* {@link ListenableFutureTask} instances to the given delegate executor. Those methods, as well
   * as {@code execute} and {@code invokeAny}, are implemented in terms of calls to {@code
   * delegate.execute}. All other methods are forwarded unchanged to the delegate. This implies that
   * the returned {@code ListeningExecutorService} never calls the delegate's {@code submit}, {@code

  }

  /**
   * This implements a "bursty" RateLimiter, where storedPermits are translated to zero throttling.
   * The maximum number of permits that can be saved (when the RateLimiter is unused) is defined in
   * terms of time, in this sense: if a RateLimiter is 2qps, and this time is specified as 10
   * seconds, we can save up to 2 * 10 = 20 permits.
   */
  static final class SmoothBursty extends SmoothRateLimiter {

 *
 * <h3>Capabilities</h3>
 *
 * <p>{@code ValueGraph} supports the following use cases (<a
 * href="https://github.com/google/guava/wiki/GraphsExplained#definitions">definitions of
 * terms</a>):
 *
 * <ul>
 *   <li>directed graphs
 *   <li>undirected graphs
 *   <li>graphs that do/don't allow self-loops
 *   <li>graphs whose nodes/edges are insertion-ordered, sorted, or unordered

 *
 * <h3>Capabilities</h3>
 *
 * <p>{@code Network} supports the following use cases (<a
 * href="https://github.com/google/guava/wiki/GraphsExplained#definitions">definitions of
 * terms</a>):
 *
 * <ul>
 *   <li>directed graphs
 *   <li>undirected graphs
 *   <li>graphs that do/don't allow parallel edges
 *   <li>graphs that do/don't allow self-loops

 *
 * <h3>Capabilities</h3>
 *
 * <p>{@code Graph} supports the following use cases (<a
 * href="https://github.com/google/guava/wiki/GraphsExplained#definitions">definitions of
 * terms</a>):
 *
 * <ul>
 *   <li>directed graphs
 *   <li>undirected graphs
 *   <li>graphs that do/don't allow self-loops
 *   <li>graphs whose nodes/edges are insertion-ordered, sorted, or unordered
 * </ul>
 *

 *
 * <h3>Capabilities</h3>
 *
 * <p>{@code Graph} supports the following use cases (<a
 * href="https://github.com/google/guava/wiki/GraphsExplained#definitions">definitions of
 * terms</a>):
 *
 * <ul>
 *   <li>directed graphs
 *   <li>undirected graphs
 *   <li>graphs that do/don't allow self-loops
 *   <li>graphs whose nodes/edges are insertion-ordered, sorted, or unordered
 * </ul>
 *

 * <p>Elements of a multiset that are equal to one another are referred to as <i>occurrences</i> of
 * the same single element. The total number of occurrences of an element in a multiset is called
 * the <i>count</i> of that element (the terms "frequency" and "multiplicity" are equivalent, but
 * not used in this API). Since the count of an element is represented as an {@code int}, a multiset

 *
 * <pre>{@code
 * Zachary: [Taylor]
 * John: [Adams, Adams, Tyler, Kennedy]  // Remember, Quincy!
 * George: [Washington, Bush, Bush]
 * Grover: [Cleveland, Cleveland]        // Two, non-consecutive terms, rep'ing NJ!
 * ...
 * }</pre>
 *
 * <h3>Views</h3>
 *
 * <p>Much of the power of the multimap API comes from the <i>view collections</i> it provides.

 * <p>Elements of a multiset that are equal to one another are referred to as <i>occurrences</i> of
 * the same single element. The total number of occurrences of an element in a multiset is called
 * the <i>count</i> of that element (the terms "frequency" and "multiplicity" are equivalent, but
 * not used in this API). Since the count of an element is represented as an {@code int}, a multiset

          // caught and forwarded to the service the task would stop executing but the service would
          // have no idea.
          // TODO(lukes): consider building everything in terms of ListenableScheduledFuture then
          // the AbstractService could monitor the future directly. Rescheduling is still hard...
          // but it would help with some of these lock ordering issues.