...
Please add additional agenda items or updates – we always welcome suggestions
Updates
- Colorado
- Cornell
- Duke
- Florida – starting to work with Peoplesoft – the way to insulate us is to provide a BizTalk interface. Would give an initial load and anytime people change something another dump of the whole record would be sent as a message. Talking about a live interface for VIVO – a web service that listens, grabs it, and writes to a folder that the Harvester would read every 2 hours.
- Indiana
- Johns Hopkins – still working on access to databases of record
- North Texas – Library iSchool conference in Vancouver went well
- Stony Brook – check out http://reach.suny.edu. Had to remove triples, but got an error with recomputing inferences – a lost connection. Upped the maxconnections in MySQL and a few other parameters and running it again to see what's happening. Got positive feedback from the first campus to look at it.
- Weill Cornell – working on reports - generated using SPARQL. Trying to see how the data are populated for each profile – sent some questions to the list about the time required to run some queries. Looking to verify that the profiles are minimally populated – trying to get a sense statistically of basic properties. Trying to improve the queries. Using the filter (bound) – but only has one optional block; could put each in its own optional block or use unions to improve performance. To get the ones that are all bound, you can take out the optional and the filter.
- Any other sites represented
Trip Reports
- VIVO Community Dayin Melbourne and other discussions about VIVO and DataStaR, a project at Cornell and Washington University to extend the VIVO ontology and code base to serve as a registry for research data sets
- Stephen Williams
- Huda Khan
- Brian Lowe
Several days with people at the University of Melbourne, plus a brief meeting with the Australian National Data Service.
The VIVO community day was attended by upwards of 50 people from universities around Australia – a mixed crowd of both developers and administrators. The afternoon had sessions for Australian universities to show off their VIVO implementation. Griffith University has not only done skinning and theming but has developed more dynamic pages using Ajax and Javascript. They had apparently had a mandate to have page load times that don't exceed a second and a half, so they had to introduce Squid caching (as a number of large websites) on the public side of their site. They don't have the caching turned on while editing since there's no easy mechanism for determining when any single element on the page has changed. They had also worked on a replacement for Flash for visualizations using JIT and HTML5. They're also using Ajax Solr to get the content back, plus SlickGrid JQuery library. Have shared this in a Git repository. Would be good to include them in these calls – 4 pm our time would be 8:00 am their time. Trying to identify a top-ten list of publications for each person, but running into some political challenges.
QUT – http://researchdatafinder.qut.edu.au/vivo/
U Melbourne – implemented at tabbed interface and modified the search results for an alternate Who's Who service to give a media person a quick reference about the person, following the original intent of their Find an Expert. They also replace their VIVO search with Funnelback. They are also analyzing people's co-author networks to group them into cliques – a pie graph for each clique with your publications highlighted. Additional line graphs available if click for more detail. David Cliff and Martin worked on it – using a Javascript library called Raphael. VIVO at Melbourne will just be for display, not editing, because the government already has a reporting requirement that has resulted in centralized systems for faculty data. VIVO will be used as a public profile and for linked data. They have licensed Symplectic Elements for publications, but that will push data into their central system rather than into VIVO directly.
Richard Ferrers at ANDS had used a program called Gource (http://code.google.com/p/gource/wiki/Videos) – can run it against the VIVO svn data to see all the commits going into the repository. Using the theory of watching logs, could watch datasets coming into their database and show how they were linking together. An organic-looking visualization.
Other universities – Curtin is working toward a production site in 1.3; University of Western Australia wants to connect VIVO with their student management system since they want to get graduate students in – for the thesis information, but they have some of the same privacy restrictions.
University of Western Sydney is using Redbox with a VIVO ontology – it has a very simple editing interface as it's driving appeal.
...
- EuroCRIS(European Current Research Information Systems) meetings in Bath, England and the new Linked Data Task Force
- Jon Corson-Rikert
Upcoming conference presentations
- American Association of Universities Data Exchange, March 12-14
- American Medical Informatics Association, March 19-24
- Stanford library development conference, LibDevConX, March 26-28
- Internet2 Conference, April 23-25
- CTSA Ontology Workshop, April 25&26
- VIVO Implementation Fest, May 14-15 or 16
VIVO Conference Planning
Simon Porter from University of Melbourne has agreed to serve as the Program Chair for the 2012 VIVO Conference, which will be held August 22-24 in Miami, Florida. Simon has a number of great ideas for speakers and for conference and workshop tracks, but would also like input from the development community.
More next week ...
Notable Development List Traffic
- finding where context nodes are missing expected people or institution links
- CSV ingest questions – Nick Veenstra each line needs double quotes around it, and each field that contains a space needs double doublequotes, and the file also had to be UTF8 for the system to be able to parse it. The CSV in the harvester uses the H2 database method so shouldn't have issues with spaces; if Nick can share an example file we may be able to reproduce it.
- The harvester uses the H2 method of digesting csv files.
- finding missing links on context nodes (authorships, positions, roles), usually to people or organizations
- implementing full text linking - Tammy appends the DOI to http://dx.doi.org/10.1016/j.prot.2010.12.012 – will depend on the user's home university subscriptions
- QRcode error message filling VIVO logs
- Tammy – deltacomputer.run error even though the recompute inferences ran with a longer wait timeout value in MySQL. Maybe something in the code is holding onto a database connection.
- request for materials in other languages – Kristi Holmes established a page for listing International VIVOs
- VIVO in Progress Google map of VIVO implementation and development activity
Call-in Information
1. Please join my meeting. https://www1.gotomeeting.com/join/322087560
...
Dial +1 (773) 897-3008
Access Code: 322-087-560
Audio PIN: Shown after joining the meeting
Meeting ID: 322-087-560