Central articles:

* Sundar Balasubramaniam and Benjamin C. Pierce, "What is a File
   Synchronizer", Fourth Annual ACM/IEEE International Conference on
   Mobile Computing and Networking, 1998.

* Norman Ramsey, Elod Csirmaz, "An Algebraic Approach to File
   Synchronization", Foundations of Software Engineering, 175-185,
   2001.

* Tancred Lindholm, "A 3-Way Merging Algorithm for Synchronizing
   Ordered Trees: The 3DM Merging and Differencing Tool for XML",
   Master's thesis, Helsinki University of Technology, 2001.


Questions:

(File synchronization)

1. A basic assumption in ``What is a File Synchronizer'' is that
    the two replicas are in the same state prior to synchronization.
    What changes are required to relax this assumption?

2. How long can chains be with respect to the <= ordering on
    file systems?

3. Suggest a method to allow the state-based synchronizer to
    synchronize more than two replicas.

4. Why does the reconciliation algorithm specify a lexicographic
    order? Does order matter?

5. How can one find a sequence of operations that transform
    a file system F into a file system F'?

6. Show that conflicts depend on a particular sequence of
    operations.

7. Give an example that shows that the algebra is not complete.
    (Can you make the algebra complete?)

(Tree matching and merging)

8. The author claims that the use of a DTD would make his tool more
    powerful. In what ways could the DTD assist the tool?

9. Is there a cleaner solution than the 128 byte pruning of matching
    edges to avoid relating (small) unrelated data?

10.Would it have helped the author to work with DAGs instead of trees?