Survival Guide
1 Why a survival guide?
CS 115 is designed to be well within the reach of all university students. There is no material in CS 115 that is beyond your abilities, and there is no reason that you should not succeed in it. However, the nature of computer science can make any introductory course a challenge to those who do not develop the right work habits.
This survival guide is a collection of guidelines, tips, suggestions, and background information to help you make the best of your opportunities. It is based on the experience of first-year students facing their first university course in computer science.
2 Course philosophy
We designed CS 115 for students in diverse programs, keeping in mind the large range of possible ways in which computer science might be integrated into future studies. By choosing to focus on fundamental concepts and study them in depth, we are giving students transferable skills on which to build (rather then providing a shallow overview of the breadth of the field of computer science).
The material in CS 115 lends itself very nicely to being built up in successive layers, where a new layer is added only when the previous one is completely explained and understood. One of our goals is to progress through the material in small steps rather than huge leaps.
When you use a Web browser, e-mailer, or other modern computer application, there’s a lot going on behind the scenes. One of our goals is to have as little “magic” as possible. You might hear us saying “Know your tools.” That means understanding not only what the tools do, but how they do it. In many cases, we will show how to implement built-in features of Scheme, rather than just describe them.
Each component of the course has a role to play: lectures, the textbook, labs, and assignments teach you the material and give you a chance to practice; assignments and the midterm provide diagnostics, giving you feedback on whether or not you are on the right track; and assignments, the midterm, and the final are used for assessment (computing your mark in the course).
Lectures illustrate concepts and techniques; labs and assignments give you practice in those concepts and techniques, strengthening your existing skills and teaching you new ones. Most of the learning in CS 115 takes place through your working through problems on your own. In many cases the final answer isn’t particularly relevant; it’s going through the process of arriving at the final answer that’s important. Doing the labs and assignments is the key to doing well in the course. Skipping labs and assignments would be like trying to learn to play a musical instrument without practicing, attempting to master a language without speaking it, or claiming to be an outstanding artist or athlete based solely on theoretical understanding of the underlying principles of the field.
In previous offerings of first-year courses, we have discovered that attendance in labs and handing in assignments were a better predictor of final grades than were high school marks. Some students make the mistake of thinking that these components aren’t important because they only account for 20% of the final grade, or that they can skip one assignment because it’s worth so little. But an assignment question worth less than 1% of the final grade or a lab question worth
—-PAGE—-
no marks could provide important insight into a concept tested by a final exam question weighted more heavily.
If labs and assignments are so important, why aren’t there marks for labs and why aren’t as- signments weighted as a greater percentage of the final grade? Your final grade is supposed to be an indication of how well you have mastered course material. That’s assessment, which occurs after learning. Lectures and labs are part of the learning process, so you don’t receive marks for attendance. The assignments really shouldn’t contribute directly to your final grade at all. But if we took them out of the final grade computation, you’d be tempted not to do them, and put yourself in serious danger of failing the final exam. So we make them worth a little bit to encourage you to do what you should be doing for your own good anyway.
3 Lectures
We’ve tried to create a lightweight implementation of the basic course outline. Instead of having a fixed set of slides for each lecture, we’ve grouped slides into thematic units we call lecture modules, with only a rough idea of how long we will take to cover each one. That way, if you ask a lot of questions, we don’t have to rush over subsequent material to catch up with a ticking clock.
The instructors work from the same set of lecture notes, though each will say different things, and possibly write different examples on the board (or overhead projector, or data projector). If you have to miss your lecture, it is better to attend another lecture than not attend any that day. However you should be aware that as the sections may drift a little out of sync, you may still end up missing some material. In addition, you should be sure that you are not taking a seat from someone who is legitimately enrolled in that section.
For possibly the first time in your life, no one is forcing you to go to lectures. There is no mark for class participation, and attendance will be taken rarely, if ever. You are free to not attend. This honour system is built on the assumption that this freedom will be exercised responsibly. Other aspects of this assumption include the assumption that students will complete the required work on time, as deadlines are firm and no “bonus assignments” are handed out to bolster marks.
You may be tempted to skip lecture because these handouts include the text of all lecture slides; surely you can do just as well reading them? If we thought that, we wouldn’t be lecturing. The slides are like basic musical chords over which we improvise a melody; the chords alone do not make up the whole song. Not everything is written down on the slides. It’s not that we’ve deliberately left things out; it’s just that we don’t design the slides to be the sole source of learning, but rather as an aid in lecturing.
The handouts contain the text of the slides so that you don’t need to scribble madly while we display them. That frees up your time so that you can take proper notes on what we say in class. We’ve seen students write down only what we write on the board, or not write anything down at all, figuring that the slides do all their work for them. That’s a mistake; you can’t possibly remember all of what is said in all of the lectures you will attend during a term. Taking notes helps fix information in your long-term memory, keeps you active, and allows you to preserve insights that you gain during the course of a lecture.
Going to lectures reserves three hours of dedicated time each week during which you are think- ing about course material, with the aid of someone who understands it thoroughly. It’s a rare stu-
—-PAGE—-
dent who is disciplined enough to skip lectures and adequately substitute for that experience. Of course, not every minute of lectures is “quality time”, but you won’t know what you’re missing if you don’t attend. Take something discreet to do (like your math homework or a good novel) during the stretches where there’s an extra example of something you already understand, and make sure you keep one eye and ear on what’s going on. You’re sharing space, so don’t do anything that might annoy or distract your classmates: don’t let your cell phone ring, don’t rustle food wrappers loudly, and don’t play video games on your laptop.
4 The textbook
CS 115 wouldn’t exist if it weren’t for our textbook, “How to Design Programs”. CS is not a field blessed with an abundance of good textbooks, due to its very short history and the ever-changing nature of the curriculum. This is one of the few books we can wholeheartedly recommend. The lectures are designed on the assumption that you’ll do the assigned readings – before lecture if pos- sible, but certainly before you attempt to do the corresponding assignment. This is not a textbook to be used for reference, to be looked at only when you run into trouble on an assignment. That will only increase the amount of time you’ll need to get your work done. (We’ll say more about doing assignments below.)
On occasion, we will use the same examples in lectures as in the textbook, illustrating different aspects. More often, we will use different examples, and at times, we will “parallel” the book by covering the same concepts using different language and a different approach. Sometimes, we will go more in depth than the textbook (it assumes eighth-grade knowledge of mathematics, and we can expect better than that of you) and sometimes we will skim a topic, leaving the details to the readings. But the textbook always informs and directs our efforts.
The textbook is available on-line, but we recommend that you buy a physical copy. Paper has more psychological weight; you will be more inclined to take your readings seriously, and get more out of them, than if you view them on a screen. It’s quite unusual for the publisher to agree to such an arrangement; neither they nor the authors are really making any money from it. Since the book is so integral to the course, we’re not going to change books; used copies will be available for you to purchase if you act fast, and you in turn can sell your copy to next year’s students.
On your schedule is a weekly lab, a time when you can work on exercises under the guidance of a tutor or teaching assistant. Each lab meeting will be a mixture of instruction and an opportunity to practice. You can view the exercises as warm-ups for assignments, on which you can work with others, ask for feedback on partially-completed work, and otherwise work out the bugs without worrying about being assessed.
If you need more time to complete the lab exercises, there are free times when the lab is open. The free times for a lab room are posted on the door of the lab. Course personnel will be happy to answer your questions during office hours as well. Assignment and exam questions often build on lab exercises, so it is well worth the time to make sure you complete and understand them all.
—-PAGE—-
The only exception to this is the optional, open-ended questions at the end of each lab. These are provided as a opportunity for enrichment.
If you are extremely fast at completing the lab exercises (including the optional, open-ended questions), you can use the rest of the time to start on your homework assignments. It is a good way to get into the habit of setting aside regular blocks of time to work, and a good opportunity to be able to have assistance from course personnel while you are working at a computer (at office hours, you don’t have the luxury of being able to ask someone to look over your shoulder while you are sitting in front of a computer screen).
Details on how to copy your files to/from the Macs, how to electronically submit your work, and some details of the program style that we expect can be found in the submission and style guide, also included here in the course handouts.
In past offerings of the course, students who have not completed the labs have found the as- signments difficult. This is not surprising, as the assignments have been created based on the expectation that previous labs were completed. Don’t make the course more difficult for you by expecting to be able to leap successfully from lectures to assignments without the intermediate step of lab work.
6 Assignments
It is on assignments that poor work habits may be most obvious, since programming, approached incorrectly, can quickly become a timesink. It’s possible to spend far too much time on CS as- signments, because it’s easy to believe that you’re really close to a solution and the next change you make to your program will have it running perfectly. Hours go by before you realize it. Don’t waste your time grinding away at a question; if you feel you’re not making progress, go on to another question, or even switch to working on another course, and come back later. We’ve had students say, “I can’t find an example in the textbook that looks like this assignment question”. We try not to ask assignment questions that look like examples, because we want you to be able to write programs from scratch, not just modify examples that you don’t fully understand.
Some students try to do CS assignments by looking at the assignment for the first time while sitting in front of their computer. They read the first question and immediately start to write a program. We’d like to suggest a more productive approach. Read the assignment away from your computer, and think about how you might solve the questions. The lectures and the textbook teach something called the design recipe. This is a process to go through in writing a Scheme function or set of functions to solve a problem. The authors of the textbook came up with the design recipe by observing students in the lab and noticing when and how they ran into difficulties, and what worked to help overcome their difficulties. We encourage you to use the design recipe by assigning marks to its various components (since they help us to see that you understand what you are doing). You can, of course, get your program running and then put in all the components worth marks afterwards. But following the process properly will save you time and reduce frustration. In CS 115, the programs you have to write are usually quite small; most of your time should be spent on thought.
The design recipe may include templates, which are function skeletons derived from data defi- nitions. Students report that the assignments took half as much time when they used the templates
—-PAGE—-
as we suggest. Finally, we usually specify an interface, which is nothing more than the function headers (name and parameter list) of each main function you are to write, spelled correctly. This prevents you from losing marks when our automarking software can’t find your answer because you spelled the name of your function incorrectly. Use this, plus the appropriate template, as a starting point when you finally start to write programs.
Public tests are available for all assignments and most lab questions. To request a public test, submit your work as instructed in the Style Guide, and then follow the instructions on the Public Tests page. You will receive feedback indicating whether your work passed certain trivial checks, such as having the right number of parameters and correctly-spelled names. Public tests are not a replacement for your own testing, nor an indication of the more thorough testing that we will use for the marking of assignments. Thus, passing public tests is not a guarantee that your work is flawless. Failing public tests, however, may mean that you have not submitted your work correctly or that you might be otherwise in danger of losing all marks for what might be a trivial error.
We have scheduled assignments (roughly one per week) so that the work is spread out evenly across the term. You are welcome to complete the assignments either on your own computer or at free time in the Mac labs. This allows you to start each assignment shortly after finishing the previous assignment, putting in a solid but not excessive amount of time. You then have time to set it aside and come back to it later. This allows you to discover difficulties and get help, and you will find that when you return to questions fresh, you will see ways to solve problems that you couldn’t see before. Although this advice holds for any course, it is particularly important for computer science.
You should try to complete the whole assignment, but if it turns out that you don’t have time to do it all, submit what you have, even if it is incomplete. You will still get written feedback from the markers, and you may get some marks as well. When you skip an assignment, you are likely to find the next one more difficult. Do your best to avoid skipping assignments, and make sure you have a sense of how to do each question, even if you don’t submit each one. Finally, when you get back your assignments, don’t just glance at the mark and toss it aside. Learn from your mistakes by reading the comments in conjunction with model solutions and post mortems.
Model solutions will be posted outside the Tutorial Centre (MC 4065) immediately after each assignment is due. In some cases several alternate solutions will be presented. Study these to make sure you understand the material; this will pay off on later assignments and on exams. This is true even for students who receive perfect marks, as even a solution that has no obvious flaw might be able to be improved in elegance or other attributes. If nothing else, knowing how to find a shorter solution will be of great benefit on exams.
Post mortems will be posted to the Web site after each assignment and exam has been marked. They contain listings of common errors found by the markers as well as tips on how to improve future work. Reading post mortems is a good way to ensure that you don’t repeatedly lose marks for the same mistake.
7 The midterm exam
The midterm exam has a diagnostic purpose: they give you feedback on what you thought you understood (but didn’t) and what you thought you didn’t understand (but did). Having a chance to
—-PAGE—-
take an exam before the final also gives you an indication of what that experience might be like. As with assignments, if we don’t make them worth some percentage of the final grade, students tend not to take them seriously, and so their value is lost. So the midterm account for a modest fraction of the final grade.
Midterm questions will resemble assignment questions, except that we can’t ask you questions that will take too much time, either to discover the answer or to write it down. You also won’t have a computer to use during the midterm (or on the final exam), meaning we can’t ask you questions for which you need to go through a process of discovery and refinement (as we can on assignments). We won’t ask you to write down memorized definitions. Obviously, you need to know the syntax of Scheme and the meanings of terms and phrases that we use. But we will test that understanding by asking questions where you need to use this material, not just repeat it from memory.
The best way to do well on the midterm is to learn the material as it is presented in class and exercised in labs and on assignments. It helps to plan for exam study as part of your weekly schedule. Start preparing review sheets as you learn material rather than waiting until right before an exam. During the midterm exam period, there is a temptation to neglect all courses except the one with the next exam; if you do this, you may find yourself behind and unable to catch up. Your studying should be active. Find questions in the textbook not already assigned and write out complete answers. Course personnel will be happy to answer any questions you encounter along the way.
We recommend not using midterms from past years as study guides. Each term, we create midterms from scratch, and the course does change from year to year. Past midterms may give you a misleading sense of what this term’s tests will be like. As with assignments, we won’t ask you to do things that are just minor tweaks on lecture examples; we try to design questions that test deeper understanding. We also recommend not cramming at the last minute. This type of study results in shallow, easily-forgotten understanding of a sort inappropriate to CS 115. Cramming works best when you have to repeat definitions and do things nearly identical to what you’ve already done, and we’re not going to ask you exam questions like that.
1A midterm marks are a shock to some students, because they’re the first confirmation that grades tend to be lower in unive