* ImmutableMultiset_CustomFieldSerializer. But that type has no methods (since it's never actually
 * used). We could probably fix the problem by adding dummy methods to that class, but that is
 * starting to sound harder than taking the superclass approach, which I've been coming to like,
 * anyway, since it doesn't require us to declare dummy methods (though occasionally constructors)
 * and make types non-final.
 */

  static final int SAMPLE_MASK = 0xFF;

  /**
   * Compute arrays of valid unicode text, and store it in 3 forms: byte arrays, Strings, and
   * StringBuilders (in a CharSequence[] to make it a little harder for the JVM).
   */
  @BeforeExperiment
  void setUp() {
    final long seed = 99;
    final Random rnd = new Random(seed);
    strings = new String[SAMPLES];
    for (int i = 0; i < SAMPLES; i++) {

then, when you've cleared all the jelly-fish out of the way--'

  `THAT generally takes some time,' interrupted the Gryphon.

  `--you advance twice--'

  `Each with a lobster as a partner!' cried the Gryphon.

  `Of course,' the Mock Turtle said:  `advance twice, set to
partners--'

  `--change lobsters, and retire in same order,' continued the
Gryphon.

  static final int SAMPLE_MASK = 0xFF;

  /**
   * Compute arrays of valid unicode text, and store it in 3 forms: byte arrays, Strings, and
   * StringBuilders (in a CharSequence[] to make it a little harder for the JVM).
   */
  @BeforeExperiment
  void setUp() {
    final long seed = 99;
    final Random rnd = new Random(seed);
    strings = new String[SAMPLES];
    for (int i = 0; i < SAMPLES; i++) {

    // we try hard to make them robust in practice.  We could additionally try to add in a system
    // load timeout multiplier.  Or we could try to use a CPU time bound instead of wall clock time
    // bound.  But these ideas are harder to implement.  We do not try to detect or handle a
    // user-specified -XX:+DisableExplicitGC.
    //
    // TODO(user): Consider using
    // java/lang/management/OperatingSystemMXBean.html#getSystemLoadAverage()

 * ImmutableMultiset_CustomFieldSerializer. But that type has no methods (since it's never actually
 * used). We could probably fix the problem by adding dummy methods to that class, but that is
 * starting to sound harder than taking the superclass approach, which I've been coming to like,
 * anyway, since it doesn't require us to declare dummy methods (though occasionally constructors)
 * and make types non-final.
 */

    // we try hard to make them robust in practice.  We could additionally try to add in a system
    // load timeout multiplier.  Or we could try to use a CPU time bound instead of wall clock time
    // bound.  But these ideas are harder to implement.  We do not try to detect or handle a
    // user-specified -XX:+DisableExplicitGC.
    //
    // TODO(user): Consider using
    // java/lang/management/OperatingSystemMXBean.html#getSystemLoadAverage()

 * by individual libraries that provide specialized support for particular use cases.
 *
 * <p>Disadvantages of EventBus include:
 *
 * <ul>
 *   <li>It makes the cross-references between producer and subscriber harder to find. This can
 *       complicate debugging, lead to unintentional reentrant calls, and force apps to eagerly
 *       initialize all possible subscribers at startup time.

then, when you've cleared all the jelly-fish out of the way--'

  `THAT generally takes some time,' interrupted the Gryphon.

  `--you advance twice--'

  `Each with a lobster as a partner!' cried the Gryphon.

  `Of course,' the Mock Turtle said:  `advance twice, set to
partners--'

  `--change lobsters, and retire in same order,' continued the
Gryphon.