@Override
  protected ArrayTable<String, Integer, Character> create(@Nullable Object... data) {
    // TODO: Specify different numbers of rows and columns, to detect problems
    // that arise when the wrong size is used.
    ArrayTable<String, Integer, Character> table =
        ArrayTable.create(asList("foo", "bar", "cat"), asList(1, 2, 3));
    populate(table, data);
    return table;

  @Override
  protected ArrayTable<String, Integer, Character> create(@Nullable Object... data) {
    // TODO: Specify different numbers of rows and columns, to detect problems
    // that arise when the wrong size is used.
    ArrayTable<String, Integer, Character> table =
        ArrayTable.create(asList("foo", "bar", "cat"), asList(1, 2, 3));
    populate(table, data);
    return table;

       *   the case that the service is using CustomScheduler). (It needs it so that it can cancel
       *   the task and detect whether it has been cancelled.)
       *
       * - ReschedulableCallable has a reference back to its enclosing CustomScheduler. (It needs it
       *   so that it can call getNextSchedule).
       *

   * 'affect' a bit(j) in the output if two inputs, identical but for bit(i), will differ at output
   * bit(j) about half the time
   *
   * <p>Funneling is pretty simple to detect. The key idea is to find example keys which
   * unequivocally demonstrate that funneling cannot be occurring. This is done bit-by-bit. For each

   *       collections intelligently fall back to a binary search tree if hash table collisions are
   *       detected. Rather than going to all the trouble of reimplementing this ourselves, we
   *       simply switch over to use the JDK implementation wholesale if probable hash flooding is
   *       detected, sacrificing the compactness guarantee in very rare cases in exchange for much
   *       more reliable worst-case behavior.

             * exception in seenExceptions. When we complete secondFuture with the same exception,
             * we want for AggregateFuture to still detect that it's been previously seen.
             */
            secondFuture.setException(sameInstance);
          }
        },
        directExecutor());
    firstFuture.setException(sameInstance);

    // This is some basic recursion detection for when people create cycles via set/setFuture or
    // when deep chains of futures exist resulting in a StackOverflowException. We could detect
    // arbitrary cycles using a thread local but this should be a good enough solution (it is also
    // what jdk collections do in these cases)
    try {
      if (o == this) {

     * <tt>S-X-Y-Z</tt> strings of fragments of).
     * <p>
     * Alternatively <tt>getSecurity(true)</tt> may be used to resolve all
     * SIDs together and detect network failures.
     * 
     * @return array of ACEs
     * @throws IOException
     */
    ACE[] getSecurity () throws IOException;


    /**

   *       collections intelligently fall back to a binary search tree if hash table collisions are
   *       detected. Rather than going to all the trouble of reimplementing this ourselves, we
   *       simply switch over to use the JDK implementation wholesale if probable hash flooding is
   *       detected, sacrificing the compactness guarantee in very rare cases in exchange for much
   *       more reliable worst-case behavior.

     * Module information for the path specified at construction time.
     * This map is usually either empty if no module was found, or a singleton map.
     * It may however contain more than one entry if module hierarchy was detected,
     * in which case there is one key per sub-directory.
     *
     * <p>This map may contain null values if the constructor was invoked with {@code resolve}