Scout wrote: > Neil R. Ormos wrote: >> Scoute wrote:
>>> The geo format does not appear to be >>> overcomplicated.
>> Beauty lies in the eyes of the beholder I guess. >> I see DIV this, Class that, a bunch of angle >> brackets, alternative forms. So much extra stuff >> for so little gain, and, bonus!, incompatible with >> prior formats.
> Arbitrary HTML can have lots of confusing > source. To me, the geo format seems like a > pretty basic way of identifying geographic > coordinates in the confusion of possible content > that an arbitrary Web page might have.
>> As one simple but effective example, the >> coordinates could be encoded in a conventional >> URL.
> There's no standard there, either. I suppose we > could pick one, say Topozone or Tiger, and use > that for the standard. But what if you want to > create a Web page that has some geographic > coordinates on it somewhere and you don't want > to create a link? How would you drop a URL in a > Web page in a way that a browser just displays > the coordinates, not a link?
Put the URL (or whatever) in an HTML comment and encode whatever you want displayed as normal. No doubt the W3C pedants will squeal like stuck pigs. But having the undisplayed data in a comment is far more honest about the fact that any semantic interpretation of the data is an undefined side effect of its presence and the serendipity of using some specific browser with some specific gadget. Unfortunately, modern web pages are unnecessarily layered in a nasty soup of client-side languages which creates all kinds of aggravation for users and results in ridiculously large, complex, and mutually incompatible browsers. Encoding non-displayable content in the HTML code of a page just adds to this miasma.
> The geo format provides one solution.
Yes it does provide one solution. I disagree that it is a "standard" solution or an improvement over using some older, common format that is (1) compatible with at least some prior work; and (2) accommodates more extensive and explicit semantics.
>> Also, there have been numerous prior XML formats >> for representing geospatial data, including >> Garmin's format and several for GIS applications.
> I was hoping you would cite one that seems > targeted at what the geo format seems to be > targeted at, namely identifying geographical > information in arbitrary Web pages. For creating > files of waypoints, something like LOC files or > GPX files solve the problem, but I'm not sure > you can embed those formats in an arbitrary HTML > page to identify a latitude/longitude that > appears somewhere on a page of text.
Why does the format have to be "targeted" at arbitrary web pages? If you don't like a TMS URL, virtually any of the well-developed existing formats could be embedded in a URL or an HTML comment. The existing formats at least have some relationship to existing applications and encode more sophisticated semantics.
Although I'm not a fan of XML, consider as just one example the Garmin XML-based format, which is already defined to encode a great deal of useful information about a point, and can also represent collections of points in the form of tracks and routes. There is no reason why geospatial information in that format cannot be encoded in comments near the relevant displayable page content, and it would be no more difficult to parse using grammer-based parsers then the alleged "standard". I point out the Garmin format not because it is necessarily superior to other formats but because Garmin has a large market share and its format probably has the largest base of users, excluding the formats directed to GIS applications.