correct position -- quel horreur!). Quite independently, quoting positions with reference to a particular datum is like quoting your height to the nearest nanometre: the extra apparent precision is probably spurious, for everyone except all those folk surveying telescope mountings, or -- well -- aiming missiles, who probably aren't the most urgent targets for mashup- facilitation (and who already know all this stuff in more detail than gives them pleasure). Thus rather than > I suggest obsoleting 2006/vcard/ns#latitude > in favor of 2003/01/geo/wgs84_pos#lat ...it would probably be more sensible to go in the other direction, and say > geo:lat rdfs:subPropertyOf vcard:latitude and nothing more. In particular, it would probably be best to leave the vCard lat/long as informally specified, and let WGS84, and ITRS, and ..., coordinates to be subProperties of them, for those few specialists who care. All the best, Norman -- ------------------------------------------------------------------------ ---- Norman Gray / http://nxg.me.uk eurovotech.org / University of Leicester, UK
Received on Thursday, 23 November 2006 15:54:54 UTC