An Assessment of Theory of Computation in Computer Science

Curricula

C. M. Keet
University of KwaZulu-Natal, South Africa
[email protected]

Computing curricula are regularly reassessed and updated to reflect changes in the
discipline. Currently, the ACM/IEEE curriculum ‘CS2013’ is under review, which
provides the main international guiding principles for curriculum development. We assess
this curriculum focusing on one of the core themes of computer science, being Theory of
Computation. We examine how it is implemented in computer science curricula around the
world, and the sentiment around teaching it. Two surveys were conducted, examining
curricula and syllabi of computer science degrees and an online opinion survey. Theory of
Computation is part of 84% of the consulted curricula around the world, but taught at only
27% of the South African universities, and these syllabi contain substantially more Theory
of Computation topics than the basic core in CS2013. The online survey not only confirms
this but also indicates inclusion of even more topics as essential for Theory of
Computation and shows that it is mostly solidly part of the degree programme as a core
course and mostly in the 2nd or 3rd year, despite that for more than half of the respondents,
the course causes issues in the university system.

Introduction
Computing skills are designated as scarce skills, yet it is important not only to produce
more computer science graduates, but also of a high quality so that graduates are equipped
with the capabilities to design novel IT solutions for the Southern African context.
Currently, there is no South(ern) African computing organisation for quality and
curriculum guidance and accreditation to assist with this, hence one is left with considering
international efforts and adapt it to the regional context (Marshall, 2011). The main
international organisations for Computer Science (CS), such as the US-oriented
Association for Computing Machinery (ACM) and the International Federation for
Information Processing (IFIP), conduct periodic updates to the CS curricula. A landmark
publication in 1989 by the ACM defined the CS discipline and curriculum topics (Denning
et al, 1989), which was followed by an internationally more inclusive and much more
detailed list of topics and the notion of ‘modular curriculum’ by (UNESCO-IFIP, 1994),
which was followed by the UNESCO-IFIP “ICF-2000” curriculum and “CC2001” by the
ACM (Roberts, 2002). Currently, the ACM/IEEE “CS2013” is under public review
(ACM/IEEE Joint Task Force on Computing Curricula, 2012; Sahimi, Aiken & Zalenski,
2010; Sahimi et al, 2012), which, as notable change to its predecessors, recognises the
diversification within CS beyond the simple hardware/software/IT divides, and describes
learning outcomes. A central theme is what falls under the banner of Theory of
Computation (ToC)—being, roughly, formal languages, automata, Turing machines,
computability, and complexity—that introduces the mathematical and computational
principles that are the foundations of CS, such as the foundations of programming
languages and algorithms, and the limits of computation; typical textbooks are Sipser
(1997) and Hopcroft, Mottwani & Ullman (2007). These themes are mostly taught in at
least one basic undergraduate course (module), but also can be split over two or more
courses, such as automata in a compilers course and complexity jointly with algorithms.
These topics were core in the original curriculum guidelines (Denning et al, 1989;

1
UNESCO-IFIP, 1994), but CC2001 and CS2013 have divided that into a smaller core and
a set of elective topics, which contributes to a steady flow of anecdotes about ‘dumbing
down’ of CS degrees (e.g., (Dijkstra, 1988; Dewar & Schonberg, 2008)). The ACM/IEEE
and UNESCO-IFIP curriculum guidelines are, however, not to be understood as templates
for course design, evaluation of subjects, teaching and learning principles, or criteria for
accreditation, but principally as a guideline by means of a list of required and
recommended undergraduate degree contents (van Veen, Mulder & Lemmen, 2004).
Given the observed difference in emphasis on ToC in the international curricula guidelines
over time and the pending update on the ACM/IEEE curriculum—as well as curriculum
update discussions at the author’s institution and elsewhere, e.g., (Sahimi, Aiken &
Zalenski, 2010)—it is useful to assess 1) how ToC topics are implemented in CS curricula
around the world at present, 2) whether there are any country or regional differences, and
3) what the sentiment is around teaching ToC in academia that may influence the former.
To gain insight in these three aspects, we conducted two surveys, which, to the best of our
knowledge, are the first of its kind. The first survey examines syllabi of CS degrees as
published online on the respective universities’ websites with respect to ToC. The second
survey was an online survey open to everybody, which asked for the respondents’ opinions
on ToC, the context in which it is taught, and what topics should be in a ToC course. The
salient outcomes are that ToC is included in the vast majority of the consulted CS curricula
around the world, except for South Africa, and these syllabi contain substantially more
ToC topics than the basic core of CS2013. The online survey confirms this, and, moreover,
the respondents generally include even more topics as ‘essential’ than that are given in the
syllabi. Also, most respondents did do ToC and it is being taught at most of the
universities with which the respondents are affiliated, and it is solidly part of the degree
programme in the majority of responses. The opinion survey also highlights that there are
difficulties in teaching ToC, exhibited by, among others, relatively high failure rates or
other issues in the university system.
In the remainder of the paper, we first describe the materials and methods of the surveys,
followed by the results and a discussion, and then we conclude.
Materials and Methods
The survey is divided into two components: the survey of extant curricula and syllabi and
the online opinion survey. Limitations of the set-up will be addressed in the discussion
section.
Curriculum and Syllabi Survey
General set-up. A selection of universities will be made, with a focus on all South African
universities designated as ‘traditional’ or ‘comprehensive’, and from each major region in
the world a subset of universities is pre-selected. The aggregations for the latter are: Africa
outside South Africa, Asia, Latin America, Europe (continent, west), and the discontiguous
region of Anglo-Saxon countries (Australia, Canada, New Zealand, UK, USA). The
selection of the international universities is based roughly on inclusion in the Times
Higher Education Ranking for countries unfamiliar to the authors. Of those regions and
countries, only those will be assessed that have a website and a curriculum understandable
in any of the languages that the author understands sufficiently in order to conduct the data
collection (being, in alphabetical order: Afrikaans, Dutch, English, German, Italian,
Portuguese, and Spanish).
Data collection. Parameters to record are: region, country, type of university (where
applicable), whether ToC is included in a degree programme in whole or in part or not, and

2
if so in which year of the degree programme, and analysis of content of the syllabi (if
present).
Opinion Survey
General set-up. The method chosen for this survey is that of an open online questionnaire
for a two-week time period. Both individual email invitations are to be sent to colleagues
inside and outside the University of KwaZulu-Natal, an announcement on the Description
Logics and UKZN/CSIR-Meraka Centre for Artificial Intelligence Research mailing lists,
and social media will be used as well (the author’s Facebook and Google+ accounts).
Respondents can leave their email to be contacted for further questions and feedback. The
survey software used is LimeSurvey, because it is free survey software that has the
necessary feature of conditional questions and extensive branching. Analysis of the results
will be carried out with the built-in LimeSurvey features and Microsoft Excel.
Survey Questions. The survey consists of eight main questions, with conditional questions
depending on the answer provided by the respondent. They are summarised here.
1. Should Theory of Computation [roughly: formal languages, automata, complexity]
be a course in a CS programme?
a. If yes: core or elective, undergraduate or postgraduate, when in the
programme (undergrad (1, 2, 3), honours(4), MSc, PhD)?
2. Is it taught at your university?
a. If no: was it, but cancelled? If yes: Why?
b. If yes: core or elective, when in the programme (undergrad (1, 2, 3),
honours(4), MSc, PhD), is it secure/solidly in the programme or threatened
to be cancelled, does it cause ‘problems’ such as being flagged for low pass
rates or negative student evaluations? Participation (indication): estimate of
average amount of enrolled students in each course offering over the past 3
years (<10; 10-30; 31-60; 61-100; more than 100), is the amount stable or
in decline (stable, slight decline, strong decline (>50% fewer students in
past 5 years)). First-time pass rate (<20, 20-40, 41-60, 61-80, 80-100%).
3. Did you ever teach, or are currently teaching, Theory of Computation or Formal
languages and automata only or Computability, complexity only?
4. Did you do Theory of Computation in your degree? If yes: which year of the
programme? If no: do you miss it/regret not having done so?
5. Do you do research in Theory of Computation topics? Do you use Theory of
Computation topics in research or at work?
6. Our survey of syllabi indicates some differences across universities. Please indicate
for the following topics whether you consider the topic ‘essential’, suitable for an
‘extended version’ of a ToC course, or ‘peripheral/may be skipped’. (For reasons
of brevity, the full list of 46 topics in included in the results section only.)
7. Comments the respondent wishes to make, name of the organisation where the
respondent works/studies, email (optional).

Results
In line with the two surveys, the results will be presented separately.

3
Computer Science Curriculum and Syllabi Survey Results
Countries and Universities. The full list of universities consulted is included in the
appendix; they are 17 universities from South Africa, 15 from Europe, 15 from the Anglo-
Saxon countries, 6 from Asia, 7 from Africa other than South Africa, and 8 from Latin
America. The lower amounts for Africa, Asia, and Latin America is largely due to the
language barrier and that they generally have less information online, except for the
relatively famous (and rated) ones. A total of 9 universities had no or insufficient data
online to assess ToC in the curriculum, being two in South Africa (Limpopo and Venda)
and 7 in Asia, Africa, and Latin America.
ToC in the curriculum. The inclusion of ToC in part or in whole in the CS curricula of the
South African universities that had such information online (n=15) with that of the
consulted universities in the rest of the world that had such information online (n=44) is
compared in Figure 1. More precisely: 27% of the South African universities—University
of KwaZulu-Natal, University of South Africa, University of Western Cape, University of
Witwatersrand—has ToC in the CS curriculum compared to 84% elsewhere in the world.

Figure 1. Comparison of ToC in the CS curriculum in South Africa and in other countries.
(FLAT=formal languages and automata theory)
Disaggregating the “other countries” by the identified regions and including also those
universities consulted for which no or insufficient material was available online, we see a
similar picture (see Figure 2). The African universities consulted that include ToC in the
curriculum are Alexandria, Makerere, Zimbabwe, Lagos, and Kenyatta partially, whereas
for Botswana, and Addis Ababa no information was available online. The absence of
online information was also an issue for the Latin American universities, but, given their
responses on the opinion survey (see below), it definitely will amount to a higher
percentage than is currently included in the syllabi data. This is unclear for Asia.

Figure 2. Curriculum evaluation on inclusion of ToC, disaggregated by region outside South

Africa.

4
The difference between Europe (93% inclusion of ToC) and the Anglo-Saxon countries
(80%) may be an artefact of the sample size, but it deserves further analysis with larger
sample sizes.
The level of consistency of presenting information online about curricula and/or course
listings and/or course descriptions and/or detailed syllabi varied widely, and in several
cases multiple options applied. For instance, ToC being core in one ‘track’ or ‘stream’, but
not another, or for some students scheduled in the BSc and others in the MSc programme.
There were 43 universities (of the 59 with data) that had sufficient information online
regarding timing of ToC in the degree programme. Of those, only 5 had it explicitly in the
MSc degree programme, with the rest mainly in year 2, 3, or 4. Five universities that offer
ToC have it spread over 2 or more courses, and the rest offers it in a single course
(occasionally, additional advanced courses covering advanced topics such as probabilistic
automata and other complexity classes). An initial attempt was made to categorise the
topics of the syllabi in more detail, but this was abandoned due to the high variability of
detail of such information being online.
Opinion Survey Results
Characterisation of respondents. The number of completed surveys was 77, of which 58
filled in their affiliation. The respondents are mainly employed at universities and research
institutes: 12 respondents indicated an academic affiliation in South Africa (Stellenbosch,
KwaZulu-Natal, UNISA, Pretoria, Witwatersrand) and thus the majority of respondents
were from around the world, including universities of Illinois, Rutgers [US], Waterloo
[CA], Bolzano [IT], Dresden [DE], Geneva [CH], Southampton [UK], Ben Gurion [IL],
Bahía Blanca [AR], CENATAV [CU], Pontificia [CL/BR], São Paolo, Rio de Janeiro,
Sergipe, Espirito Santo, and 8 others from [BR], Macau [CN] and Indonesia [ID], and at
least three respondents are employed in industry (Boeing, Google, and SRI International).
35 respondents indicated to have taught ToC, 37 formal languages and automata only, 25
computability and complexity, and 16 closely related courses such as algorithms, logic,
and compilers. 41% of the respondents (32) conducts research in ToC topics and 72% (56)
use it in research or other work (including 2 from the aforementioned companies). Twelve
respondents neither conduct research in ToC topics nor use it. The spread of respondents is
too broad to merit statistical analyses about whether responses vary significantly by
country, by continent, or by predominant language at the organization.
Answers relating to ToC. The anonymised raw question answers with percentages, sorted
by question (exported from LimeSurvey) are online at http://www.meteck.org/
files/tocsurvey/, and anonymised answers to each survey question in per-question-format
with answers by respondent as a spreadsheet are available upon request; the following
paragraphs summarise the results.
The basic ToC course statistics are as follows. 76 of the 77 respondents are of the opinion
ToC should be in programme, 74 (96% of the answers) have it currently in the programme,
and 82% had it in their degree programme when they were a student. Of those 14 who did
not do ToC during their studies, 86% misses not having had that opportunity and the
remaining respondent did not do ToC, does not miss it not having done so, said that it
should not be in the programme, and never uses it at work.
Considering when it should be taught in the degree programme, and taking note of the 10
comments in the comments field of the survey describing it has been divided over several
courses, there was a clear tendency for undergraduate compared to honours, MSc, or PhD
(ratio yes/no 0.69 for undergrad versus 0.45 for honours and 0.33 for MSc/PhD), and the
3rd year in particular (ratio yes/no year1 0.13, year2 1.03, and year3 1.57). This roughly

5
corresponds to when the respondents themselves did ToC and when it is taught at the
university. Given there must be several calendar years difference between being a student
and that the survey respondents are graduated since a while (see previous paragraph), there
is thus little change over the years, as can be seen in Figure 3. The 5% increase in teaching
ToC in year 3 comes mainly from the decrease in honours/year 4 and PhD.

Figure 3. Comparison between when the survey respondents did ToC in their studies and in which
year in the study programme it is taught at their university (where applicable).
ToC is a core course in the curriculum in 67 of the 74 answers (90%), and secure in the
programme for 57 out of 66 responses (86%). Only few reasons were provided for “under
threat/other”, being that it has been removed from some specialisations but not all (in part,
due to the computer science vs. information systems tensions, as elaborated on by one
respondent), or threatened due to low enrolment numbers, resulting in one case ToC being
taught only every other year. Enrolment numbers vary greatly, from less than 10 students
(6% of respondents) to classes with more than 100 students (16%), with the main 31-60
students/year (40%) and then 11-30 students (24%). 25% (15 of 60) record a slight decline
in enrolment.
Given the plentiful anecdotes, hearsay, and assertions in other ToC teaching papers about
difficulties with ToC teaching and learning, we also asked about that in the survey. The
data provided by the respondents do substantiate the existence of issues to some extent.
While 44% of the respondents answered that there are no issues and everything runs
smoothly, 32% note it causes problems in the academic system each year and another 24%
reported that management/student affairs has gotten used to the fact there are problems,
i.e., a slight majority of respondents faces issues. Several respondents provided additional
information regarding the issues, mentioning low pass rates (3), that students struggle
because they do not see the usefulness of ToC for their career (4), that it also depends on
the quality of the teacher (2), and low enrolment numbers (2). We considered three
variables present in the result set that may influence there being issues: pass rates, class
size, and content of the course. Course content was extrapolated from the answers given to
the ToC topics, where more topics denoted as ‘essential’ was assumed to result in a
heavier course load, which need not be the case, and no correlation was found. Data on
the interaction between pass rates, participation, and issues were analysed: for 45%, the
first-time pass rates remain below 60% and with 80% of the respondents, the pass rate
remains below 80%. There is no clear trend between pass rate, class size and issues, except
that pass rate 41-60% and class size 31-60 have comparatively more issues (72% and 64%
of the reported instances, respectively), and that the correlation between pass rate and
issues is 0.79; n is to small to draw any conclusions for the other combinations.
ToC topics. The final set of main questions concerned the topics that should be part of a
ToC course. 46 topics were listed and for each one, the answer [essential/extended/
peripheral/no answer] could be given. The responses were analysed in two different ways:
calculating the percentage of a response value out of the total responses given and by

6
assigning values to the responses (essential = 3, extended = 2, peripheral = 1) and ordering
topics according to the average over the given answers. The order of the topics is roughly
the same for the essential topics and varies only by a few places at the tail end; thus, there
is a consensus about which topics are important regardless the measure. The complete list
of ToC topics ordered on percent ‘essential’ is shown in Table 1.
Table 1. Ordering of the 46 ToC topics, by calculating the percentage of responses that
marked it as ‘essential’ out of the given answers.
Topics ordered on percent ‘essential’ cont’d
Regular expressions 98.55 Show problem to be decidable/un- 61.90
decidable(RE, non-RE, RE but not
Rec.)
Deterministic Finite Automata 95.71 Push-down Automata (PDA,deter- 61.54
(DFA) ministic and non-deterministic)
The Turing Machine (basics) 94.37 Converting RG NFAs 60.61
Context-free Grammars (defini- 94.20 TM acceptors 57.14
tion, ambiguity, simplification,
derivations)
Non-Deterministic Finite 88.41 Pumping lemma for CFLs 54.69
Automata (NFA, epsilon-NFA)
Equivalences & conversion RE 85.29 Equivalences & conversion PDA, 51.61
and automata CFG
Problems a computer cannot 85.29 Show problem to be P, NP, NP- 50.75
solve complete, NP-hard, co-NP
Halting problem 82.81 State minimization 49.25
Properties of Regular languages 80.30 Closure properties of CFLs 49.21
Regular grammars (RG) 80.00 Recursively Enumerable, non-RE, 46.38
RE but not recursive languages
Examples of some undecidable 78.13 Decision properties of CFLs 46.03
problems/languages
Church-Turing thesis 77.94 Non-deterministic TM 45.45
Computability and decidability 76.81 Cook’s theorem 38.71
Equivalences & conversion DFA, 73.85 Diagonalization language 34.38
NFA, epsilon-NFA
P, NP, NP-complete, NP-hard, 73.53 Multi-tape TM 30.88
co-NP
Universal Turing Machine 72.06 Rice’s theorem 30.00
Undecidability 68.57 Post correspondence problem 26.67
Pumping lemma for Regular 68.18 PSPACE, EXPTIME, ... 26.23
languages
Some well-known NP problems 68.18 Moore Machines 21.31
(e.g., TSP, SAT, Node cover)
Chomsky normal form, hierarchy 67.16 Mealy Machines 21.31
Reductions 65.15 Programming tricks for TM 20.59
(storage, tracks)
Proving undecidability 62.86 TM transducers 17.74
Polynomial time reductions 62.69 Hot/fun topics and their 15.87
complexity classes (e.g., games)

7
Topics that received most (50-60%) ‘In an extended course’ are: TM transducers; Post
correspondence problem; PSPACE, EXPTIME, ...; hot/fun topics and their complexity
classes (e.g., games); ‘programming tricks’ for TM (storage, tracks); and Cook’s theorem.
Four respondents used the comments field to add other topics, being: weighted automata
and transducers with applications in a basic course and tree automata in an extended
course, Quantum Turing Machine, Savitch Theorem, PSPACE-completeness, and parallel
complexity classes with PRAM NC P-complete.
Discussion
The responses of the opinion survey—77 being a substantial amount for an open survey—
show an overwhelming agreement about the need for a ToC course in a CS degree
programme, regardless whether they have done the course themselves, teach or have
taught it or a similar course, or conduct research in it, or do not us it at all. As such, it re-
confirms ToC’s place in the curriculum from the perspective of, mostly, academics.
Concerning topics of a ToC course, it is perceived decidedly that formal languages,
automata, Turing machines, complexity, computability and decidability themes form part
of one coherent offering, but that the detail of the sub-topics covered may vary. For
instance, including Turing machines in a basic ToC course, but transducers, storage and
tracks only in an extended or advanced ToC course, including Deterministic and Non-
Deterministic Finite Automata, but not Mealy and Moore machines, and covering Context-
Free Grammars, but not decision and closure properties. This contrasts quite markedly
with the Strawman/CS2013 outline, which is depicted in Figure 4.

Figure 4. Proposed CS2013’s ToC topics in the Strawman draft (layout edited).

However, the feasibility of imparting a real understanding of complexity classes P and NP

without also touching upon computability and Turing machines is limited. In addition, the
hours indicated in Figure 4 have to be understood as minimum hours of fact-to-face

8
lectures, which amounts to 8 lessons at a South African university (8 * 45 mins = 6 * 60
mins), or at least almost 3 weeks of a standard 16 credit semester course, which, if this
minimum is adhered to, amounts to a very superficial treatment of partial ToC topics. We
focus explicitly on CS programmes, which should have most time dedicated to ToC
compared to other computing specialisations, but even then, the voices from the field
clearly demonstrate putting a higher weight on ToC than the ACM/IEEE curriculum
developers allot to it. Why could this be so? Arguments can be heard that ToC matters
more for CS than other recently recognised specialisations within computing—e.g.,
software engineering, net-centric computing, information systems, and computational
biology—, that this diversification has to be recognised by the curriculum developers
(Rosenbloom, 2004; Sahimi et al, 2012), and that it should result in putting more or less
weight on the core topics (see (ACM/IEEE Joint Task Force on Computing Curricula,
2005) for a detailed analysis on sub-disciplines within computing and a proposed
weighting of curriculum themes). This is not reflected as such in the Strawman draft: the
different tracks within CS all have to do the core, with 100% for tier-1 and >80% of tier-2
core, and an undefined amount of the elective topics to facilitate track-development,
although no tracks have been defined in the CS2013 yet. It is noticeable only when one
compares it with the Software Engineering curriculum guidelines (ACM/IEEE Joint Task
Force on Computing Curricula, 2004), as those guidelines include only a little bit on finite
state machines, grammars, and complexity and computability in the “Data structures and
algorithms” and “Discrete structures II” themes. It may be the case that, in praxis, those
degree programmes called “computer science” indeed do contain the more fundamental
topics, such as ToC (and logic, formal methods etc.), and that other ‘tracks’ actually have
been given different names already, hence, would have been filtered out unintentionally a
priori in the data collection stage of the curriculum survey.
Concerning issues teaching ToC, on an absolute scale, that 56% faces issues with their
ToC courses is substantial. It deserves a comparative analysis to uncover what the other
half does so as to not have such issues. Comments in the survey and offline (in the form of
follow-up emails) by survey respondents suggest it may help to demonstrate better the
applicability of ToC topics in the students’ prospective career, have experienced good
teachers, and appropriate preparation in prior courses to increase the pass rates. Further, it
might be related to the quantity and depth of material covered in a ToC course with respect
to nominal course load. The data hints also to another possible explanation: even with a
80-100% pass rate and no low enrolment the ‘gotten used to the issues’ was selected
occasionally, and vv., with a 41-60% pass rate that everything runs smoothly, thereby
indicating that having issues might also be relative to a particular university culture and
expectations of students, academics, and management.
Addressing possible limitations of the survey set-up. There are several limitations of the set
up of the surveys that may affect the results. First, the non-South African universities were
selected largely based on reputation, such as the Times Higher Education Ranking, to the
extent one might ponder whether the fact that it is skewed toward the ‘good’ universities
may have an effect, and one might argue that they include ToC because they have a good
insight in designing a high quality curriculum. Or, that the sample is too focussed on
education systems in ‘the West’—which holds for the syllabus survey, but not the opinion
survey—and that such curricula are, or have to be, adapted to the local context in a yet to
be specified way. The latter ought not to entail omitting core material from a discipline’s
curriculum anyway, and, moreover, CS and development of novel and good quality
software requires an understanding of ToC topics; e.g., in order to develop a correct
isiZulu grammar checker or parser, scalable image pattern recognition algorithms to

9
monitor wildlife tracks with pictures taken in situ, or an ontology-driven user interface for
the South African Department of Science & Technology’s National Recordal System for
indigenous knowledge management. These are broad-sweeping statements and this
research does not provide, and did not aim to provide, an answer to these questions.
Second, when there were detailed syllabi, they often did not provide detail regarding the
hours spent on each topic and no credit comparison was made, which can introduce a
larger variability than what is presented in the results. A case in point is the detailed
evaluation of the curricula by academics from Imperial College London, TU Delft, ETH
Zurich, and RWTH Aachen that gave ToC a relative importance of 10%, 7%, 4.5%, and
9.3%, respectively (IDEA League, 2001). However, this does not change the substantial
difference between the presence/absence of ToC in the curricula in South Africa versus in
other countries.
Regarding the opinion survey, the distribution of the invitations was skewed toward fellow
scientists who work in similar fields, both regarding individual emails and the distribution
lists, most of whom use, or know they benefit from, ToC topics in their work, and perhaps
a bit skewed toward computability and decidability as most important topics. Also, it is an
open survey, hence, it may exhibit the tendency that those who are more passionate about
ToC will be the main proportion of respondents. However, the cross-check with the
curricula and syllabi survey did not demonstrate a clear difference, and the necessity of
including at least complexity is also reflected in the Strawman/CS2013 draft; hence, the
opinion survey does not indicate the existence of this possible bias. Moreover, the
respondents come from each continent of the world and from many different universities,
therewith providing a valuable snapshot of various aspects of ToC, in particular regarding
the basic ToC course statistics, perceived importance of topics, and data about prevalence
of issues, which is, to the best of our knowledge, the first of its kind.
Returning now to the original three questions posed in the introduction, we can answer
them as follows. Regarding the first question, on how ToC is implemented, it can be
observed that ToC topics in the actual international curricula are more in line with the
older curriculum guidelines of Denning et al (1989) and UNESCO-IFIP (1994) than the
more recent versions that put less weight on ToC topics. The timing in the curriculum
regarding when to teach ToC remains largely stable. The results show there are
country/regional differences, with the most pronounced one being that ToC is taught at
only 27% of the South African traditional and comprehensive universities versus its
inclusion in 84% of the consulted curricula elsewhere in the world. Even including the
ones with partial ToC coverage does not make up for the differences with elsewhere in the
world and any of the proposed curriculum guidelines. Other geographic or language-based
differences are not deducible from the data, or: based on the data, region does not matter
substantially regarding inclusion of ToC in the CS curriculum, except that the slight
difference between Europe and the Anglo-Saxon countries deserves further attention.
Opinion on ToC is overwhelmingly in favour of having it in the curriculum, and primarily
in the 2nd or 3rd year. Also, a large list of topics is considered to be ‘essential’ to the course,
and this list is more inclusive than the recent international curricula Strawman drafts’ core
for ToC topics. Despite noted issues with the course, the voices from the field clearly
indicate that ToC is here to stay.
Conclusions
Both the survey of the international curricula and syllabi and the opinion survey show an
overwhelming agreement that Theory of Computation should be taught and is being
taught, and a majority has it scheduled in the 2nd or 3rd year in an undergraduate computer

10
science programme. The course is mostly solidly in the programme as a core course. There
is agreement on the typical topics that are considered as essential to Theory of
Computation, covering regular and context-free languages, automata, Turing machines,
undecidability, computability and complexity, where the subtopics covered vary. This is in
line with older computing curricula guidelines, but less so with recent proposals that,
comparatively, downplay Theory of Computation topics even for pure/‘majoring in’
computer science curricula in favour of a smaller core and multiple tracks. About half of
the respondents note there are issues with the course, for various reasons, including, but
not limited to, low pass rates and low enrolment, where roughly half observe first-time
pass rates below 60%. This nevertheless does not to have an effect on the curricula thus
far.
Given that, practically, Theory of Computation is solidly in the CS degree programme, and
perhaps ought to be introduced more widely in South Africa, our future line of work
pertains to assessing reasons behind the noted issues with teaching Theory of
Computation, and what makes the course successful and running smoothly in some cases,
including aiming to obtain a better specification of its prerequisites, and therewith other
undergraduate courses.

References
ACM/IEEE Joint Task Force on Computing Curricula. (2012). Computer Science Curricula 2013
Strawman Draft (Feb. 2012). ACM/IEEE, Online: http://ai.stanford.edu/users/sahami/CS2013/.
ACM/IEEE Joint Task Force on Computing Curricula. (2005). The overview report. ACM, AIS,
IEEE-CS, September 30, 2005.
ACM/IEEE Joint Task Force on Computing Curricula. (2004). Software Engineering 2004. ACM,
IEEE-CS, August 23, 2004.
Denning, P.J., Comer, D.E., Gries, D., Mulder, M.C., Tucker, A., Turner, A.J. & Young, P.R.
(1989) Computing as a discipline. Communications of the ACM, 32(1), 9-23.
Dewar, R.B.K. & Schonberg, E. (2008). Computer Science Education: where are the software
engineers of tomorrow? Crosstalk: The Journal of Defense Software Engineering, January
2008, 28-30.
Dijkstra, E.W. (1988). On the cruelty of really teaching computing science. Unpublished
manuscript EWD 1036.
Hopcroft, J.E., Motwani, R. & Ullman, J.D. (2007). Introduction to Automata Theory, Languages,
and Computation, 3rd ed., Pearson Education.
IDEA League (2001). Report on comparison of curricula in Computer Science. ICL, TUD, ETH &
RWTH. Oct, 12, 2001. Online: http://www.automata.rwth-aachen.de/~thomas/Ideareport01.pdf.
Marshall, L. (2011). Developing a Computer Science curriculum in the South African context. In:
G. van der Veer, P. Sloep & M. van Eekelen (Eds.), Computer Science Education Research
Conference (pp9-19). Heerlen, the Netherlands, 7-8 April, 2011. New York: ACM Conference
Proceedings.
Roberts, E. (2001). Computing Curricula 2001: Computer Science Final Report, New York : IEEE
Computer Society.
Rosenbloom, P.S. (2004). A new framework for computer science and engineering. IEEE
Computer, 37(11), 23-28.
Sahimi, M., Aiken, A, & Zalenski, J. (2010). Expanding the frontiers of computer science:
designing a curriculum to reflect a diverse field. In: G. Lewandowski, S. Wolfman, T.J. Cortina
& E.L. Walker (Eds.), Proceedings of the 41st ACM technical Symposium on Computer Science
Education (pp. 47-51). Milwaukee, USA, March 10-13, 2010. New York: ACM Conference
Proceedings.
Sahimi, M., Roach, S., Cuadros-Vargas, E. & Reed, D. (2012). Computer Science curriculum
2013: reviewing the Strawman report from the ACM/IEEE Task Team. In: Proceedings of the

11
 43rd ACM technical Symposium on Computer Science Education (pp. 3-4). Raleigh, North
Carolina, USA, February 29 - March 3, 2012. New York: ACM Conference Proceedings.
Sipser, M., Introduction to the Theory of Computation, PWS Publishing Company, 1997.
UNESCO-IFIP. (1994). A modular curriculum in computer science. UNESCO and IFIP report
ED/94/WS/13. 112p.
van Veen, M., Mulder, F., Lemmen, K. (2004). What is lacking in curriculum schemes for
Computing/Informatics? In: R. Boyle, M. Clark, & A. Kumar (Eds.), Proceedings of the 9th
Annual SIGCSE conference on Innovation and Technology in Computer Science Education
(pp186-190). Leeds, UK, June 28-30. New York: ACM Conference Proceedings.

Acknowledgements
We are grateful for all the people who took the time to fill in the survey, and especially
those who provided additional offline/email feedback and information. We also would like
to thank Leonard Els for setting up the survey software.

Appendix A: Universities consulted for the syllabi survey

South African universities (17): UCT, UFH, UFS, UKZN, Limpopo, NWU, UP, RU, Stellenbosch,
UWC, Wits, UJ, NMMU, Unisa, Univen, WSU, UniZulu.
European (continent, west) universities (15): Free University of Bozen-Bolzano and “la Sapienza”
University of Rome [IT], Technical University of Dresden, Technical University of Rhine-
Westphalia in Aachen (RWTH Aachen), and Technical University of Munich [DE], Free
University of Amsterdam and Technical University of Eindhoven [NL], Catholic University
Louvain and Ghent University [BE], Vienna University of Technology [AT], Linköping University
[SE], ETH Zurich and EPFL [CH], Polytechnic University of Madrid and Polytechnic University
of Catalonia [ES].
Universities in Anglo-Saxon countries (15): Oxford University, Manchester, and University of
Edinburgh [UK], Toronto and Vancouver UBC [CA], MIT, Stanford, Penn State, Yale, Harvard,
Stony Brook, and CMU [USA], University of Melbourne and Australian National University [AU],
and the University of Auckland [NZ].
Asian universities (6): University of Malaya [MY], National University of Singapore, Peking
university and the University of Hong Kong [CN], University of Tokyo [JP], and the Indian
Institute of Technology, Bombay.
African (non-SA) universities (7): Alexandria University [EG], Makerere [UG], University of
Zimbabwe, University of Lagos [NG], Kenyatta University [KE], University of Botswana, and
Addis Ababa University [ET].
Latin American universities (8): UCI and University of Havana [CU], Bolivarian University of
Venezuela, Pontifical University of Chile and University of Chile, State University of Campinas
and University of São Paolo [BR], and National Autonomous University of Mexico.

12