Talk:CRIG Unconference

From DigiRepWiki

From last night

  1. Will repositories care about ORE?
  2. Does sceario-oriented planning prevent out of the box thinking?
  3. Make more mistakes rapidly
  4. Standards are like sausages: if you like them you don't want to see how they are made.
    1. (unless you are an engineer, in which case, you don't want to see a machine failure!)
  5. Will there be repositories in 10 years, 5 years, 2 years. How do we get there?
  6. Any arguments about packaging/metadata formats invokes Godwins law.
  7. Different types of repository software are better for different content types
    1. Does this mean we're giving/creating developers artificial horizons?
    2. Is this a problem or a feature
  8. Institutions need policy. It doesn't really matter where store it, institutions have to have a coherent approach. As many repositories as you want but managed at a policy level not a technical one (though this can help).
    1. Can the people who can't understand the problem make policy decisions?
  9. Data is preservable! Metadata is too version specific to be preserved.
  10. Access management is related to the core business of a University.
    1. Most institutions do not consider the value of digital assets, especially those relating to teaching/learning.
  11. NO MORE STANDARS: The ones we have are less than adequate
  12. This is as good as it gets. Development is over. Accept it.
  13. What can we learn from the porn industry, they are always the first to implement the technology that any idiot can use.
  14. Forget Google: they are not competitors or our friends.
  15. Move past DC -> Object profiles
  16. Author list changes from version to version on a paper.
  17. Telepathic metadata capture
  18. Public / Private control in Repositories
    1. Dark Archive
  19. Don't prevent people from getting on with it!
    1. Lower the barrier to participation
  20. Repositories are just digital shelving
  21. Its just data.
    1. Each dataset needs an API
    2. Common datasets should have common APIs
  22. Dark Archives: niches for expressiveness, privacy (ultimately), viewable but version control, can hide creative process, considered private until chosen to be public.
  23. What's the point in reinventing the wheel.
  24. It's ok I have solved everything (4 beers)
  25. We need to start with a model that works and make it better
    1. Google works, Flickr works: why don't we just do it like them?
  26. if it can't be done in a year it's never going to happen