This page presents ideas for short projects that contributors may wish to consider implementing
Project 1 |
|
Name |
3d Visualisation |
Description |
|
Provide a 3d view of the model projection by superimposing it onto a 3d visualisation widget such as Marble. The implementation process would consist of adding a new model view to the openModellerDesktop user interface and incoroporating Marble widget into this view and then loading each model into the 3d view when it is selected in the experiment tree. The user will then be able to interactively rotate and zoom the globe to visualise the model. Additional logic would be needed to cater for non global projections and non geographic projections. For non geographic projections the 3d view should be disabled. For non global models, appropriate padding should be added to the model projection to facilitate display on teh 3d globe view. Ideally the presence / absence points should be overlaid on this 3d view. The work would involve incorporating new code into openModeller Desktop and possibly the Marble project (in collaboration with the marble developers) if needed functionality does not exist. A preview of the desired outcome is available here: http://www.flickr.com/photos/timlinux/1071996497/ |
|
Links |
|
Advisor |
Tim Sutton |
Allocated to |
not allocated |
Completed |
not started |
Project 2 |
|
Name |
Semantic layer set matching |
Description |
|
When we run models using different environmental scenarios, the environemnt layers need to be of the same type and order e.g. temp, rainfall, elevation. Currently this validation is left up to the user and when the layers are not properly matched errors occur. This project would involve researching an implementing a way to perform a semantic match on two sets of layers to gauge how closely matched they are. When layersets are obviously not matched we will prevent the user using them in the same experiment. When there is uncertainty, the user should be informed but still allowed to use the two datasets together. |
|
Links |
|
Advisor |
Tim Sutton |
Allocated to |
not allocated |
Completed |
not started |
Project 3 |
|
Name |
Peered architecture for collaborative experiment processing |
Description |
|
The idea with this project is to allow mulitple computers on a network to participate in computing a single experiment. One computer would be a designated master. Peers would poll the master for the next job in the experiment, compute the job and then submit the results back to the master. The master will be responsible for the assembly of the descrete jobs into the experiment again, and for polling peers to detect time outs so jobs can be reallocated if needed. |
|
Links |
|
Advisor |
Tim Sutton |
Allocated to |
not allocated |
Completed |
not started |
Project 4 |
|
Name |
Post processing tools : model comparison |
Description |
|
This project would seek to implement a tool for comparison and display of two model projections. The idea would be to create a 'difference' map where cells are coded with a high value where there is a large difference between the two rasters, and a low value where the cells are very similar. |
|
Links |
|
Advisor |
Tim Sutton |
Allocated to |
not allocated |
Completed |
not started |
Template
Project |
|
Name |
|
Description |
|
|
|
Links |
|
|
|
Advisor |
|
Allocated to |
|
Completed |
|