Adrian wrote: "It seems like we have a problem of understanding."
This is likely due to my not explaining myself properly. :] Adrian wrote: "If you are working with geographic data (lat/long) using JTS (which only works in a Cartesian 2D system, then you have *necessarily* projected your data." Understood. Adrian wrote: "If you have not controlled your projection, then you are undoubtedly working in a particular projection, often called Plate Carre (with various spellings). For doing anything analytically useful, that projection sucks. It is neither conformal nor equal area nor does it have any other redeeming characteristics. If you are hooking into geotools anyhow, that's something you might want to think about. Projecting coordinates is pretty trivial and takes your work from the 'random calculation' realm into geographic information systems proper." In this particular scenario unprojected lat/long data is exactly what I need. I'm not producing any maps at this stage in the game. I'm just collecting position data using GPS. My data will be available for other people to make their maps. They can take my lat/long data and project it into whatever coordinate system works for them. I understand that JTS is a plane based coordinate system. That is why I was asking Martin about the Ellipsoid stuff. If I need to do any geometry calculations I'll be doing them on the surface of an ellipsoid, not on the plane. JTS is just serving as a convenient coordinate holder at this point. Thanks for all of the advice. When I do move to the map production stage, I'll likely use UTM, or maybe state plane coordinates. The Sunburned Surveyor On Thu, Nov 20, 2008 at 1:09 AM, Adrian Custer <[EMAIL PROTECTED]> wrote: > Hey, > > > Sunburned Surveyor wrote: >> >> Adrian wrote: "It would be interesting to hear what you mean by >> testing your geotools code live from openjump since I've always >> thought those were very different projects. How does openjump deal >> with gps data, in plate-carre space?" >> >> I'm working with TIGER data that is all in lat/long anyways. So I'm >> not dealing with any map projections. Not at this stage at least. >> Right now I'm just worried about building features from lat/long >> coordinates. Later I'll worry about displaying those features with a >> map projection. >> >> > > It seems like we have a problem of understanding. If you are working with > geographic data (lat/long) using JTS (which only works in a Cartesian 2D > system, then you have *necessarily* projected your data. > > If you have not controlled your projection, then you are undoubtedly working > in a particular projection, often called Plate Carre (with various > spellings). For doing anything analytically useful, that projection sucks. > It is neither conformal nor equal area nor does it have any other redeeming > characteristics. If you are hooking into geotools anyhow, that's something > you might want to think about. Projecting coordinates is pretty trivial and > takes your work from the 'random calculation' realm into geographic > information systems proper. > > >> Adrian wrote: "Wow! Your asking this question means either you have >> not understood what geotools is trying to do or you are distracted and >> so are not thinking." >> >> I think it is the first one. Right now my GPX module is basically >> stand-along. I want to make future releases of it play better with the >> rest of GeoTools. So I sort of went backwards. I have reinvented the >> wheel to get something working, now I want to slowly replace some of >> my home grown stuff with standard GeoTools components. I know this >> sounds like a silly way to design something, but I find GeoTools >> complex. It is easier for me to get a simple working design and then >> see how I can integrate code from the larger library. >> > > Great! Whatever keeps you working is good. The ever diplomatic Martin kept > his response open for possible changes, I see. Indeed, if there were a > structured block of functionality that was being proposed, he would welcome > it. But we are a long way from that in this initial work. > > Glad that GeodeticCalculator looks useful. That class is a utility class, > specific to Geotools, so if you propose new code for that class, those > changes are much more likely to fit the design of the library and be > welcomed in. Do look around if you are trying to do something new, there's a > lot of power in that there library. > > --adrian > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Geotools-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/geotools-devel
