KNOWN-PROBLEMS

MAIN INSTANCE:

  - VALUELESS TAGS

    Creating tags with no value is failing against both the main instance
    and the sandbox.   There is a known problem with FluidDB causing
    this, which has been fixed internally, but not yet pushed to the
    sandbox or the main FluidDB instance.

    So in the current release, one test fails against the main instance
    and also against the sandbox (though see below).   This test is

	* TestFluidDB.testAddValuelessTag


  - NAMESPACES

    Although FDB will happily read tags with long names (njr/foo/bar/bas)
    it will not currently create them, nor can it set tags with such
    names.   (This is simply unimplemented.)

  
SANDBOX:

  - FAILING TESTS

    Although

	fdb test -sand

    will now cause the tests to run against the sandbox, several fail
    because they assume fixed object IDs that are correct in the main
    instance but different in the sandbox.   This is deliberate, but
    at some point I plan to change it so they will/can run against
    any intance.

  - HANGING TESTS

    Several tests cause a hang against the sandbox, making them hard
    to use.   I believe these are because of (new) bugs in the sandbox
    when working with non-existent IDs.   Though, I could be wrong, and
    and they could simply be bugs in FDB (though it seems OK against the
    main instance).

    So in practice, the only tests that can be run against the sandbox
    right now are the utils (which don't actually talk to FluidDB).

	python fdb.py testutil -sand

    Sould produce 4 passes at present.