Thursday, September 19, 2019, 10 AM US Eastern Time

Connection Info

To join the online meeting:

Attendees

  1. Ralph O'Flinn
  2. Mike Conlon
  3. Christian Hauschke
  4. Tatiana Walther
  5. Anna Kasprzik
  6. Ralph O’Flinn
  7. Brian Lowe
  8. Marijane White

Resources

Ontology Interest Group Google Folder https://drive.google.com/drive/u/0/folders/1RGBh4fDZdzpJdwyiUMO8OPWwkcmVYrI0

Agenda

  1. Updates
  2. Ontology at the conference
    1. Poster on domains https://vivoconference.org/vivo2019/schedule/#session-805
    2. Talk on ontology development  https://vivoconference.org/vivo2019/schedule/#session-36
  3. Discussion of Early Thoughts regarding representing people in VIVO.  http://bit.ly/31GTqD2
  4. Next meeting October 3.

Google doc for notes: http://bit.ly/31Cr8cU

Notes

  • Short sneak preview of the “Early Thoughts on an Impact Ontology” by Svantje, Eva, Tatiana and some others at TIB.
  • Presentation of VIVO Ontology 2 proposal at the VIVO conference. Now software and ontology are too close together. Needs separation to improve both in a more easy way. Additionally it would be an opportunity to get to a better ontology
  • Thoughts on how to rewrite or reuse existing scripts have to be considered in the process. Idea: Should work automatically and mechanical.
  • Early thoughts on persons in VIVO: https://docs.google.com/document/d/1wd-53Bus0974d7CvNgm8inQ5bbiw-JilbLpAvT_ydnU/edit
    • We understand types
    • It’s possible to have roles without specifying which role (person has role). That would be really helpful.
    • Multiple images should be possible. Preferences should be possible in other contexts, too. Rank is problematic in this context, if it’s not specified which kind of ranking it is (for example preference ranking).
    • Brian: That’s part of the application, not in the ontology domain; Mike: Preferences are part of the world, so should be part of the ontology. 
    • Brian: This has to be discussed in the future. If this is part of the domain, we have to be careful that someone can say that is my ranking with allowing conflicting views on the ranking. 
    • Addresses: Main issue with vCard: It’s not related with any other ontologies. A lot of stuff we do not need (romantic dates etc.). A lot of duplicates etc. due to the use of vcards., suboptimal modeling of things like telephone numbers. But VIVO needs an address model for virtual and physical addresses. MJ: Schema.org did a good job in this domain. Should not reinvent something existing.
    • Email addresses: Schema.org has to be investigated. Email for persons, orgs, contact points etc.
    • Websites have been someone’s work until now in VIVO. Things have websites, other things have websites. This distinction has to be made.
    • Names: Names are more complicated than we can cover now. Pseudonyms. Groups of people who publish under a single name (https://en.wikipedia.org/wiki/Nicolas_Bourbaki).
      • Earlier VIVO versions were friendlier regarding nicknames.Nicknames are very popular in the USA. Latin American names are also very specific. There could be e.g. two last names. Example: Lionel Andrés Messi Cuccitini. No one in the world uses the last name, only name 1 and 3. .
    • We should be better at generalizing the recording of names in VIVO. It needs more research on how to do this.
    • RDFS:label is fine for display. Schema.org is very FOAF-like in this domain. 
    • Titles could come from wherever and names are given. Titles are often confused with academic titles.
    • Positions are more formal. Positions have titles. People and organisations have roles in positions. It would be good if we could represent positions in a general way. We should reexamine the idea of a position. Position is an intermediate object between a person and an organization. Are positions things or proccesses?


Action Items

  • No labels