AP CS Ad Hoc Committee
Notes from the Meeting on April 27-29, 1996
During April 27-29, 1996, the AP CS Ad Hoc Committee held its last
formally-scheduled meeting. From the standpoint of the participants, the
efforts of this Committee have been most productive. Not only did this
work allow the AP CS Test Development Committee to refine the C++ subset
and clarify several issues of content, but the interactions also raised
some interesting possibilities and suggestions for new directions and
initiatives for SIGCSE. The following notes highlight a variety of ideas.
- As announced previously, Pascal will be used on both versions of the AP
CS examinations in 1997 and 1998. Both examinations will shift to C++ in
1999.
- The Ad Hoc Committee, together with the Test Development Committee,
refined the specification of a
subset of C++
for use on the AP CS
examinations. For the most part, the subset follows the constructs
proposed in a draft "Call for Comments", although some adjustments were
made. The "Call for Comments" had been circulated at the 1996 SIGCSE
Symposium and via the World Wide Web.
- A commentary
was written, adding details about the C++ subset and
explaining why a variety of choices were made.
- Three parenthetical notes may be of interest.
- While the Committee had many long discussions on details of a C++ subset,
there was great agreement about the final result; the Committee members
seem agreed that the C++ subset seems well constructed. While Committee
members might quibble about a few details now and then, these points do not
seem significant -- largely reflecting personal tastes.
- Unfortunately, very few people chose to respond to the "Call for Comments",
in spite of the circulation of the materials and the request for
feedback. The Ad Hoc Committee reviewed all comments very carefully, and in
many cases, the AP CS C++ subset was modified (and improved) in response to
the comments. In a few cases, the suggested changes seemed inconsistent
with the goals of AP CS or a CS1/2 course, or the suggested changes seemed
to add unnecessary additional syntax.
- The resulting AP CS C++ subset has many elements in common with the
description of "Essential C++" on pages 3-14,20 of the June 1996 SIGCSE
Bulletin.
The AP CS Topic Outline
- While the AP CS Test Development Committee expressed its interest in
receiving extensive feedback about the AP CS Topic Outline over the long
term, the Committee indicated that some preliminary work is necessary.
- Due to the very heavy workload of the AP CS Test Development Committee,
the current AP CS Course Description is badly dated -- the
Course Description does not adequately indicate the content and emphasis of
the current AP CS exams. Thus, at the April meeting, a considerable
amount of time was spent by the AP CS Test Development Committee in
dramatically overhauling the current AP CS Course Description to reflect
current content and emphasis. While some members of the Ad Hoc Committee
helped in this process (most folk who had served recently on the AP CS Test
Development Committee), this work largely was done by the current AP CS
Test Development Committee. In this work, the Test Development Committee
emphasized repeatedly that they were not changing the current course --
rather they were trying to bring the topics outline and course description
up to date with current practice.
- With various time constraints involved with ETS publishing, the AP CS Test
Development Committee wants a corrected AP CS Course Description to appear
this summer or fall (in a special mailing to the entire AP CS community, to
high schools, and to colleges) and then next spring in the new AP CS Course
Description booklet. With various production constraints, it seems that
there is a six-month or more time lag between when the Committee finishes
its work on material and when revised materials can appear in published
form.
- Once an up-to-date AP CS Course Description is available and circulated,
then the AP CS Test Development Committee can start the process of
considering appropriate changes, new directions, and other issues of course
content. The time frame for this work is not yet clear.
Teacher Training
- Work has already started on a revised Teacher's Guide for a C++ based
AP CS course. Fran Trees of the AP CS Ad Hoc Committee will be a principal
editor and writer.
- Plans also are being made for 2 sets of training sessions in the Summer,
1997. These sessions will be designed for AP consultants who give regional
workshops to CS teachers, so the leaders of various AP workshops around the
country will understand the important elements and issues regarding the
change to C++. Note that ETS already has budgeted these training sessions,
and various leaders (people who have recently served on the AP CS Test
Development Committee) are already talking about specific content and
schedules for these sessions.
The Relationship Between AP CS and SIGCSE
Several hours during the meeting considered how to promote and expand
appropriate interactions between AP CS and the more general CS community,
with particular consideration of the relationship between AP CS and SIGCSE
and/or ACM's Education Board. Altogether, this discussion produced many
interesting suggestions for further directions and explorations, although
some specific actions may require additional actions by the SIGCSE or other
groups. Some important points from this discusison follow:
- The AP CS Test Development Committee genuinely wants to receive
feedback on content, directions, and details of the AP CS Course
Description.
- There is a need for high school and college faculty to come together on a
regular basis for ongoing discussions. High school teachers need to know
about curricular directions in colleges and about college expectations for
incoming students. College teachers need to understand opportunities and
constraints for teaching in high schools. Unfortunately, this contact
rarely takes place (except at AP CS readings), and some college faculty may
have a tendency to pontificate without listening.
- Meaningful dialog between high schools and colleges about CS curricula
can take place only if the faculties of these groups interact, and SIGCSE
is a natural place for discussions to occur.
- While SIGCSE has talked about expanding its membership in many ways
and while SIGCSE has made some attempts to reach out to secondary school
teachers, it has not been terribly successful in attracting large numbers
of high school teachers. Some other groups may be organized for this
audience (e.g., ISTE and a new SIG), but SIGCSE can serve as an essential
role in bringing groups together.
The discussions within the Ad Hoc Committee identified some specific
opportunities and suggestions for SIGCSE to help foster dialog among
teachers at all levels and to expand participation by secondary school
teachers. These suggestions seemed practical and consistent with the aims
of SIGCSE as the premier organization for faculty interested in computer
science education.
These ideas have been forwarded to the SIGCSE Board, where a detailed
review is in progress. Within SIGCSE, such suggestions are particularly
timely, since SIGCSE has undertaken a general effort to reach out to a
broader audience -- both internationally and within the United States --
and to include a wider range of interests. The goal here is not to limit
SIGCSE's audience to introductory-level courses, but to expand discussion
at all levels, with AP CS as one important part. Thus, the work suggested
by the AP CS Ad Hoc Committee is meant to complement, not limit, other
initiatives. (Other SIGCSE efforts might even be directed at high-level,
undergraduate courses or even at graduate education.)
With on-going cooperation among AP CS, SIGCSE, and other groups, the Ad Hoc
Committee believes there is a good chance of expanding interactions between
college and high school faculty, reinforcing communication channels between
AP CS and the general CS Community, and increasing SIGCSE membership.
Your
SIGCSE representatives on the Committee welcome your comments.
Suggestions concerning this World Wide Web page also are welcome.
created June 6, 1996
last revised June 7, 1996
Henry M. Walker
(walker@math.grin.edu)