notesThis is a featured page

Searching for a mission statement

First stab: Defining the necessary and the awesome characterstics of a killer app. What's a "flickr for linguists" like?

Thinking about what part of the research process would be most suited for a killer app:
  • data creation: maybe not so much, too diverse?
  • data sharing: yes
  • data analysis: yes. possibly built on top of data sharing platform
so could a platform for data sharing be the "flickr for linguists"?
  • immediate benefits:
    • backup
    • in the light of the first panel discussion: maybe the place where help with licensing could be provided


No user avatar
robert_forkel
Latest page update: made by robert_forkel , Jul 18 2009, 11:46 AM EDT (about this update About This Update robert_forkel Edited by robert_forkel

33 words added

view changes

- complete history)
Keyword tags: None
More Info: links to this page
Started By Thread Subject Replies Last Post
nancyide Please look at LAF/GrAF 0 Jul 18 2009, 12:31 PM EDT by nancyide
Thread started: Jul 18 2009, 12:31 PM EDT  Watch
I would like to ask you to have a look at the Linguistic Annotation Framework, which provides a graph-based "pivot format" which is intended to serve as the lingua franca for annotations schemes (in terms of form, not content). The idea is that everyone uses his/her own format, provided it conforms to an abstract data model, and then there is a tool that can transform this into the generic graph format (GrAF). In this form, annotations can be easily merged and compared, and -- most interestingly for your group -- transduced to any other format that a given tool needs. Or, one would assume, someday tools might all use GrAF as input. GrAF is designed to be perfectly generic, and has been shown to be able to handle all kinds of annotations. We already have "Proff of concept" tools to transduce many formats into GrAF and tools to export GrAF to a wide variety of formats.

Even if you do not use LAF/GrAF itself, please look at the LAF documents (links to them on the wiki pages of Working Groups 1 and 7) because the ideas and principles are the result of 20 years of thinking on this topic.
Do you find this valuable?    
Keyword tags: None
Showing 1 of 1 threads for this page