CS145 - Fall 2011
Introduction to Databases

Assigned Work

Assignments in CS145 include four different types of work:

Automated quizzes
Challenge problems
Database exercises using (directly or indirectly):
- xmllint, for validating XML data sets
- An interpreter for relational algebra
- SQL on the SQLite, MySQL, and Postgres open-source relational database systems
- XPath, XQuery, and XSLT on the Saxon system
A three-part programming project, using the SQLite database library and the PHP scripting language

All work in CS145, including the programming project, is completed individually, not as teams.

Automated Quizzes

The purpose of the automated quizzes is to ensure that you have learned the basic material in CS145. You will perform the exercises using a quizzing approach pioneered by Gradiance. (Gradiance was launched at Stanford in a 2002 offering of CS145, it did time as a commercial product partenered with Pearson Education, and now it's offered as a free service.) The automated exercises are based on "root questions," from which the system generates different instances of the quiz each time you take it. Thus, you can test yourself on the same material multiple times, and you receive immediate feedback for each attempt. We strongly encourage you to continue testing on each topic until you complete the quiz with a perfect score at least once.

Challenge Problems

Challenge problems are more difficult and open-ended than the exercises in the automated quizzes. The challenge problems are designed to make you think more deeply about the topics covered. Some (but not all) of the problems are by design very challenging. We do not expect most students to completely solve every challenge problem in the course. Grading of challenge problems is on a relatively coarse scale, and credit is given if you make a good attempt at a problem in the right direction. (See Challenge Problem Grading for more details.)

Database Exercises

For many of these exercises (but probably not all), we will provide an automated workbench for you to interactively develop, test, and submit your solutions.

You will create Document Type Descriptors (DTDs) for XML data sets, validating the data sets against your DTDs using the xmllint tool.
You will write relational algebra queries and execute them using a relational algebra interpreter called RA. The RA system was developed by Prof. Jun Yang at Duke University.
You will write queries and other commands in the SQL language and execute them using the SQLite, MySQL, and Postgres open-source relational database management systems.
You will develop and run queries over XML data sets using the XPath, XQuery, and XSLT languages supported by the open-source Saxon system.

Programming Project: AuctionBase

Your programming project is to build an online auction system called AuctionBase. The project is in three parts, included in three of your later assignments. At the back-end, AuctionBase manages all of its data using the SQLite database management library. At the front-end, AuctionBase provides a friendly web interface to its users. AuctionBase is modeled roughly after the eBay online auction system. In fact, we will give you a considerable amount of real data downloaded from eBay quite some years ago (courtesy of the University of Wisconsin), to initially populate your AuctionBase system.

We will expect certain minimal functionality in each AuctionBase system -- beyond that, possibilities are unlimited. Minimal functionality includes a variety of queries and browsing capabilities over the current items up for auction, the bids on those items, and the sellers and bidders. AuctionBase also must provide a means for entering bids, and for concluding auctions on individual items. Various integrity constraints must be monitored. Although AuctionBase uses SQLite and therefore has transaction support, multi-user issues are not a focus of the project.

Web programming skills are not required for the AuctionBase project. We will provide skeleton PHP code for most aspects of the web interface. Those students savvy in web programming may rewrite or extend the interface as they please.

At the end of the course, there will be a small contest in which we select a few of the best AuctionBase systems to be demonstrated in class and linked to the course home page. No extra credit is given, but winners will be invited to enjoy lunch at the Stanford Faculty Club with Prof. Widom, the TAs, and the other contest winners.

The scope of the AuctionBase project is such that it can easily be completed by each student in the class individually -- partners or teams will not be used.

Programming and System Issues

To develop the project all students will use the open-source SQLite relational database management library and the PHP scripting language, running under Unix. (See Interfacing with Stanford's Unix Machines.) An initial part of the project involves standalone parsing and processing of XML data, which is typically done using Java. We will assume students are proficient already with Unix and Java, but we are not assuming familiarity with web or database programming -- help sessions will be provided.

In general, students are welcome to use alternative languages, systems, and tools if they wish. However, only SQLite and PHP on the Stanford Unix Corn machines will be supported by the course staff, and the specifications for the project must be met regardless of the languages or tools used. We will provide an XML parser in Java for processing the initial XML data, but students may use other tools or languages if they prefer.

Important note: Students using alternative languages, systems, and tools must be able to submit all of their code using the standard submission procedure. Furthermore, the TAs must be able to compile and run all submitted code on the Stanford Unix Corn machines without any additional packages or runtime support, and with no additional effort over that required for projects using the supported languages and tools.

Part 1: AuctionBase Schema and Data
You will be given a large volume of data downloaded from the eBay web site and stored in XML files. You will examine the data and design a good (we hope) relational schema for it. You will then write a program or scripts to transform the data from its XML form into SQLite's load file format, conforming to your relational schema. You will create your schema in your own SQLite database and load the data.
Part 2: AuctionBase Data Integrity
You will incorporate data integrity checking into your AuctionBase database using SQLite constraints and triggers. You will also add a "current time" feature.
Part 3: AuctionBase Web Site
As a baseline, you will design an appropriate set of queries and updates for your AuctionBase system and create a simple web interface for them. Ambitious students may migrate the simple front-end into a user-friendly web interface that looks like a real auction site, and may exploit triggers, views, or other SQLite features for additional functionality. Use your creativity. Several projects will be selected to be demonstrated on the last day in class (for glory, not grades).

Late Policy

11:59pm on the due date is always the deadline.
Automated quizzes and exercises: Your score for each automated quiz or exercise set is the highest score achieved as of 11:59pm on the due date. No credit is given for automated quizzes or exercises submitted after the due date.
All other work: All other assignments are also submitted electronically. Assignments may be submitted multiple times, but the last submission before the late deadline is the only one that is used. Submissions after the deadline but less than 24 hours late are accepted but penalized 10%, and submissions more than 24 hours but less than 48 hours late are penalized 30%. No submissions are accepted more than 48 hours late.
Emergencies: Since emergencies do arise, each student is allowed a total of four unpenalized late days (four periods up to 24 hours each) for work other than the automated quizzes and exercises. (But note that no single assignment may be turned in more than two days late.) If you exceed this amount and are penalized for late work, we will automatically apply the four unpenalized late days to the work it benefits the most, i.e., to the work that is worth the most in calculating your final grade.

THIS LATE POLICY WILL BE STRICTLY ENFORCED.

Regrade Policy

Automated quizzes and exercises cannot be regraded (but if you find a significant technical error you'll definitely receive extra credit). If you believe you deserve a regrade in other work, please contact the course staff, explaining precisely why. Please do not ask for a regrade on a challenge problem until you have studied our sample solution. Include your name, @stanford username, and an exact pointer to the work in question. The TA responsible for grading the work will contact you.

Please note that sometimes during regrading we discover a grading error or misjudgement in which too many points were awarded. In this case, your score will be lowered. In other words, a regrade request may result in a higher score, a lower score, or no change.

Honor Code

Under the Honor Code at Stanford, each of you is expected to solve the automated quizzes and exercises independently, and to submit your own original work for challenge problems, database queries, and the programming project. On many occasions when working on assignments (but never exams!) it is useful to ask others -- the instructor, the TAs, or other students -- for hints, or to talk generally about aspects of the assignment. Such activity is both acceptable and encouraged, but you must indicate on all submitted work any assistance (human or otherwise) that you received. Any assistance received that is not given proper citation will be considered a violation of the Honor Code. In any event, you are responsible for understanding, performing or writing up, and being able to explain on your own, all quiz answers and other work that you submit. The course staff will pursue aggressively all suspected cases of Honor Code violations, and they will be handled through official University channels.

If you have any questions about this policy or about the degree to which we will pursue Honor Code violations, please discuss your concerns with the course staff immediately.