Martin, I don't want to Shanghai this thread but I feel like I need to add my 2 cents here. I firmly believe that it's very valuable to have a toolkit for processing complex scientific data. I like your example of two seemingly different groups of scientists (eg. physicians and oceanographers) being able to work together to map cholera propagation through the use of common data/format standards. Having said that, I also think that there needs to be tools made available for the common user so that they can perform the same (or similar) analysis that the seasoned scientists can. Neogeography[1] is the term I know of that identifies this as the "the usage of geographical techniques and tools used for personal and community activities or for utilization by a non-expert group of users" This is very appealing to a lot of end users who may not have the skills needed to code up a robust solution but also know what problem they need to solve. I would love to have my cake and eat it too meaning that SIS would have both feature rich low level as well as high level tools and libraries that would appeal to all audiences. What is being added now adds to to the lower-level architecture, so what roadmap items can we lay down to start building tools and services on top of SIS for the causal user (eg. [2][3])?
A [1] http://en.wikipedia.org/wiki/Neogeography [2] https://issues.apache.org/jira/browse/SIS-67 [3] http://goo.gl/aZyRZ On Feb 17, 2013, at 9:08 AM, Martin Desruisseaux wrote: > Hello Chris, Adam and all > > Le 15/02/13 17:44, Mattmann, Chris A (388J) a écrit : >> Guys, I need to prepare a short talk on Apache SIS for this NOAA meetup next >> week. >> >> I'm going to do it over Skype. Would love any input on slides or material >> you'd like me to include. > > I can send you by email some LibreOffice/OpenOffice.org slides that I have. > Or maybe there is some space that we can use somewhere for this kind of > material? Those slides would surely need to be adapted however. > > > Le 16/02/13 21:42, Adam Estrada a écrit : >> I think that it is important to note that what is currently being worked on >> not only helps to promote interoperability, its helping to define it by >> adhering so rigorously to ISO and OGC standards. Your's and Martin's >> involvement with the OGC GeoAPI working group is huge for the geospatial >> community because you can help to promote the use of these standards across >> all communities. I really believe that the key takeaway should resemble >> something like this. > > I also believe that our efforts in participating to standards are a very > important aspect. GeoAPI is one aspect, but hopefully would not be the only > one (after all, we still don't know if GeoAPI will succeed). I think that > participation to the meteo-oceanography group is also important and seems > close to the concern of many SIS members. Actually, I would love to present > SIS as a library having a focus on science :-). One real use case that I have > see is crossing data from physicians with data from oceanographers, for a > study of cholera propagation. Because physicians and oceanographers are two > quite different communities, crossing those data without standards is uneasy. > The standards required here are more complex than the usual WMS (we need > quality and uncertainties estimations, etc.). > > An other aspect that we could said is that SIS benefits from the experience > gained from the previous projects: we have learn what worked and what have > been bad idea. Every class that we port to SIS is an opportunity to revisit > it and sometime to simplify it. When a feature seems overly complex, we have > the possibility to search in the previous projects how this feature was used. > This help us to determine how it can be simplified. > > An other reason for revisiting classes one-by-one is that the Java landscape > as changed: JavaFX and Android need to be keep in mind, which imply for > example to try to isolate Java2D dependencies in dedicated classes. Some > factories may be replaced by JSR-330 (dependency injection). Support for > supplementary Unicode characters become worthy, since they now include > geographic symbols. OSGi support became a goal, etc. > > > Le 17/02/13 00:45, Mattmann, Chris A (388J) a écrit : >> Guys: another thing to note. Fast Feather tracks just opened up at >> ApacheCon NA 2013. It's an opportunity for a short, but meaningful talk: >> >> http://s.apache.org/BqE >> >> Maybe we should do a Fast Feather on SIS? Any interest? > > Do we have someone who can be present? > > Martin >
