|
This is version 7.
It is not the current version, and thus it cannot be edited.
[Back to current version]
[Restore this version]
Early Career Faculty Training Notes
Attendance (Katz): Shawn Bowers, Dan Higgins, Samantha Katz, Bill Michener, Deana Pennington, Mark Schildhauer
Goals for this training to have users return from training and be able to:
- Create a scientific workflow before they start in kepler. This is something that can help organize their thought process of conducting research. This workflow could and we should then be created in kepler
- Use their own dataset in a pre-existing kepler workflow
- Create their own simple kepler workflow using r.
- Enter their own data using either morpho and the knb data registry and get that data into a kepler workflow -- a simple exercise that exploits
- Show how kepler utilizes the EcoGrid and how that functions into the importance of well structured data sets and databases -- which included QA/QC
- KR--SMS should show the functionalities that allow kepler to work -- showing the semantic, semantic annotation and ontologies and how that works for them.
Review agenda
Discussion on linking the training together with the same dataset used throughout the training
- update presentation with the common dataset;
- Deana will send out npp dataset
- link Dan Higgins presentation on r with Kristin's presentation on QA/QC -- Dan will try to illustrate some of the approaches Kristin covers in her presentation -- share screenshots Sam will send email Dan the QAQC presentation and email Kristin and Dan to facilitate
- Grid activity -- provide example that illustrates the distributed execution notion and link back to Kepler and how that is executed. Will and Mark will work on this
- Visualization in Kepler (demo + exercise) -- will cover a summary of the r, image j -- raster type and photoshop -- pdfs : tables; images; time series example show them the examples in kepler imaging Dan and Deana will work on this
- Taxonomic resolution in Kepler –Stewart -- will cover GBIF; Biota; query into DIGIR and IT IS, Vegbank demo Sam will work with Aimee on presentation
-
Documentation Notes
Review Getting Started Guide documentation for the release
- First example -- Simple Addition changed to Simple Statistics -- needs reviewed for edits and completeness -- Need Dan to review and save as workflow for the Beta3 release
- Need to finish updating section 5.3 Director and Actor Icons with most recent icon family categories for the Beta3 release
- sam used the 3 "all-the-actors actor in workflows" directory that Matt made and went back to svg and made some adaptations)
- Laura needs to be reviewed and make suggestion
- Need to finish the screenshots in the GSG.Working with Monica from UCSB people on creating screenshots. Will work with Monica on making sure the GSG images are the same as the Kepler interface -- go over images with Laura
- Adding director and extra r workflows? Should we just keep expanding the GSG?
- Review actor classifications
- List of list SEEK actors and workflows -- corresponds
- Review and request developer level actor documentation -- need help Dan, Ilkay and Efrat cannot be the only people reviewing
- Removed the following actors from the actor documentation: File to String Converter ; Invoke server; ServerExecute, TempActor;
- Naming convention -- camel case conventions
- Incorporating documentation into application -- working with Chad on the strategy mentioned below
Strategy: Use the Actor right-click menu to update the individual documentation screens (i.e., update the actor's MOML files). Use HTML <p>, <i>, and <tt> tags to preserve formatting. Save the updated actor files as KAR file and then overwrite the existing files in the /Kepler/src/actors directory in CVS.
In the current system, the application will display the Java doc, unless there is doc info in the corresponding MOML file, which will override the Java doc. The system also examines the class and displays that name. An example of an actor with a MOML doc is the Scatterplot actor.
Documentation about ports and parameters should only be pasted into the actor if it differs from its base class. Will this work? Now that I'm writing this up, I'm a bit confused. We're updating the MOML file, right? Will the class MOML file overwrite a child's Java docs?
- KR- SMS ontology -- what kind of documentation is needed? Why? -- example -- ask laura for appendices and one-page description.
Eml dataset and link to morpho
Attachments:
|
