Housekeeping

  • Legal notice: the instructor lacks Pennsylvania ACT 153 child-protection clearance, so nobody under 18 may be in the classroom (0:33).
  • Course infrastructure (2:50):
    • Course web page — announcements, projects, homeworks.
    • Piazza — all discussion and questions.
    • Gradescope — homework/project submission and autograding.
    • Canvas — mid-semester and final grades.
    • Room cap is 140 ("before the Fire Marshall shows up"); 6 seats open at lecture time.
  • Non-CMU viewers can access projects via a Gradescope code, but must not publish solutions on GitHub — a "Wall of Shame" is threatened for PRs containing solutions (3:30).
  • Question policy: interrupt during lecture; the instructor refuses to answer lecture questions after class (5:09).

Agenda for this lecture

  1. Background — what database systems are (4:01).
  2. The relational model and relational algebra — the bulk of the lecture, and the bedrock for most systems covered in the course (4:01).
  3. Alternative data models — document/JSON and vector databases, treated as ancillary (4:23).
  4. Open Q&A at the end (4:56).

Framing

  • The course's stated bias: SQL and the relational data model are the focus; core implementation ideas still transfer to other data models (4:23).
  • Next lecture is "Modern SQL" — basic SELECT * is assumed knowledge; the lecture targets advanced features and where the SQL standard breaks down (1:12:18).