Report on Undergraduate Forum of Feb. 26, 2003

The CIS Undergraduate Studies Committee (UGSC) organized the annual CIS undergraduate forum on Feb. 26, '03 in DL 317. Originally scheduled to last from 6:00 pm to about 7:00 pm, the forum continued a bit beyond 8:00 pm, with most students staying on till the very end. Indeed, many of the discussions had to be cut short to ensure time for the various topics that were raised. The summary below tries to describe the main points that were made during the discussions but it defintely does not capture the enthusiasm of the participants. Thanks to all who came to the forum and made it such a success.

Students:
Daniel King, CSE; Keith Rustan, CSE senior; Matt Schwaberow, CSE junior; Dave Menninger, CSE junior; Nick Hurley, CSE senior; Michael Bernstein, ECE senior; Karl Kornel, CIS junior; Andy Stamper, Linguistics 2nd year; William Triest, pre-CSE; Jeremy Fincher, pre-CIS; Robert Sheets, CSE sophomore; Hans Rohr, pre-CSE; Nick Dimidole, CSE junior; Brad Moore, CIS junior; Dipal Bhatt, CIS senior; Emily Howe, CIS senior; Michael Campesino, CSE rank 4.

Faculty: Wayne Heym, Rick Parent, Paul Sivilotti, Neelam Soundarajan, Bruce Weide, Stu Zweben.
Advising office: Ming Liu, Cory Matyas, Peg Steele.

1. GPA requirement for admission to the major: Some of the faculty explained the history of the enrollment management scheme. A number of the students suggested various ideas for changes; for example, using grades in CIS courses instead of overall GPA to determine admission to the major; using a combination of overall GPA and grades in specific CIS courses; using an essay or personal interviews to handle cases of students who are on the borderline with respect to meeting the GPA requirement; etc.

   The advantage of using the overall GPA is that it is a uniform requirement for both pre-CSE and pre-CIS students, whereas using grades in specific (non-CIS) courses would introduce differences because these two groups of students don't take the same set of courses; for example, CSE students have to take Phys 131, 132, whereas CIS students can take Phys 111, 112 in place of 131, 132. This problem wouldn't exist if we used grades in CIS courses (since all students take the same CIS courses) but it would probably give an undue advantage to students who have extensive prior programming experience. Using interviews or essays to determine which students to admit has the advantage that it takes account of the individual student, but introduces the problem of subjectivity into the process; plus, it requires considerable faculty resources to administer (when the shortage of these resources is why we have had to institute enrollment management in the first place).

   One other point to note is that whatever scheme is used has to be approved by the university; and that is a fairly drawn out and difficult process. On a positive note, the demand for the major has been moderating somewhat, so the gpa requirement has recently been reduced from 3.2 to 3.0. UGSC will continue to monitor this and recommend changes (to the CIS faculty) as warranted by the demand. The committee will also consider the suggestions for changes in the enrollment management mechanism.

2. RESOLVE in the curriculum and in particular in the 221-sequence: As was the case in the forums of past years, there was considerable discussion of this topic. Opinions ranged from, "this whole RESOLVE approach is pointless and hurts students because it doesn't develop strong code development skills in the student" to, "I used to think RESOLVE was terrible but now that I am a senior and about to graduate, I see the point of it all and am really glad I went through it".

   Again as in past forums, the question of what programming language should be used in the 221-sequence was prominent. Many students argued that it would be better to use "standard C++" in this sequence since that is what they would probably have to use in later courses and later in their professional careers. But the idea of the 221-sequence is to instill good software engineering habits in students early in their computing career, and using a notation that doesn't focus on these issues would hurt this key goal of the sequence. Many of the more senior students in the room seemed to agree with this point but they were still very concerned with what they saw as the inadequate skills that the 221-sequence imparts to the students in developing medium and large systems, especially when the student has to develop the system from scratch.

   One suggestion that several students seem to like was the idea of introducing another course between CIS 321 and 560. This course would presumably help develop students' coding skills in a standard, widely used language (such as C++ or Java) while at the same time trying to ensure that the software engineering lessons of the 221-sequence are put into practice. The labs in this course would be much simpler than the 560-project helping the students to come up to speed with respect to developing code from scratch without having to worry about the demands that 560 places on such aspects as extensive documentation. While this is certainly an interesting idea and the course might indeed help students mature as independent software developers, there are some important practical hurdles: If such a course is to be added to the curriculum what will we remove (given that we do not want add hours to the programs)? How will we find the additional resources, including especially faculty, to teach such a course? UGSC will consider these issues in its future meetings.

3. Proposed change in the name of the department: (Faculty provided some background for this topic: The EE dept. has proposed changing its name to the Dept. of Electrical and Computer Engineering. This proposal is currently being considered by the College of Engineering. If the college approves it, the proposal will then go up to the university level. Primarily in response to this, the CIS department has proposed changing its name to the Dept. of Computer Science and Engineering. The reason for this was the concern that if the EE dept. is renamed, people outside the university might come to think of that department as "the" computer engineering department. The CIS faculty feel this would be a very undesirable development, given that the term "computer engineering" does not any longer mean just hardware as it used to at one time, but has come to represent, in the minds of many people in the field, a large segment of the field of computing; hence the proposal to rename our department the Dept. of Computer Science and Engineering, especially if the EE name change goes through.)

   The students seemed comfortable with the proposed new name; some students noted that the word "information" in the current name of the department and in the BS CIS program seems to confuse recruiters who tend to associate that word with MIS programs.

4. Changes in CIS and CSE curricula: There were no major comments about the recent changes in the CSE curriculum (such as the omission of Math 415 from, and the addition of Math 566 to the list of required courses). Students did have some comments on the recently proposed omission of Math 254 from and the addition of Math 566 to the CIS curriculum. While there was general consensus that the addition of a second discrete math course to the CIS program would be a good idea, there was also some feeling that omitting Math 254 would result in the students not really seeing the point of the Math 151-152-153 sequence since it is in 254 that the calculus concepts presented in the 151-sequence are actually put to use.

   This seems to be a difficult choice. On the one hand, a second discrete math course would not only be of intrinsic value, it would also allow CIS 680 to present more interesting algorithmic issues, and in greater depth; moreover, since CSE students are required to take Math 566, there is currently an imbalance in CIS 680 since some students have taken Math 566 and others have not; indeed, this was one of the key reasons for the proposal to remove Math 254 from, and add Math 566 to, the CIS curriculum. On the other hand, students, at least those present at the forum, seem to feel that Math 254 is also of some value. UGSC will consider these before making a proposal to the faculty.

   [After the forum, a student who could not attend the forum sent an email suggesting that Math 254 is a worthwhile course and it would be useful for students to take it. His point too was that it is in 254 that the basic ideas and techniques introduced in the Math 151-sequence are actually used; hence he had concerns about the proposed omission of 254 from the CIS program.]

5. Feedback on (CSE) capstone design courses: Rick Parent mentioned the new animation course (temporary number: CIS694F; permanent number to be announced) that is likely to be approved as a new capstone design course. It has CIS 581 (or 681) as pre-req.

   Some students suggested that it might be useful to require CIS students to also take a capstone design course. Currently, CIS students in the Software Systems option are required to take one of CIS 662, 681, 756, or 758 as their "lab elective"; CIS 758 is a CSE capstone design course but the others are not (CIS 762, the follow-up to CIS 662 is); moreover, CIS students in the Information Systems and Scientific Computing options do not have a "lab elective" requirement. Again while the idea of requiring all CIS students to take a capstone design course is appealing, we also have to consider the questions of number of hours in the curriculum, as well as resources to offer additional sections of the capstone courses.

6. Possible new courses: There was a range of different ideas suggested here but also a lack of consensus. For example, it was suggested that a course focusing on "web development" might be very useful; on the other hand, some students felt such courses are really "technology" courses and not appropriate as part of the BS CIS or BS CSE programs.

   There seemed to be more consensus about the need for new networking courses or possibly changes to existing ones. In particular (and relating to the question of capstone design courses), there was some feeling that a capstone design course on networking would be very useful and very popular. In the past, the networking faculty have talked about revising the networking curriculum. One of the changes they had proposed was to reorganize the curriculum to introduce a standalone course that students who want to get an introduction to the field can take; and a networking sequence for students who want to get into the field in greater depth. A capstone course would certainly be worth considering (subject again to the ever-present resources issue) when and if these changes are implemented.

   On a more general note, faculty pointed out that ideas for new courses or for changes in existing courses should be brought to the Curriculum Committee. The undergrad student representative on the Curriculum Committee is Dennis Dalessandro (email: dalessan@cis). Similarly, ideas for changes in both the CSE and the CIS programs should be brought to the Undergraduate Studies Committee. The CSE student representative on UGSC is Davis King (email: davisking@mail.com) and the CIS student representative is Erin Dean (email: dean@mps.ohio-state.edu) .

7. Research Opportunities: Many students are interested in doing an independent project, either involving original research or perhaps a serious system-implementation activity. Some students felt that there was no single person or place they could turn to who would help them identify such projects. Others pointed to the Europa [ http://www.cis.ohio-state.edu/europa/] group which is a good training ground for such projects. It was also suggested that UGSC could maintain a central up-to-date web site that would contain pointers to possible undergraduate research opportunities throughout the department.

   After the forum, someone noted that another site that students interested in research might find useful is:
   http://www.cis.ohio-state.edu/~weide/sce/now/projects.html

8. Additional: Several people provided, after the UG Forum, additional information or made other comments relevant to one or more of the issues raised above:
   • Two new courses, one on Network Security and another on Information Assurance have been proposed and are currently in the process of being developed. Detailed information about these courses is available at: http://www.cis.ohio-state.edu/~xuan/courses/new/.
     Another course on wireless networking (versions of which have been taught as a "seminar" course) is also under development and is expected to be offered within the next few quarters.
   • Some courses, or at least some instructors, seem to assume students in their class know material that is not part of the specified prerequisites for the course. For example, some CIS 541 instructors assume their students are familiar with differential equations when the math prereq for the course is just Math 153. While it may be alright to devote a small amount of a course to material that not all students can handle because of lack of prerequisite knowledge, it is clearly a bad idea for a large part of the course to depend on such knowledge, especially when the course in question is a required one.
   • The 459.xx classes seem a bit weak; meeting once a week and doing a handful of small programming tasks is not sufficient to get a real working knowledge of serious languages, especially if the language is quite different from the ones the student is familiar with. Perhaps these can be expanded to 2 hour classes?
   • Students interested in recent ideas and technologies such as Design Patterns that are fairly widely used in the software industry should take a look at the new Applied Software Sequence. [It should be noted that the first course in that sequence is not intended for CSE/CIS majors, and cannot be taken as tech elective courses by these students.]