[Geotools-devel] RE: [Geoapi-devel] Comparision of Propsals

[Paul S. <pa...@to...>]

Interesting to see some of these discussions.

The day, I learnt Feature is nothing but "properties" of
something spatial, my life became simpler.

Can someone tell me if there is any feature, be it GML3 or GML100,
that cannot be modelled by something as simply as this?

interface Feature
{
	object GetProperty(string path);
	void SetProperty(string path, object obj);
}

For the implementation? another simply thing, just put one
Hashtable in there and case closed.

class DefaultFeature implements Feature
{
	Hashtable properties;
...
}

It might give me ideas as to whether to worry about if ft
is derived from Schema or Type or not.

Best regards,
Paul.

> -----Original Message-----
> From: geo...@li...=20
> [mailto:geo...@li...] On Behalf=20
> Of Jody Garnett
> Sent: Friday, August 26, 2005 5:28 AM
> To: David Zwiers
> Cc: geo...@li...; Gabriel Rold=E1n; geoapi
> Subject: [Geoapi-devel] Comparision of Propsals
>=20
> David Zwiers wrote:
>=20
> >Aside from the obvious 'Complex' is a overused noun ... what=20
> is it in=20
> >you proposal. This still just looks like a rename to me. Is there an=20
> >actual plan as to what this would look like (am I missing one of the=20
> >many links floating around?)?
> > =20
> >
> It is more that the email & links cannot keep up with the=20
> discussion (that is mostly happening on IRC and privately).
> I am doing my best to keep up the email/documentation:
>=20
> To that end I have given up keeping discussion split out in=20
> terms of the modeling issues and gathered up everything into one page:
> - <http://docs.codehaus.org/display/GEOTOOLS/Feature+Model+Discussion>
>=20
> This should be the only link that matters until a decision is=20
> reached.=20
> My proposal is gathered together is interfaces, gabriels is a=20
> UML diagram. There is a short Q&A section and links to IRC=20
> discussions where progress has actually been made.
>=20
> >I am still in favor of gabrial's email earlier ... it's alot=20
> simpler.=20
> > =20
> >
> We are working from the same design card: separate out Type=20
> from Schema.
>=20
> >Oh, and so far as 'Type' goes ... well there are alot of=20
> classes called=20
> >that too, which one are you refering to ... (same applies if this=20
> >referes to a concept too).
> > =20
> >
> Understood, I felt I had to rename it as AttributeType is=20
> already lodged in peoples brain.
>=20
> >Page is blank ...
> > =20
> >
> Confluence is suffering hit reload.
>=20
> >>I hope this discussion helps,
> >>   =20
> >>
> >Well ... it actualy confused me ...=20
> > =20
> >
> Fair enough, I don't fully understand Gabriels propsal yet=20
> (but I am working on it).
>=20
> Jody
>=20
>=20
> -------------------------------------------------------
> SF.Net email is Sponsored by the Better Software Conference &=20
> EXPO September 19-22, 2005 * San Francisco, CA * Development=20
> Lifecycle Practices Agile & Plan-Driven Development *=20
> Managing Projects & Teams * Testing & QA Security * Process=20
> Improvement & Measurement * http://www.sqe.com/bsce5sf=20
> _______________________________________________
> Geoapi-devel mailing list
> Geo...@li...
> https://lists.sourceforge.net/lists/listinfo/geoapi-devel
>=20