define a scope for git_batch

[bjuergens]

Is git batch only the receive-hook, or also the run-script?

(I say the former, till says the latter)

[bjuergens]

in this repo will be:

• 1 receive hook per scheduler
  • at
  • condor
  • qsub (via condor)
  • generic (das was jetzt gerade da ist)
  • docker
• 1 simple generic example for testing
  • the run.sh will be independent from the scheduler
  • it will just execute nbconvert on some dummy-notebook with some dummy data
• 1 short example for each hook (maybe part of the global readme?)
  • examples in the readme are all about hooks and scheduler
  • they are not about dirhash and reproduction
  • they are only a little bit about sdil

We assume, that the user will use these hooks as examples/boilerplates and change them according to their needs.

Open questions:

• Will the modified receive hooks be on the same level as the scheduler-specific hooks?Or will they be created in the tutorial?

[bjuergens]

the current tutorials about dirhash and reproduction will be moved to a new repository. The new name is not yet decided. Maybe "tutorials-for-reproducible-datascience" or "example-workflows" or something like that