|
This is version 4.
It is not the current version, and thus it cannot be edited.
[Back to current version]
[Restore this version]
- Taxon group
- Considering whether relationships should have global identifiers, and implications for how taxonomists use the system
- Mammal data from ITIS put into the exchange standard, waiting on hickory data
- Kepler group
- Substantial discussions on user interface design and ability for semantic abstraction of workflows
- Preliminary actor for running R scripts in Kepler completed
- EcoGrid group
- Almost complete with new resultset format on all four platforms (Metacat, SRB, DiGIR, Xanthoria)
- EOT group
- New website prototype ready, will be placed in production next week (replacing the existing SEEK web site). Comments appreciated, see the prototype web site
 .
Completed tasks
- tested Xanthoria EcoGrid codes on Axis, without globus on windows machine. It works fine.
Current work:
- Currently, I am working on two things:
- Making changes to existing codes to make it compatible with latest resultset schema (getting rid of AnyRecordType)
- Trying to install and configure GT3.2 on a development machine (linux box), so that I can test Xanthoria and then maybe check in the codes to seek cvs.
Significant issues
- Currently I am interested in the following issues:
- Configuring GT version 3.2-how to use the certificates? Will the older ones work or do we need to create again? What’s the procedure for installing and configuring GT 3.2?
- Is anybody working on setting up GIIS and MDS? What is the current status with this respect?
Completed Tasks/ Bugs Fixed:
- Contributed to the SEEK/Kepler UI discussion.
- Highlights: There is consensus building that a higher level conceptual design tool will be needed that more closely represents how models are constructed by the scientists.
- Dan constructed several screen shots for talking points at http://www.nceas.ucsb.edu/~higgins/BoxesAndArrows
- The filling of the usability engineering position is key to future progress on all UI issues (although there is a lot of progress yet to be made).
Current Work and Bugs:
- Start on Bug 1595 - Create Advance Query Builder UI
- Continued investigation into which Swing features can be leveraged to create the UI.
- In general, the top level of components makes it easy to quickly prototype the general window. But as always the devil is in the details. Here is a small sample of some of the issues:
- The JDesktopPane was not made to expand and shrink as elements are placed "off screen", and items "off-screen to do paint correctly when dragged back
- Painting is problematic in the desktop because of clipping issues
- Hardest: Because Swing uses a single delegate to do all the rendering into a buffer, there are no "real" visual "frames" for each item in the JList that represent a field in a table. This makes it extremely difficult to track the scrolling of table and make the "connection lines" between the tables scroll appropriately.
Significant Issues
Completed Tasks/ Bugs Fixed:
- Attended design & implementation meeting at Chapel Hill about the Concept Review System
- reported current status
- presented a demo
- discussed revision & next-step developemnt
- Tested and revised auto-email function
Current Work and Bugs:
- developing Auto-Manager for Concept Review System to automatically complete some system administration tasks, such as sending emails,assigning proposals to peer-viewers.
Significant Issues
Completed tasks/bugs fixed
Current Work and Bugs (In order of time/effort)
- GrOWL editor: integrated new version of OWLAPI, redesigned and began re-implementing OWL file saving.
- Continued working on the ontology for biodiversity indices.
- Review/comment on the inter-ant parasitism semantic mediation prototype paper. Paper is now submitted.
Significant issues
Completed tasks/bugs fixed
Current Work and Bugs
- working on Bug 1576 - implement enterProviderData. Robert has converted some ITIS data to the current Taxonomic Exchange Schema (still waiting on hickory data) to test parsing the TES.
Significant issues
- Still concerned with the definition of a GUID-bearing concept as containing relationships. This seems to reduce the utility of our service to scientists wishing to markup or search on datasets with a name string.
Completed Tasks/ Bugs Fixed:
- Revised the collaboration and class diagram draft design for bug 1584 and 1586
- Started to code for data query engnie with hsql
Current Work and Bugs:
- Working on the code for data query
Significant Issues
Completed Tasks/Bugs Fixed:
Current Work and Bugs:
- Continuing to work on the concept relationships manuscript, now turning to final sections; the work is directly related to discussions between KU and Napier last week.
- In exchange with KU and Napier about these issues (see e.g. http://tdwg.napier.ac.uk/phpwiki/index.php/What%20Gets%20a%20Guid and http://tdwg.napier.ac.uk/phpwiki/index.php/RelationshipTypes).
Significant Issues
- It seems like we're converging on a consensus that concept relationships may be authored (as an option) without also authoring concepts. KU made this a requirement, in essence, and Napier is now working on ways to accommodate it within the taxonomic exchange schema.
Completed Tasks/ Bugs Fixed:
- Completed some additional examples of ecology specific pipeline (Predetor/Prey detailed model) --- Bug 1336
- Completed a Simple R actor using the existing CommandLine Actor and and input script file (Bug 1342). R creates an image file which is then displayed by existing Ptolemy actors.
- Created a web page comparing various 'box and arrow' systems
Current Work and Bugs:
- Continuing work on a more general R-actor system. I have succeeded in getting a java system for running interactive command line systems like R running. This should make it easier to create customized actors from command line systems without having to write system specific jni code for every case.
- Working on updating to a Kepler system based on Ptolemy 4.
Significant Issues
Completed tasks/bugs fixed
- Growl
- Sketch of owlDLCompliance function
- Enabled editing of class/instances properties save for uri base
- Replaced ascii text for math symbol by respective standard math symbols from Unicode, which enhances look and feel
Current Work and Bugs
- Redesigning property panes and enabling the editing for constraints , properties, and other things on the toolbox
Significant issues
- Does anyone know how to gracefully (without raising an exception) check if a string is a valid URI without using external java libraries???!!! (this problem slows down my work on editing properties)
Completed Tasks/ Bugs Fixed:
- Bug 1604 - Convert ITIS database dump to TES
- Converted ITIS mammals only, as the requirements were to have a testing instance of the TES for implementation of enterProviderData
Current Work and Bugs:
- Minor data model changes to incorporate the schema fully
- Bug 1603 - Implementation of addConcept
- Bug 1604 - Implementation of addRelationship
Significant Issues
Completed Tasks/ Bugs Fixed:
- fixed the second bug mentioned in the last week report: the previous conflicted requirements on java version from running multiple threads on windows and the XQuery engine being used in our schema matching tool.
Current Work and Bugs:
- Work on 2 paper drafts
- Fix more bugs of the tool:
- make details of nodes accessible from the GUI,
- correctly handle union types of matches in the GUI,
- input domain thesauri from user specified text files instead of builtin ones
- Implement abstract schemas and abstract schema morphisms to make the tool more general.
Significant Issues:
- Regarding connecting the schema matching tool with Kepler, there is no problems of java version and performance on Windows machines any more.
Completed tasks/bugs fixed
- Working on user management of taxon server. Logging in and out works fine in struts on tomcat with mysql.
Current Work and Bugs
- Working on bug #1597 - Logging in and out works fine. I sent a note to Steve about what I'm doing, but haven't heard back yet - I emailed him not too long ago. He mentioned using the hibernate persistance layer in one of his emails a while ago, so I'm incorporating that into the server. It's slowing me down a bit, but I should be over the hump tomorrow.
- Finished the paper with Shawn Bowers for the Semantic Web and Databases conference.
Significant issues
- Wondering if there's any kind of unified authentication system for SEEK? I'm using struts for the authentication stuff I'm doing, so there's going to be a clean division between the view and the model. Who's the best ecogrid person to ask about authentication from the ecogrid perspecitve?
You can see the status reports for other time periods as well.
|
