Integrating Technology into Computer Science Education

By Jacob Palme, e-mail: jpalme@dsv.su.se, at the research group for CMC (Computer Mediated Communication) in the Department of Computer and Systems Sciences at Stockholm University and KTH.

The conference on "Integrating Technology into Computer Science Education" was organized by ACM and Uppsala University in June, 1997. These are my personal votes from some of the sessions during that conference. Statements below are often quotes from what the speakers said at the conference, and may not represent my personal opinions.

MANKIND project

Hermann Maurer, Graz University, Austria

Not remote lecturing. More asynchronous than synchronous communication. Asynchronous communication allows students to think, look up words, and then put their questions and read the replies at times suitable to them.

Server with educational material in small modules, so that modules can be combined for the needs of different students. Courses contain multimedia and interactive software. Student can add notes to the lecture material they find on the web. They can put questions asynchronously. Note: Most web pages are not modular, they contain lots of links in and out which means they can only be used in the environment where they were created.

Annotations can be personal or group, and there can be annotations on annotations to any level. Tree-structured discussions can occur.

Communication: One-to-one, group, shout for help (when you need immediate help on a problem). Shouts for helped should immediately be routed to approporiate people who can help you, not to everyone, not to no one active when you shout.

Traditional WWW structures show the same structure to everyone, which is stone age according to the speaker, the structure of information should be different for different recipient needs.

Meta-information is needed: Creation date, atuhor, expirantion date, etc. The kinds of different meta fields should be open-ended.

Structure should automatically update itself, not have a fixed structure which humans have to update manually. For example: Future becomes past automatically, no one has to edit the calendar page once a day or once a month to update future into past.

Structure: Documents are collected into collections, collections can become elements in other collections, etc. The speaker had the opinion that structure should be enforced automatically by the software, so that no thick manuals are needed to adhere to the agreed structures for a web page. Goto-s and pointers should be avoided an replaced by collections where the software keeps items in their right place in the collections. Links should be bi-directional, the speaker said. Only in that way can we avoid links which do not work any more. Links should not be embedded in documents, then they will not be updated automatically when for example the document you are linking to is changed or deleted.

You should be able to add links to documents, which you do not own, this is not allowed by the traditional WWW structure.

www.hyperwawe.com implements all this, earlier known as Hyper-G.

Questions from the audience: How can the author design a page if the author does not know which links will be in the page for different recipients. How can you automatically remove links from documents without potentially invalidating the whole document?

My personal comment: The speaker seemed very envious that his system, Hyper-G, had not become as popular as the World Wide Web. Both systems were created at about the same time, and Hyper-G was in his opinion much better. However, he did not seem to understand that it is the simplicity of the WWW protocols which has made it so popular. WWW is a kind of modern "virtual terminal" protocol, filling similar roles as 3270 terminals did a decade ago. The advantage with WWW is that it is so easy to map so many existing structures on the WWW protocol.

Teaching via the Internet: The Impact of the Internet as a communication medium on Distance Learning Introductory Computing Students

Linda Carswell, Maths & Computing Faculity,The Open University, UK
 
Different kinds of distance learning:
Sattelite Teaching
All students together, but at a different campus.
Correspondence Teaching
All interactions by post, mainly paper based, no face-to-face meetings
Supported Distance Learning
different media, range of presentation media, face-to-face meetings, counselling, student support groups

Comparison: Her study compared two groups of students. One group was studying by the Internet, the other using conventional distance learning methods. Both groups got the same learning material by the post (not web-based). The difference was that the Internet students communicated with the teacher through e-mail and asynchronous conferencing. The conventional students communicated through phone and face-to-face group solving sessions. The only difference was the communication medium for communication between students and between students and teacher.

Subject of the course: Introductory computing.

Assessment method: assignment and exam grades using standardized and validated exam assignments.

Result: Very small difference, conventional students slightly higher average grades, but difference too small to be able to draw conclusions that one medium is better than the other. Conclusion: Learning otcome does not depend on the medium difference in the setting tested in this course.

Does this mean there is no difference? No, students reported:

Using Multimedia Communication Technologies in Distance Learning

Thomás Robles, Davied Fernández, Encarna Pastor and Santiago Alamillo, Universidad Politéchnica de Madrid, Spain.

Aim: To evaluate the adequacy of multimedia communication in distance education.

Teaching methods: Students at local workstations connected by ISDN to the teacher. The ISDN could send computer-generated images, video and still pictures.

Student perception: Like TV. Students experienced live lectures the same way as pre-recorded videos, only 4 % asked questions, a very passive student situation.

Conclusion: We have to break the "glass barrier". Offer visual variability. Alternate the methods of interactions. Short sessions, not too close together.

Social, Ethical and Professional Issues of Computer Science Education

C. Dianne Martin

The speaker produced a large two-dimensional matrix with columns for different ethical issues, and rows for different computer applications and computing technologies. How narrow should computer science ethical courses be? Should ethical issues in other disciplines be included (like organ transplant priorities) or should the examples be from the computer area?

Access control on the Internet: How to protect children from information which some people believe can be dangerous to children. The solution: Content-labelling systems, content-advisories, blocking systems. But these solutions have their own new etchical problems.

Example: Content labels should be descriptive, non-evaluative. Does it just desdribe, or also give a value judgement. Typical example: Age-based limits are evaluative, since the opinion of what is suitable at different ages is a value judgement where opinions may differ.

Labels can be deterministic or non-deterministic. Deterministic are ratings that different people will make the same rating, while non-determinstic are ratings which will differ depending on who makes the ratings.

RSAC: An independent, non-profit organization set up in 1994. Its mission is to empower the public, especially parents, to make informed decisions about electronic media (initially: computer games) by means of an objective system.

Result: Two rating systems: One for computer games, one for computerized game platform games. Developed by different organisations.

RSAC (general computer platforms): Voluntary, objective, anyone can look at the ratings and how they were made (RSAC = Recreational Software Advisory Council).

CDA (game platform games): Evaluative, age limits, decided by anonymous votes in rating groups. In 1998 all televisions must have a vchip with scales: garbage, pure garbage, 13 years garbage, revolting, naughty cable. Disliked by producers, they prefer factual labels. Television industry dislikes all this, they believe they would lose advertising revenue on pages rated not suitable for children. The industry is threatening to try to invalidate such laws by invoking the first amendment protection of free speech. Similar to schemes for rating television programs.

RSACi: Internet version of RSAC. Oriented towards protecting free speech and empowering parents to make the final decisions. Sponsored by big computer manufacturers. Can be expected to be found unconstitutional.

RSAC: Separate figures for nudity, violence, etc. Easy to adopt to the Internet.

PICS is not a system for one particular rating scheme, it can be used for any kind of rating, even indexing systems. The users of PICS decides which scales to use.

RSACi (PICS based): Four categories; Violence, Nudity, Sex and Language, each categories has five levels from 0 to 5. International interest. Ideal: An international, culture-independent labelling system. In addition, some cultures may want to add their own additional scales.

Granularity problem: Rate pages or folders. Advice: Try to rate your own web site with RSACi!

Rating can be done at the national level: Not allowing certain web pages to anyone in an entire country. Example: Singapore might want to do this. Thus, this becomes an instrument for censorship although not meant for that.

Very clever kids can bypass blocking software.

New version of PICS: Encryption, cryptographic seals and signatures on labels.

Netscape has refused to incorporate PICS into Netscape, because one person there regards it as censorship. Microsoft has implemented it in the Microsoft Internet Explorer.

Esther Dyson in Jan, 1997, Release 1.0: "We do need governance on the Net..."

Three dimensions: Techical, Social (individuals, communities, organisations, cultures, institutional, global/national), Ethical: Responsibility, Quality of life, Power to control other people, Risk and liability, Privacy, Equity and access, Honesty and deception.

My personal comment: A very intersting and thought-provoking speech. A reaction from me can be found at URL: http://dsv.su.se/jpalme/reports/right-wrong.html.

Using Network Simulation Package to Teach the Client-Server Model

Michael W. Dixon, Tanya J. McGill and Johan M. Carlsson.

Students can learn the client-server by programming it themselves. But then they have to learn a lot of programming techniques first. This did not work well in practical application. So how can they be taught by demonstrations which do not have the problems of letting them actually write programs.

Students like graphics. The simulation package OPNET was chosen. OPNET is a descrite network simulator that is comprehensive enough and has good visual display of what is happening.

Teach students to understand what is happening through animation in event state finite state graphical depictions. Students can see what is happening, can play it back slowly to see what is really happening. Students can define modules, tcp and ip encapsulation, ethernet connections, build all the layers to understand what is happening at each layer.

Students can also see the performance effects of various design options.

Students say they have learnt much more in this way than with conventional lectures.

Recursion and Grammars for CS2

Viera K. Proulx, Northeastern University, U.S.A.

The main idea of this lecture is that students are taught to write programs using more and more top-level concepts to use modules defined at lower levels. But the students are then shown the actual low-level instructions which will be executed when this is executed. Students are shown how inefficient certain recursive algorithms can be. In this way, the students learn to understand what is really happening behind the program they are writing. This is shown graphically to make students understand it better.

Is technology harmful?

A. Joe Turner

Too much innovative technology can be harmful.

Bandwagons in previous years of computer science education:

None of these ideas are actually wrong, but too much emphasis on one of them, when it was very popular, might have been harmful, especially when these bandwagons were promoted with religious zeal.

Pressure to use technology: Everyone wants us to be more efficient, press us to use technology, sometimes without knowing what we really want to do with it. Example: It may not always be cost-efficient to put all information on the web, and then have the students print it on the laserprinters in the lab. We could print this more efficiently in other ways.

Technology has made programming teaching easier by better debugging and visualisation tools, and communication among students and teachers has been facilitated. But there has not been any real "wow" type of gain in increased efficiency of education.

Current direction: Lots of work keeping courses current, inclusion of social/ethical context and other professional issues. There has been more interest in really understanding teaching/learning paradigms and how they work.

One area where computers are becoming useful is automatic testing and assessment of the programs developed by students. Another area is class management, electronic submission of tasks, providing feedback and assessment electronically, etc.

Why is this not progressing faster? Because it is a lot of work, especially since courses in our area get out-of-date within a few years. It is much easier for the teacher not to develop electronic tuition facilities in his/her courses. It can also be incovenient for students, because access to facilities can cause problems. Most students print a copy of the web pages and study from the printed copy, and then all the neat animation and Java you are using is of no value.

People claim this is solved by technology advancements. They have always said that. But is it true? The average cost of a typical personal computer has been constant, 5000 dollars, in the last fifteen years. It has not been going down! Increases in performance has been counteracted by increases in the requirements of computer power in the software.

Algorithm animation software: Makes students too passive, there is no challenge in it for the students, they just run and observe. Students learn more if they build the animation themselves, not just observe animation built by the teacher.

Conclusion: Technology will be more beneficial if it is not just a new vehicle for delivering the same information, but supports new processes. Examples: Electronic funds management, automatic teller machines. Can we do the same in the educational sector? Paradigm shifts are needed to gain benefits.

A controversial talk, lots of discussion after the talk.

On-line seminars

Jaqueline Taylor, Bournemouth University, U.K.

An evaluation of student's experiences when seminars were conducted asynchronously using a text-based conferencing system.

Aims: Student should be able to participate more, and learn about CMC.

Lots of interesting references on impact of CMC on inter-personal communication.

Communication skills and preferences likely to affect the use and perception of CMC.

CMC reduces anxiety, beneficial to some students, claim by Adrianson.

Result: Students found on-line seminars more rewarding, but also requiring more work from them.

Negative: Technical problems, network access problems, speed problems.

Strongly extravertive people are more dissatisfied with CMC, they need the more rapid feedback of F2F communication.

Groups, technology and inter-cultural exchange

Anita Jawari, Christiana Birchek, Susan Strack Vargo

Learn art of giving clear instructions, art of persuasion.

Task: Design a timing device, and then sell it to the lecturer, including persuading the lecturer that the tool they had developed really worked. Analyse their own communication in the group using Barker 1993: Communication (Sixth edition, Prentice Hall, pp 213-26). Describe the roles played in the meetings.

Clture collission between students in Australia and the USA. Words used by the Australians considered offensive by some of the Americans (example: "Black music", should have been "African-American music"). Many did not have English as their native language, and the students did learn about words which are difficult to understand by non-native readers (example "bull dog clip", "spiral binder").

Students believed e-mail could solve all problems. They expected immediate feedback, not have to organize their ideas before writing them, etc.

Use of computer conferencing to teach a course on people, computers and society

Jacob Palme and Sirkku Männikö

This was our own contribution to the conference. You can find the full text at URL http://dsv.su.se/jpalme/distance-education/ITICSE-Report-88.html.

Ethics, programming and virtual environments

Michael Houle and Simon

How to teach the social, legal, ethical, professional issues relevant to the computing disciplines?

How to explain privacy to crackers? How to explain copyright issues to students who have pirate copies of software?

Platform: Moo, a variant of MUD.

A MUD system can be seen as a small interactive simulation of reality. Typical features: rooms, users move from room to room, the current room and its inhabitants and contents appear as text-based descriptions. Users in the same room can converse.

Traditional MUD usage: Role playing in games.

Common actions:
Exploration: go, look, read, examine
Communication: say, page, emote, whisper
Manipulation: take, put, open, close

MOO = Mud Object Oriented. All entities are objects. Hierarhcy of objects. Verbs written in MOO language. Users can program their own virtual reality objects.

Example: A student wrote an automated tour guide with verbs like "hello", "follow me".

Uses of MOOs:

This course: Partly empty building, students can create new rooms with their own equipment.

2 hours F2F lectures/week plus on-line usage.

The MOO was:

Tasks for the students:

Stage 1: Familiarizaon
Create a persona for yourself
Explore the environment
Learn MOO
Some learned MOO programming
Stage 2: Social interaction
Cooperation in building rooms in the MOO
Conflict: Impersonation, spoofing, rude behaviour
Spoofing: Impersonating other people on the net
Stage 3: Debate
Standards of behaviour: Is this just a game, or is it serious business?
How can standards of behaviour be enforced?
Can different standards of behaviour coexist?
Stage 4: Investigation
MOO communication, leading to
MOO system security, leading to
Understaanding tradeoffs of functionality and security
Understanding issues of freedom & speech
Stage 5: Implementation
Codes of conduct
Danger zones and safe zones
Voting machines
Stage 6: Integration
Tie everything together
Reinforced in final exam

Outcome: Postive student feedback. Sometimes too much work, some students seemed to be addicted to the system. Many students continued improving it during the summer vacation.

Multi-Media integrated into CS2

Ursula Wultz, Daniel Domen and Michael McAuLiffe

Learn programming a large system together and have fun at the same time: This requires graphics and interactivity. Students were given the task of developing an interactive childrens story. Consider also ethics of what you give to children. Topics integrated: Data, algorithms, project dynamics, ethics, multi-media, story-telling. Difficult task: integrating sound with animation, flow control issues. Distinctions between plagiarism and cooperation. Can you take other people's pictures and use them?

Miscellaneous interesting ideas picked up in various presentations

Problem: Different users have different size screens, windows and web fonts, which means that web pages which look nice for the sender may not look nice for the recipient.

Solution: Provide a calibration page first, the student is asked to modify his browser settings so that this page looks nice.

http://www.webct.com popular educational software systems, available for low cost for educational usage.

Working groups

An important part of the conference was six working groups of experts on various areas. Each working group was to develop, during the actual conference, a report. The working groups met a whole day before the start of the conference, a whole day after the end of the conference, and at various times during the conference.

I participated in a group whose task was to develop a report on the use of collaborative software as educational tools. I think the report from our group will be good, but I do not have the full text yet. Unfortunately, ACM seems to claim some kind of copyright on the group reports, so that these cannot be put up on the web. This caused a large controversy between me and the chairperson of the ACM SIG, since I claimed I was not willing to participate in development of a report if the report does not become publicly available.