Author: Olof Hagsand, NADA.
On-line version of course evaluation
Number of credits: 5p
Written examination TEN1 : 2p
Laborations LAB1 : 2p
Recitations HEM1 : 1p
Course dates: period 3, 2005
Course instructor: Olof Hagsand
Guest lecturers: Roland Elverljung (KTHNOC)
Måns Nilsson (KTHNOC),
Malin Carlzon (KTHNOC)
Laura Marie Feeney (SICS)
Kurt-Erik Lindqvist (NetNod)
Anders Hillbo (KTHLAN)
Per Nihlen (KTHNOC)
Lab assistants: Roland Elverljung, Gunnar Kreitz
Number of lectures: 12 (34 hours)
Number of recitations: 5 (10 hours)
Number of laborations: 8 (32 hours)
Number of recitation groups: 1
Number of lab groups: 1
Web page: http://www.nada.kth.se/kurser/kth/2D1490
TCP/IP Protocol Suite, Second Edition (Behrouz A. Forouzan)
Printed course material (kurshäfte) of RFCs, lab instructions, etc
Number of registered students: 18
Number of students directly aborting course: 4
Number of students passing TEN1: 14 (2005-03-10)
Number of students passing LAB1: 18
Number of students passing HEM1: 18
Number of students passing re-exam TEN1: ? (2005-04-08)
Degree of performance: 78%
Degree of examination: 78%
The first part is an addition to the course since two years back when the introductory course (2D1392) was removed. There are other introductory courses (2G1316, 2G1305) but none of the students had taken these.
The introduction part followed the course book (Forouzan), and the lectures were given by Roland Elverljung (2) and Olof Hagsand (1). These lectures do not really belong to the course - since they are too fundamental. It is difficult to give a complete internetworking coverage in a few lectures. Instead, this should be prerequisites to the course, such as the re-introduction of 2D1392.
The main lectures (internal routing) was given by Olof Hagsand (3) and Malin Carlzon (1). The lectures are the weak point from a pedagogic perspective in this course. While the recitations and labs are very much to the point, the lectures were more problematic in conveying the subject to the students. It would be better to spend more lectures on the central parts of the course. Still, the student review gave good grades to the lectures, with some exceptions.
There were five guest lectures on subjects related to the course, or to internetworking in general. The students had varying comments on these lectures, and many students chose not to be present. Many did this since they were not covered by the exam, lab or recitations. It is questionable to have such an extensive guest lecture program for a handful of students - at least out of respect for the guest lecturers.
Some of the lectures were scheduled to three hours. For some lectures, it was good to have some extra time, but the general view was that three hours is one hour too much.
The recitations were popular by almost all students, and worked well to convey the subject. Much better than the lectures. They were also excellent to prepare the students for the labs.
The labs are made in room SAM at Teknikringen 14. The students were partitioned into five groups, with each group having four work-stations running FreeBSD, and four CISCO routers. The labs are very much "hands-on" including connecting all wires and networks for each lab.
One observation is that there might be too much emphasis on OSPF, one or two of the intermediate OSPF labs could be removed and replaced by other topics, such as ISIS and Multicast.
Four out of five recitations were also mandatory, as were the homework for each recitation. All registered students passed.
The exam was slightly enhanced and more difficult compared to previous years. However, since it followed the recitations and labs quite closely, the result was good. One exam question (RIP) was unfortunately not well designed, the question was more advanced than intended and this caused some confusion among the students.
Three persons got grade five, three got grade four, six got grade three. One person did not reach grade three, but obtained grade three after an oral exam. Three persons chosed to make the exam at a later point - an oral exam will be given on April 8.
A problem with the written exam follows the layout of the course. It is only the central parts (internal routing) that is possible to make a written examination on. The introduction is too fundamental, and the related work is too diversified.
Therefore, the course design might benefit from removing the introductory part and making the guest lecture part smaller.
Therefore, the book gives a good background to the subject, but is not really a help to the students when reading for the written exam.
The central literature for labs and exam is the standardization documents: RFCs. These documents are inherently difficult to read, but the students managed to do this quite nicely. However, it is essential to have good reading instructions - these were available on the web.
It can be noted that actually reading a standardization document might be useful for future engineers, since they will probably often encounter such documents in their future proifessional career.
Appendix 1: Result of student course evaluation