Hello 👋👋! This is the "about" page for MGT858 "Database Systems" at the Yale School of Management in Spring 2024. This is the third time I've taught this course. Thanks for joining!

MGT858 introduces students to the database systems used in modern technology companies. You will emerge from this course an SQL ninja, well-prepared for tech sector roles including data scientist, product manager, and technical program manager. The first half of the course covers relational databases (after a brief discussion of content-addressable storage) and the latter half covers other kinds of systems including key-value stores, document databases (noSQL), distributed databases, graph databases, and vector databases. We will discuss the business case for each of these database such that you ought to emerge from the course able to think of a software product or service and describe what kind of database systems ought to be used to build that product. You will not learn how to build such systems, though you will learn a little bit about their internals. You will definitely learn how these systems can be used to answer to the kinds of analytics questions you will encounter as a manager. The coursework includes numerous quizzes, homework assignments, and a final examination, all of which require writing code and must be submitted using the git version control system. (Experience with git is not required; however, some experience programming is necessary. The particular language does not matter.) Students can expect to spend perhaps three to four hours per week on homework.

You can be admitted to the class by bidding in the Yale SOM bidding system. I asked for a big 'ole room this year, so you should get in for "free". Anybody at Yale can take the class. But, you have to figure out what is required by your program. For example, Yale College students require some kind of form.

Because our class info and policies are tracked using git, you can see if and how they change over the course of the semester. Further, you have the ability to make changes yourself---please send me a pull request. Similarly, if you feel these documents are lacking, please open an "issue" on GitHub for this repo and we'll address the issue. Finally, because our git repo is hosted on GitHub, you can easily receive push notifications of changes to this repo. (Woot!)

Your first stop for help should be the classroom discussion on GitHub. Please only email if you need private help. Please, check with the TAs first. (I care greatly about you, but my inbox is often a hot mess and I'm not always the best at responding to email. 😞)

The office hours for each of the teaching staff is shown below. Though, if you are reading this before the start of the semester, it is possible some staff are missing because they did not yet determine their schedule.

Feel free to send an email if these times do not work for you.

You will need to write code a lot of code this semester, mostly domain-specific query languages like PostgreSQL-flavored SQL. Most of the homework assignments and the final exam will require using the git version control system for submission. You'll sign up for GitHub, join the yale-mgt-858-spring-2024 organization on GitHub and submit your assignments (in part) by pushing up code to GitHub. Usually I'll give you some starter code like empty .sql files with examples queries. When you complete your assignments, you'll make git commits and push those commits up to GitHub so that I can download your answers and test them. Obviously this requires knowing a little bit of git. I'll point you to some git tutorials, but I'm not going to teach git in class. You are forewarned 😉.

You can complete the homework assignments on any kind of computer: Windows, Mac, Linux, whatever. You'll almost certainly need to install some software in order to do so. Obviously you'll need git. You might also need _ clients_ for different database systems. For example, you will almost certain need a PostgreSQL client or a MongoDB client. The $XYZ client is a piece of software that connects to the $XYZ server in order to send the server queries (or other commands) and interpret the responses. (Here, $XYZ is a stand-in for PostgreSQL, Redis, MongoDB, Clickhouse, Elasticsearch, etc.) Now, it could be that you don't want to set up the $XYZ client software on your computer, which would be totally reasonable: it can be a pain in the rear-end. For this reason, you might want to use GitHub Codespaces for your assignments. You can find a lot of tutorials for doing that online.

I am receiving a few emails from students about how best to prepare for the class. There's a few ways to think about doing so. One is to prepare narrowly for the content you're going to encounter. For SQL, in my person opinion, the best way to do so is going through the exercises at SQLBolt. I think that website is fantastic and it's part of the assigned reading in class. For the other database systems we'll use, the answer is less clear. You can read about ElasticSearch (the representative full-text search database we might use); ClickHouse or DuckDB (the representative columnar stores we might use); Redis (the representative caching database we might use); MongoDB (the representative noSQL database we'll likely use); and Neo4j (the representative graph database we'll use).

A second way to prepare is to level-up your computer skills in general, particularly your *nix skills. A good resource for doing so is MIT's missing semester course.

Finally, you might think about getting your personal computer set up for writing code. That usually starts with choosing a package manager (homebrew on Mac OS and Chocolatey on Windows) and choosing a good code editor. I use VSCode and NeoVim.

Also 😜, you'll want to be familiar with git. You can find many tutorials online. Your use of git this semester will be super simple because you won't generally have collaborators. If you like, you can use a git GUI. I use GitHub Desktop often, mostly for line-by-line staging of changes.

There is no textbook for the course. Readings will be posted on the class website.

This class will be an inclusive environment. If you see behavior that you feel is discriminatory, particularly from the professor or teaching staff, we encourage you to avail yourself of one or more of the following resources.

• Preventing and Responding to Sexual Misconduct at Yale Booklet
• Yale Sexual Harassment and Assault Response & Education (SHARE)
• Diane Temple in SOM Human Resources and Sheri Scully in the SOM AASL.
• Yale Title IX Coordinators (SOM’s Title IX Coordinator is Rebecca Udler)
• Yale's Live Safe App
• Office for Equal Opportunity Programs

We will usually have two homework assignments per week and an on-paper quiz at the start of every class. That's a lot of work! All of the homework assignments will be submitted to GitHub using git.

You should totally be using AI assistants such at GitHub Copilot (best for writing code) or ChatGPT/Mistral/Claude/Gemini/Llama. I ask that you not copy from other humans in "real-time". You are welcome to use the advice of other humans through media such as StackOverflow or our class discussion/qa and also to work on your homework with your colleagues. Given such parameters, what do I think would constitute a violation of the honor code? I can think of a few things: 1) blatant copying, particularly in serial; 2) wide spread dissemination of answers to quizzes or homework assignments; 3) copying from a human in real-time; 4) use of unauthorized resources during quizzes or exams; 5) efforts to thwart faithful grading, such as fibbing about time of submission. What's the bottom line? We're all adults. I want you to learn, I want you to have fun, and I want to faithfully report a grade that reflects your "real world" subject matter competence at the end of the term. Don't do things that mess that up.

Most classes begin with an on-paper, ten-minute multiple choice quiz. Alas, these are a crude and imperfect assessment of the degree to which you understand the pre-class reading material. The following are my reasons for giving pre-class quizzes:

1. It makes you think about the upcoming class!
2. Empirical evidence shows that the testing effect increases your understanding and retention of material.
3. They give you a low-stakes, super-small reward for coming to class prepared.
4. They ensure that your classmates are better prepared for class and therefore more fun to speak with.
5. You can drop your lowest two scores, so they are not stressful.

A few notes about the quizzes...

• The quiz questions are not uniformly drawn from material in the pre-class reading. Some pre-class reading will be over-represented and some under-represented either, usually because of heterogeneous importance of the pre-class reading material.
• We want the quiz questions to test your conceptual understanding. We do not want to ask "gotcha" questions or simple recall questions. However, we will fall short in those aspirations often.
• Some quiz questions will have obvious answers. We do this in part to remind you of important concepts.
• If you feel like a quiz questions is unfair or that the answer is wrong, please tell Kyle. It is easy for Kyle to "fix" a bad question.
• Questions that have multiple correct answers are indicated as such.

Grading will follow the official Yale SOM grading policy, which is as follows.

In my opinion, it's a quite generous curve (though the ethics of grading on a curve are dubious). I will compute the grades for non-SOM students by overlaying their school's grading scale on top of the SOM curve. There's not a 1-to-1 correspondence. Your letter grade will be based on your sum grade and your sum grade will be determined roughly as follows (I reserve the right to make changes. The most common change I make is dropping assignments or quizzes.)

The ethics of counting participation in the grade are also a bit dubious 🙁. But, I'm including it for consequentialist reasons: your participation makes the class more fun for all of us and most importantly for me 🤣.