6
$\begingroup$

I am going to teach a course in object-oriented programming in Java, and am looking for materials - textbooks and slides. It is a course for 2nd-year undergrads. In the 1st year, they already learned an introduction to programming course, also with Java. So, books that just teach Java will not be good textbooks. What are good textbooks for such a course? Additionally, are there existing courses where I can get slides and other course materials, so that I do not have to prepare everything from scratch?

$\endgroup$
  • 3
    $\begingroup$ Can you say a bit more about what the students have already done and what they can be expected to know? Also, are you pretty free to design the course or does the existing design impose constraints on you? Just edit your question to add more if possible. Also, if you are new to this site welcome. We will try to be as helpful as we can. $\endgroup$ – Buffy Aug 13 '17 at 12:41
4
$\begingroup$

I'm going to jump the gun a bit here and make a suggestion with less than complete information. Assuming that the students have seen a fair amount of Java and its libraries and also assuming that the book they used isn't terrible on OO concepts then I'd suggest that you run a project course. There are a fair number of possibilities for projects

One is to use your own doctoral research interest as stated in your profile and have students build a software system to solve fair cutting problems.

Another possibility is to use an old APCS Case study that was quite well developed and used for a few years in the APCS course in the US: Grid World. The study comes with a starting code base and a student manual.

You might also join the Greenroom and look at the various projects available there. It is also a community of educators who might be able to offer assistance. The Greenroom resources use the greenfoot IDE which is simple but complete, but the advantage in your situation is the various projects that you might be able to develop into a course.

The BlueJ IDE is a bit more sophisticated and the community there also publishes a number of projects via the Blueroom. Many of them might be considered as the basis of a project course.

Of course, another consideration is how well you know Object-Oriented programming yourself, as you don't say. If you are less than an expert, you might consider for yourself to look at something like Polymorphism: As It Is Played by Joseph Bergin. I don't think it would be suitable for a course text, but it might give you some background in OO thinking if you haven't already acquired it.

I also note that in a project course you have the opportunity to introduce students to some of the additional personal practices they will need as they advance, such as teamwork and, perhaps, pair programming and test first development. Moreover, to ease your own workload, a project course could be very flexible, with students starting work on early requirements while you develop later ones. I'm thinking of something like an agile methodology here, of course, with yourself as the students' "customer".


Of course, if you want them to learn OO style and not just build code, then you need to watch over them. There are a few ways you can do this. I'm assuming that your class is reasonably sized (say 20 or less). I'm not sure how to make it work for 80. You can use the class time as design sessions for coding. You can have students work in groups for a while and show you what they come up with. Initially you can suggest designs for sections of the project. The idea is to "nudge" teams in the desired direction, rather than to lecture on the ideal.

I used to teach the Compiler Course in which a goal was also to improve OO thinking. I would review the student's code every two weeks. Student pairs (or teams) would submit printouts in a folder every two weeks. They needed to use a highlight marker to mark the changes from the previous report. It is relatively quick to go through 10 or so folders if marked up. You don't need to note every detail in these, nor do you need to grade them, but just give any needed feedback. For many groups as check mark is enough. But for some, you know you need to provide more guidance.

If teams are larger (say around 5-6) I would do two things. One is that I'd appoint a "captain" who was responsible for keeping me informed (not for team management). I could discuss the periodic report with him/her. Depending on the number of students, you could have a weekly "captain's meeting" as part of class or office hours, perhaps. I also named teams with names they though clever to give them a sense of belonging. "Lions" and "Tigers" might work, or "Earth" and "Sky". I made up the names to avoid possible improper "creativity."

$\endgroup$
  • $\begingroup$ A project course looks like a great idea, but I am not sure how to combine it with presenting the abstract OOP concepts. Isn't a project course encouraging students to just quickly writing code that works, rather than to think about OOP practices that only pay out in the long run? $\endgroup$ – Erel Segal-Halevi Aug 13 '17 at 14:10
  • $\begingroup$ Great ideas for a project course, I could use some of this. It should be possible to introduce OO topics along with appropriate parts of the projects. Also, a bit of refactoring later is good experience too. Reviewing the code is good, I have seen my students begin to write OO by passing member variable values around as arguments, missing the entire point. $\endgroup$ – user737 Aug 16 '17 at 11:55
1
$\begingroup$

I have compiled a set of notes for second Java course that is heavy on OOP. Since the department is switching to Python soon, perhaps these will be useful for somebody else.

Lecture notes

Course itself

$\endgroup$
  • $\begingroup$ Thanks a lot for sharing! I am sure it will help me a lot next time I teach this course. $\endgroup$ – Erel Segal-Halevi Jan 13 '18 at 16:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.