I agree that neither OsGEO nor the communities are meant to reproduce such full featured proprietary architectures. I think thay should point to give a strong, common, backgorund to even enable (someone) to build richer, integrated platforms. I know it's a hard target, and something is going on with this (ie the cartographic library), but it would help a lot to guarantee a low level concistency between the softwares. Ok, I'm talking for the C/C++ side of the world, but it's the same for the Java one. A good exemple for the latter is the integration of SextanteGIS inside the major projects (uDig, OpenJUMP, JGrass, Gvsig, etc.).<br>
<br>giovanni<br><br><br><div class="gmail_quote">2009/9/15 Michael P. Gerlek <span dir="ltr"><<a href="mailto:mpg@lizardtech.com">mpg@lizardtech.com</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Thinking aloud, a possible contrarian view:<br>
<br>
A goal like "to produce a comprehensive suite of tools [that do X or Y]..." doesn't likely fit with OSGeo's broad membership and interests. We are an umbrella organization representing a number of projects, each with its own unique goals and agendas. It is unlikely OSGeo would be able to produce a specific tool just because (hypothetically) the Board says we should: open source folks often don't take top-down direction well, unless it meets their own personal needs and agendas.<br>
<br>
Which is not to say that an analytical tool suite is a bad idea, just that it seems unlikely to be a worthy goal at that level of the hierarchy.<br>
<font color="#888888"><br>
-mpg<br>
</font><div class="im"><br>
<br>
> -----Original Message-----<br>
> From: <a href="mailto:discuss-bounces@lists.osgeo.org">discuss-bounces@lists.osgeo.org</a> [mailto:<a href="mailto:discuss-">discuss-</a><br>
</div><div class="im">> <a href="mailto:bounces@lists.osgeo.org">bounces@lists.osgeo.org</a>] On Behalf Of "René A. Enguehard"<br>
> Sent: Monday, September 14, 2009 2:35 PM<br>
> To: OSGeo Discussions<br>
> Subject: Re: [OSGeo-Discuss] Next 5 years for OSGeo<br>
><br>
</div><div><div></div><div class="h5">> What I'd like to see within the next 5 years would be more analytical<br>
> tools. Most of the projects in OSGeo are very much enablers: they put<br>
> the facilities in place for people to program their own tools. However,<br>
> as I have noticed over the years, people are reluctant to move to open<br>
> source implementations of geospatial software because they are, in<br>
> effect, losing capabilities. Yes, there is still the potential for the<br>
> same capabilities to be put back in, but the fact remains they just<br>
> aren't there. For example, I have never seen any MCDA, PCA, HotSpot<br>
> Analysis, CART or neural network tools in open source packages. If we<br>
> were to produce a comprehensive suite of tools offering the standard<br>
> analytical tools as well as some more advanced ones, then these<br>
> proprietary offerings wouldn't look as appealing. Moreover, if we had a<br>
> consolidated toolset which could be used on a multitude of project we<br>
> would not have to re-invent the wheel for each separate project.<br>
> Currently, proprietary software generally offers advanced analytic<br>
> capability out-of-the-box and open source software does not. I see this<br>
> as a bit of a stumbling block.<br>
><br>
> Another thing, and I was chatting about this in the lab today, is that<br>
> for particular needs, open source implementations of geospatial<br>
> software<br>
> generally don't have much to offer. The generic capabilities are there,<br>
> or at least enabled for others to program, but special-needs cases<br>
> there<br>
> is not much. The example used today in the lab was CARIS HIPS or SIPS.<br>
> What, if anything, exists in the open source community that could come<br>
> close to the processing capabilities of this?<br>
><br>
> Still another area with a lack of development is 3D and 4D modeling /<br>
> rendering / analysis, something like ESRI ArcGlobe with the 3D Analyst<br>
> package or Myriax Eonfusion. There has been very little work in these<br>
> domains which are of particular interest to me. Perhaps the amount of<br>
> people working in these areas is much smaller than the amount of people<br>
> using something more like general analytic capabilities, but it is an<br>
> area that "needs work" nonetheless.<br>
><br>
> The point, and I'd like to make this clear, is not the I'm bemoaning<br>
> the<br>
> lack of features and projects in the open source community. I think<br>
> OSGeo and the open source community have done a tremendous job and<br>
> should feel, rightfully, proud at what they have accomplished. However,<br>
> when asked what I'd like to see on the agenda for OSGeo, this is it.<br>
> I'd<br>
> like to see a hard push towards analytics to make the various projects<br>
> we have to offer more directly useful to the average GIS user. In the<br>
> end, it's really about market penetration. The more useful open source<br>
> software is, the better a "deal" it looks like to outsiders and the<br>
> more<br>
> people we'll attract.<br>
><br>
> Please note: I don't presume to speak for anyone but myself, IANAL,<br>
> just<br>
> my two cents, your mileage may vary, et cetera, et cetera, ad nauseam.<br>
><br>
> Tyler Mitchell (OSGeo) wrote:<br>
> > Hi everyone, a recent chat I was asked about our vision for OSGeo<br>
> over<br>
> > the next 3 and 5 years. I'd really like to hear thoughts on the<br>
> matter<br>
> > and pool a few of the ideas together for further discussions amongst<br>
> > committees, projects, chapters and the board.<br>
> ><br>
> > It's also a good way for the board nominees in the upcoming election<br>
> to<br>
> > get a sense of where other members are thinking these days.<br>
> ><br>
> > Best wishes,<br>
> > Tyler<br>
> ><br>
> > _______________________________________________<br>
> > Discuss mailing list<br>
> > <a href="mailto:Discuss@lists.osgeo.org">Discuss@lists.osgeo.org</a><br>
> > <a href="http://lists.osgeo.org/mailman/listinfo/discuss" target="_blank">http://lists.osgeo.org/mailman/listinfo/discuss</a><br>
> ><br>
><br>
> _______________________________________________<br>
> Discuss mailing list<br>
> <a href="mailto:Discuss@lists.osgeo.org">Discuss@lists.osgeo.org</a><br>
> <a href="http://lists.osgeo.org/mailman/listinfo/discuss" target="_blank">http://lists.osgeo.org/mailman/listinfo/discuss</a><br>
_______________________________________________<br>
Discuss mailing list<br>
<a href="mailto:Discuss@lists.osgeo.org">Discuss@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/discuss" target="_blank">http://lists.osgeo.org/mailman/listinfo/discuss</a><br>
</div></div></blockquote></div><br>