This is an old revision of the document!
This section depicts what is missing inside RTS2. In states expected human (and other) investments, profile of people doing that and a brief description of the problem. If somebody will be interested in taking problem and solve it, we will move it to extra page under ideas: namespace and provide a link there.
Final integration of meteorite detection code to RTS2
Expected costs: 1 student, 1-2 weeks
Profile: image processing, C++, Subversion
I have meteorite detection code for RTS2, however it is not integrated in RTS2. The task of the student will be to take the code, merge it with current RTS2, fix bugs and run it on our widefiled CCDs.
GUI
Expected costs: 1-3 students, 2-6 months
Profile: GTK, PyGTK ideal, XML-RPC, Subversion, GUI design, user documentation
There is prototype implementation of GUI, however I do not have time to finish it. It is in PyGTK, uses XML-RPC to communicate with RTS2. I am able to give away code and provide guidance what shall be implemented.
RTS2 Web
Expected costs: 1-3 students, 2-6 months
Profile: Google Web Toolkit, Web interface design user documentation, Subversion
There is a prototype implementation of RTS2 web, which have some limited functionalities. I would like to see extension of this piece of software which should result in usable and useful WWW interface for RTS2.
Comparsion of various XML-based interfaces for remote telescope
Expected costs: 1 student, 3 months
Profile: computer science theory, C++, Subversion
We went through an interesting history none → SOAP → XML-RPC regarding external access to RTS2. I just recently read through Atom specification, which is widely used by Google (so that was not XML-RPC - my mistake) and do not know if that is what we wanted. I like few characteristics of Atom, but dislike few others (apparent lack of C++ library to handle Atom requests is most probably the most serious one). The tasks of this mainly theoretics work is in comparing those various possibilities, try/use simple implementation with RTS2 and compare them.
Better scheduling
Expected costs: 1 student for 6 months
Profile: process scheduling (mathematics or computer science), some programming
I am currently working on making ultimate scheduling, which will answers all problems which we have in scheduling. Having somebody else working on some topics, but using different algorithms (I am using genetics algorithms as this is requirement) can be useful and possibly beneficial for both parties.