Vitali Diatchkov wrote:

Hello!

I am thinking about various object models, libraries, frameworks, application implementations related to GIS domain.

You did miss one - and an important one at that. GeoAPI is a project that is only about defining interfaces allowing the various toolkits and projects to work together. As an example they define CRS interfaces used to some extent by GeoTools, gvSig, Deegree. You may also want to look at GeOxygen which seems to be a machine translation of the xml schemas (so they have a TP_Object and GM_Object etc...).

Over the coming months geotools is looking at transitioning from its feature model to the one in the geoapi project - for a comparision of the different feature models you can look into the research on the geotools website.

There are useful sites such as freegis.org where lots of useful open source projects are listed.. My issue is more related to the development process where object models, APIs, etc are described, compared, what is the purpose, target of usage, etc.

I have faced with a problem to choose the existing (if exists) Topology Object Model (java implementation of Topology Data Model)

So far I have not seen a clear leader for a topology object model, there is however plenty of interest.

Doesn't it mean the open source community needs some efforts to improve "topology"-related direction of GIS development? Doesn't it mean the community needs, let's say, one powerful object model more with existing geometry object model (JTS), Feature Object Model (GT), DataStore object model - the "Topology Object Model"?

Needs, has and wants are all different concerns. For much of this code what you see is the result of someone (company or student) writing enough API to solve a specific problem. Please understand that generalizing and working together on a great solution is the work of volunteer hours for many and progress is as progress does.

for the following development in some contexts (topology data store (as a data store concept for nodes, edges, faces), topology fixing of inconsistent data sets, topology layer hierarchy aligning, etc.). Related to topology seems there is no good and full implementation of TDM as open source while there are different kinds of topology (polygonal coverage, network topology, layers hierarchy), only parts of it are used in one or another project (JTS is an example where T - topology - is used locally to build geometry graph to facilitate predicate operations and spatial processing).

Doesn't it mean the open source community needs some efforts to improve "topology"-related direction of GIS development? Doesn't it mean the community needs, let's say, one powerful object model more with existing geometry object model (JTS), Feature Object Model (GT), DataStore object model - the "Topology Object Model"?

I'm crazy about not full topology framework while it takes a lot of work from the community, but to have the opportunity for simple feature model and topology data model co-exist . It means whenever some topology-based processing is needed, TDM can be build locally in the area of interest from SFM to serve as a auxiliary structure for editing operations, efficient querying with topological predicates, whatever needs some topological information. This task is reachable. The first step is to choose (or develop) object model that will state of the art in its area.


Persuse the API in the geoapi project (where you have access to expert feedback), implement in whatever community you are happy with and welcome to the Java GIS community.

Cheers,
Jody



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Geotools-devel mailing list
Geotools-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to