The following paragraphs describe the requirements for the paper.

• Topic
  
• Grading
  
• Deliverables
  
• Due Date
  
• Guidelines
  

Topic.

Your paper shall be a review and discussion of some recent topic in databases. The topics are not confined to the course syllabus. It must directly discuss (and reference!) at least two related papers published within the last six years from reputable technical journals or proceedings from related conferences. Note that books or chapters from books, by themselves, do not satisfy this requirement.

You must register your topic with me by January 30, 2001 (via email) and submit the final paper by 5 P.M. on March 1, 2001 (hardcopy).

To register, you must submit a title, a short abstract, and a bibliography. The bibliography must include at least the papers you are planning to discuss. All of the above must be e-mailed to my account (srini@cis.ohio-state.edu).

If you find that you are having difficulty locating materials, please see me about it.

Grading.

My evaluation of your paper will, to some extent, be subjective. However, there are certain eternal verities.

You will receive a letter grade (using an A+ through F scale) for your paper. Because of the importance of clarity in the presentation of information, approximately 40% of your grade will be based on how you wrote your paper, with the remaining 60% based on what you wrote. When determining the style component of your grade, I will consider (among other things) your spelling, punctuation, and grammar, as well as the overall readability of your paper. (To forestall the inevitable question, yes, I do know that this isn't an English class. The ability to express yourself clearly in written works, however, is of critical importance in today's world regardless of your chosen profession, hence its inclusion as part of your grade on this assignment.)

Note also that plagiarism will not be tolerated; if you feel the need to include portions of a textbook or article in your paper, remember to attribute them properly (footnotes or endnotes are helpful here). There will be a 10% penalty, per day late, for late registration as well as for late submission (5:01PM on March 1 will be deemed late). Be on time.

Deliverables.

Your paper should have a cover page stating the title of the paper, your name, the course name, the quarter, my name, and the due date. Please do not use a binder of any sort; staple the paper once, in the top left-hand corner. You must turn in two copies of your paper - one for grading, and one for my records. In addition, you will also be expected to turn in copies of the primary journal/conference articles that you used; these will be returned if you ask for them.

Due Date.

Both copies of your paper, and copies of your journal articles, are due in class on March 1, 2001 at 5 PM (sharp). Papers will not be accepted after that time.

Guidelines.

The following guidelines are presented for two reasons: to help guide you in determining how to write your paper, and to help simplify the grading process for term papers. In particular, the format restrictions mentioned below are intended to make it easier for me to read your paper (for grading purposes) while not making it harder for you to write your paper; as such, I will expect you to abide by them. In most cases, the restrictions are similar to the standards of many professional publications in this field.

1. Quantity. The paper should be long enough to say what needs to be said and short enough to be comprehensible in my lifetime while giving reasonable coverage of the topic. It is hard to say anything of real substance in less than 8-10 pages (double spaced, 11/12 pt); rarely, though, should a paper for this assignment require more than 15 pages.

2. Quality. At a minimum, your paper should have a cover page, the text itself, and a bibliography. You need not provide a table of contents, although one would certainly be appropriate if you choose to organize your paper into sections and subsections. You may use either footnotes or endnotes, although footnotes are preferable.
   The question of coverage of broad topic areas typically arises. In general, a paper which covers fewer major ideas but goes into each idea in more detail is better than a paper which attempts to touch on every idea in an area but doesn't get into any idea in depth.
   The content of your paper will be evaluated as follows: First, I will evaluate whether you have understood the papers under discussion and have expressed yourself appropriately on the strengths and weaknesses of the results. Second, I will evaluate you on your ability to critique the work under discussion. Along these lines, some questions that you might want to address could include: a) what are some of the assumptions of the study, and are these assumptions realistic; b) what are the limitations of the study and how can these limitations be overcome; c) how does the work described compare to prior work; and d) what is/was the impact of this work.

3. Format. Your paper must be typewritten on 8.5" x 11" paper. Line-printer paper is acceptable, provided that it is the appropriate size. Each page must be double-spaced (not single or triple spaced!), and must have margins of one inch on all sides. Pages should be numbered.
   The bibliography often gives people problems. Minimally, a bibliography entry for a book should include the author(s), the title, the publisher, and the year of publication. When including a reference to a journal or magazine article, the volume and/or issue number should also be included. Any of the ``standard'' bibliography formats are acceptable; e.g., for a book,

   [BOND76] BONDY, J. A., and MURTY, U. S. R.. Graph Theory with Applications. North Holland, N.Y., 1976.

   or for a paper,

   [FORD56] FORD, L. R. Jr. and FULKERSON, D. R.. Maximal flow through a network. Canad. Jour. Math., 8 (1956), 399-404.

   There are numerous technical writing textbooks available in the library which will give you more information on bibliographies.
   You may want to investigate using a text formatting system to prepare your paper. Most text formatters will automatically handle such details as numbering of pages, creating of section and subsection headers, and generation of footnotes. If you use a formatter which allows you to choose the font for your paper, select one intelligently; many Macintosh formatters, for instance, will quite happily let you select fonts which are utterly unreadable. If you can select the type size to be used, use 11/12-point type; anything smaller is hard to read, and anything larger wastes paper (and artificially inflates the page count).

4. Journals.

   The following is a sample list of journals and conference proceedings. Don't restrict your search to these. However, they may provide a good place from which to start your search.

   SIGMOD Newsletter
   Very Large Databases (VLDB)
   ACM SIGMOD conference
   IEEE Transactions on Knowledge and Data Engineering
   ACM Transactions on Computer Systems
   International Conference on Data Engineering
   ACM SIGKDD conference on Knowledge Discovery and Data Mining
   IEEE Computer