Draft Introduction

[RichardLitt]

Plan

• Definition of computational linguistics, and linguistic tooling
• Code as it pertains to lrl
• state of the field linguistically
• State of the field computationally
• Lack of sharing code or storing it usefully, due to factors: funding, academic cycle, inability, scope, lack of knowledge of domain
• Some shared code

In this paper, I will talk about:

• Open source code
  • Longevity of linguistic scholarship and work
  • Data, rights, liability, and privacy
  • Funding
  • Institutional bottleneck
  • Linguistic colonialism
  • Ethical and moral concerns for military usage
  • Ethical and moral concerns for big business usage
• Open Source work currently available
  • Case study on GitHub, SourceForge, some archival sites (UPenn, Max Planck, DFKI)
• Case study using endangered-languages repository
  • Clean up resource
  • Add all listed resources in issues
  • Contact and create the LSA CELP Technology Subcommittee
  • Clone all SourceForge repositories
  • Rename to low-resource-languages
  • "List quality"
  • "the pages and subpages are often dead"
  • Get diagnostics on the state of the links I've found:
  • What percentage have been updated when
  • Downloaded, etc.
  • Review Excel results
• Peer-to-peer solution for sharing code
  • Stub out example
  • Build a web searcher for automatically getting and sharing code Further Work:
  • Open source data repositories (touch on)
  • Working with Ethnologue Conclusion

[RichardLitt]

Largely done or superseded by other plans.