I taught many courses, but only these are in machine-readable form. All but CS3743 and CS4313 were online.

For programming projects, see Projects.

Fall03, Spr01, Spr98.
• The latest two iterations of this course had mostly standard content, course materials, and method of teaching it. The materials were well-prepared, but not the way I would have taught the course on my own.

Fall01, Fall98, Spr97.
• The most recent iteration had required standard content and course materials, which were well-prepared but not the way I would have taught the course by myself.

Spr05.
• I only got to teach this course once. The students had already seen Java through data structures, and now we were teaching them C. My idea was to do C well, but to give them an overview of the differences and similarities between C, Java, and C++. I planned to improve the notes at the next iteration.
• I loved my material about self-reproducing programs (see Recitation 7 or Self Replicating C Programs).

Fall04, Spr04, Fall03, Spr03, Fall02, Spr02, Fall01, Spr01, Fall00, Spr00, Fall99, Spr99, Fall98.
• The first time I taught this course (Fall 1998), it was particularly challenging because I had never studied the material. (I've never been much of a hardware person.) That first time I leaned heavily on the materials that Kay Robbins had prepared the semester before. I continued to use some of her materials the whole time I taught the course.
• As time went on, I eased up on the difficulty of the recitation material because of all the copying that occurred with harder recitations. (See MIPS Single-cycle Simulator.)
• I liked teaching the course and admired the textbook. The last time I taught this course, they came out with the 3rd edition (with my name in the Preface!).

Fall99, Spr99.
• For many years, UTSA had 1-hour language courses. Eventually these were replaced by a 3-hour course covering 3 languages, initially Ada/C/Lisp. By the time I taught the course, the three languages had become Java/C++/Lisp.
• The first time I taught the course, I used a Lisp project that asked them to write an interpreter for "pure" Lisp. The resulting interpreter included functions with parameters, so it wasn't a completely toy system. This assignment was first conceived by William F. Dowling at Drexel University. I used the assignment myself at that time, and we worked on it together, but the details were mostly his. The assignment and a solution (in C) are at: Lisp Interpreter. My solution seems to have some problem with "defun". Since it once worked, there must be some simple problem.

Fall12, Spr12, Spr04 (9 weeks).
Normalize / denormalize (interesting and related examples that came up).
• I taught the first 9 weeks of this course because the regular instructor took a leave to take care of a new child. Initially it was to be 6 weeks, but that stretched out to 9 weeks, complicating the course organization.
• In Spring 2012, because of a shortage of instructors, UTSA asked me to teach the algorithms course, which I did twice, long after I had retired.

Fall14, Spr14, Fall13, Spr13, Spr05, Fall04, Spr04, Fall03, Spr03, Fall02, Spr02, Fall01, Spr01, Fall00, Spr00.
• I taught programming languages and compiler courses starting in 1977. This was not my field, but I got started and continued with them because of a lack of instructors.
• Before the Spring of 2000, UTSA had a required course "Topics in Programming Languages", CS 3323. For the Spring 2000 course, most of my students had already taken CS 3323, so I decided to make it more challenging by covering mainly compiler topics. I liked these topics more anyway, and thought they were more worthwhile. (See All Compilers.)
• In Fall 2002, we added a required 1-hour recitation. I restructured the homework so that there was something each week. I also came to realize that if the recitations were too hard, some students would just copy.
• During the lifetime of this course, a high priority for me was to teach syntax-directed translation. Each time I taught the course I made this part easier and added other material. The final iteration (Spr05) still has a legitimate compiler, but it is feasible to complete in just 3 recitations. The language is structured so that are tokens are just single characters, so that a scanner is trivial to write. (See Simple Compiler.)
• Starting with Spring 2004, I added to the course the Ruby language as an interesting object-oriented example. This was before Ruby became so popular.
• Starting with Spring 2013, UTSA had lost 2 of its programming language instructors, and I got to teach this course 4 more times, long after I had retired.
• In Spring 2014 I switched from Ruby to Python, mainly because Python had become much more popular than Ruby. In Fall 2014 I had the class use only Python for all their programming.

Fall93.
• I no longer have a machine-readable copy.
• The main interesting feature of this course is a programming project to implement a relational database management system based on the relational algebra. (See Database Management System.)

Spr97.
• I no longer have a machine-readable copy.
• The main interesting feature of this course is a programming project to implement a recognizer of arbitrary regular expressions. (See Regular Expression Recognizer.)

Spr07 (5 classes), Spr03, Spr02.
• This is one of my areas of research. The course is complicated by the need for mathematics background that few of our students have. In the end I partially completed a textbook for this course (not published): The Laws of Cryptography.