[EMAIL PROTECTED] wrote on 06/21/2006 02:04:06
AM:
> Patrick Webb wrote:
> > I'm definitely warming up to the idea of the pluggable 19117 protocol.
> You guys have to communicate what 19117 *is*, remember we have not all
> got access to the doc :-) So you guys want to tell me what 19117 *is*
> and then I can consider getting excited as well....
I think I answered this in my last email, but was too chatty :). Let me
try to condense it some.
1] It's not a complete rendering system.
2] It is a data model framework for describing the API of a portrayal
service independent of implementation, and *including* natural language
descriptions of functions, parameters, and algorithms.
3] The API's may be collected into PortrayalCatalogs. catalogs have unique
names. Similar concept to FeatureCatalogs.
4] A rendering engine (external to 19117) may use the services defined by
these PortrayalCatalogs.
So, if I'm on the right track, the effect of integrating this architecture
into GT would be:
1] Anything that "draws" (points, lines, polygons, rasters, wind barbs
etc.) would be factored out into a PortrayalService.
2] Renderers would be responsible only for:
2a] connecting data with relevant PortrayalServices
2b] configuring PortrayalServices (e.g. line styles...)
2c] merging output of PortrayalServices
2d] maintaining a z-order.
3] Functionality in #2 could be factored out of the Udig or SLD renderers.
4] Specific renderers (e.g., SLD, Streaming) would be primarily responsible
for parsing their configuration files (e.g., mySLDmap.xml) and would
inherit most of #2. Alternatively, we could define a PortrayalConfigurator
interface which could translate from a specific format to the Portrayal
framework.
5] Most important, the majority of portrayal code becomes an implementation
of GeoAPI interfaces (say org.opengis.portrayal.*), therefore shareable.
6] Most important #2, this is a long-term architecture which supports both
production and development.
Coming at this from a professional standpoint, $107 (actually, with the GSA
discount, it's ~$90) to get a framework which a lot of highly paid people
have scrutinized is a bargain. :) I don't think I could come up with a
framework design for less than $107, and I'm _sure_ I couldn't get a
contractor to produce one for that. :) However, for those without a budget
line for this kind of thing, the first step will be to produce
well-documented GeoAPI interfaces. Perhaps we could even convince Patrick
to write a "19117 Primer", or a user's-guide-to-the-framework or even post
his master's thesis on the wiki.
In the near term: I'm not an expert, but I can look up answers if people
have questions...
Bryce
_______________________________________________
Geotools-devel mailing list
Geotools-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel
