Text Box: Project Number: 	Telematics for Research 4007

Project Title:	Multimedia Education and Conferencing Collaboration 
	over ATM Networks and Others (MECCANO)

 

 

 

 

Title of Deliverable:                           Description of, and conclusions from, the

                                        seminars carried out in the second year.

 

Deliverable:                                R8.2

 

Produced by Workpackage:       WP8

 

Contractual Date of Delivery:      June 2000

 

Author(s):                                  Kjell Åge Bringsrud and Tarik Cicic,

                                        University of Oslo

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

Abstract:

 

The MECCANO project built on the results of the MERCI project, further integrating technology components into the tools to improve their utility for multimedia collaboration.

 

The tools used by the project were proven in the activities of the validation workpackage (WP8, Consolidation of Applications); the feedback from these activities was used to improve the tools.

 

A vital aspect of these activities was the use of the tools under the “real” conditions of remote lecturing, seminars and distributed virtual meetings. This deliverable describes several examples of the use of the MECCANO multimedia conferencing tools in such activities during the second year of the project, together with comments on, and snap shots of, the actual events.

 

Keyword list:

 

Multimedia Conferencing, Seminars, Distance Education, Mbone.

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

 

 


Contents

 

1.     Overview................................................................................................................. 3

2.     Mbone seminars...................................................................................................... 3

2.1 UDLR (Unidirectional link routing)......................................................................... 3

2.2 Hewlett-Packard Science Lectures......................................................................... 4

2.2.1 C_60 Buckminsterfullerene: Not Just a Pretty Molecule................................... 4

2.2.2 Topology and Quantum Physics: from Knots to Quarks................................... 5

2.2.3 Tools for Brains: how Technology Created Human Consciousness................... 5

2.2.4 Tangles, Bangles and Knots............................................................................ 6

2.3 The dlb seminar..................................................................................................... 7

2.4 Non-secret encryption and public-key cryptography............................................. 10

3.     ISS UW lectures to SSEES................................................................................... 11

4.     Demonstrations at the TERENA-NORDUnet conference....................................... 11

5.     Seminar announcement and assessment support...................................................... 12

5.1 Announcement and assessment system................................................................. 12

5.2 Assessment results............................................................................................... 14

6.     Evaluation results................................................................................................... 14

7.     Acknowledgements................................................................................................ 15

8.     Links..................................................................................................................... 15

 

Figures

 

Figure 1: UDLR seminar screen........................................................................................ 4

Figure 2: Participants at seminar on June 21st................................................................... 5

Figure 3: Video transmission of overhead slides, June 21st seminar.................................... 5

Figure 4: Screen shots from “Tools for The Brain”............................................................ 6

Figure 5: Snapshot of the seminar on September 6th 1999 recorded at ACC.................... 7

Figure 6: Screen shot of the Digital Lecture Board............................................................ 8

Figure 7: Using Chat and RAT in the dlb session............................................................... 8

Figure 8: VIC video transmission in dlb seminar.............................................................. 10

Figure 9: SSEES students in London listening to UW lecturer in Warsaw......................... 11

Figure 10: Participants in VIC and RAT during the UW lecturer in Warsaw..................... 11

Figure 11: Technical discussions in NTE......................................................................... 11

Figure 12: The MECCANO stand at the TERENA-NORDUnet conference................... 11

Figure 13: Seminar calendar........................................................................................... 13

Figure 14: Section of the MECCANO assessment questionnaire..................................... 14

 


 

1.    Overview

 

The primary objective of work package 8 “Consolidation of applications” was to validate the technology developed by the project within the project itself. The quality and acceptability of MECCANO tools for different user communities was assessed. However, this process is dependent on the network infrastructure on which the tools rely. Development and maintenance of this infrastructure are outside the scope of MECCANO.

 

In particular, international Mbone connectivity has been poor or nonexistent during much of the reporting period. This has affected these project activities including even the regular Mbone project meetings. Having in mind that these project meetings are carried out by experts in the area of distributed multimedia, providing validation activities for wider, international user community has proved to be difficult.

 

As an example, HPLB undertook to investigate the usage of the Mbone for seminar multicasts. This involved setting up a multicast transmission facility within HP labs in Bristol to formally transmit seminars across the Mbone. These seminars were to be based on a series of lectures delivered by BRIMS (Basic Research Institute in Mathematical Sciences) and the IRI (Internet Research Institute).  The primary issues raised in this project were those encountered in establishing a consistent set of seminars to broadcast and the difficulties in establishing reliable connectivity. The latter was caused by a restructuring of HP's connection to the Mbone, which was changed in May of 1999, causing these connectivity problems, and only provisionally fixed  in October  1999. A summary of these issues is included in deliverable 3.3, “Final report on the usability of the MECCANO tools and other components”.

 

Typically, the European Mbone has been very unstable and partly down for longer periods of time during the last year than previously. This has caused much testing and repairing activities within the MECCANO community, in most cases resulting in the use of fall-back solutions, such as the use of various reflectors, to obtain connectivity for the video-conferencing meetings and seminars.

 

However, many validation and consolidation activities have been carried out, both nationally and internationally. The rest of this document provides insight in the activities, and presents the support infrastructure provided. Related work by the project in support of the use of the tools by other groups is presented in the deliverable 9.2, “The support provided to to validation projects during Year 2 (June 1999 – May 2000)”.

2.    Mbone seminars

These seminars were multicast on the Mbone and aimed at wider audience than just project members.  Unfortunately, the poor Mbone connectivity has prevented many users from attending the seminars.

2.1 UDLR (Unidirectional link routing)

Date: May 17th 1999

Speaker:  Walid Dabbous, INRIA Sophia Antipolis

Description: The Internet is mostly composed of networks interconnected via heterogeneous, but bidirectional and symmetric links. Internet routing and transmission control protocols have been designed to optimize the transmission on such communication links.

 

Recently, new transmission media have been proposed to provide higher bandwidth to the Internet. Satellite links with receive only antennas and cables offering respectively unidirectional and asymmetric access to the Internet have been proposed. Both offer high speed bandwidth downstream, with zero bandwidth upstream for the satellite link and a lower bandwidth upstream for cables. Broadcast satellite links may therefore be used to provide alternative high speed receive only access to the Internet, with the return trafic sent through a low speed access. There are many specific problems to be solved in the case of unidirectional links, such as dynamic routing and transmission control. The talk adressed some of these problems as well as giving a general overview of the concept.

 

Figure 1: UDLR seminar screen

Walid Dabbous is (co-)chairing the IETF “UDLR Working Group” which focusses on  the support of alternative unidirectional links on top of a bidirectional internetwork <1>. A screen image from the talk is given in Figure 1. The connectivity during the lecture was rather good, although some participants experienced a high amount of packet duplication. Some participants got their feed from the TEN155 network or regular Mbone. At the most 11 participants attended the lecture.

2.2 Hewlett-Packard Science Lectures

The Basic Research Institute in the Mathematical Sciences (BRIMS) was set up by MECCANO partner Hewlett-Packard in 1994 as part of an initiative to widen the Company's research base. BRIMS is located in Bristol, England and is attached to Hewlett-Packard Laboratories Bristol, the European arm of the Company's corporate research structure <2>.

 

During 1997 and 1998 a series of seminars given by BRIMS was recorded on video tape and some of the series were multicast from UCL during the summer and autumn of 1999. The first three were sent via a TEN-155 VPN link to INRIA and then via the INRIA DBS satellite facility provided by EUTELSAT. This facility has been used by MECCANO to experiment with high bandwidth multimedia conferencing tools for education and research. The last was transmitted using the normal Mbone.

 

The BRIMS-seminar series was seen mostly by MECCANO partners equipped with satellite receiver equipment. Poor regular Mbone connectivity made earthbound multicast difficult. The perception was excellent due to the fast satellite links. Even with good quality links, the need for a distributed publishing tool is apparent: video transmission of overhead slides gives only the moderate results shown in figure 3.

2.2.1 C_60 Buckminsterfullerene: Not Just a Pretty Molecule

Date: First transmitted June 7th 1999; resent June 21st. 1999

Speaker: Professor Sir Harry Kroto FRS, Royal Society Research Professor of Chemistry, University of Sussex

Description: In 1985 C_60 Buckminsterfullerene (the third allotropic form of carbon) was discovered during experiments designed to unravel the carbon chemistry in Red Giant Stars. The molecule has a beautiful and elegant symmetry, as well as other attributes, which was further explained by Dr. Kroto during his talk.

 

Figure 2: Participants at seminar on June 21st

Nine people from the MECCANO community attended the multicast transmission of the lecture, mostly with good receiption quality. Screen shots from the lecture are shown in figures 2 and 3.

 

Figure 3: Video transmission of overhead slides, June 21st seminar

2.2.2 Topology and Quantum Physics: from Knots to Quarks

Date: July 5th 1999

Speaker: Sir Michael Atiyah OM FRS, Master of Trinity College, Cambridge

Description: Classical physics has a long and intimate association with geometry, culminating in Einstein's theory of general relativity.  The topologist's knots have been enlightened by quantum ideas and the mysteries of quarks are being clarified by Dr. Atiyah in the same framework. This seminar was originally delivered and recorded May 27th 1997.

 

Because the TEN 155 link between UCL and INRIA was down, multicast of this seminar had to be abandoned.

2.2.3 Tools for Brains: how Technology Created Human Consciousness

Date: Originally scheduled for August 16th 1999; rescheduled for September 6th.

Speaker: Professor Daniel Dennett, Department of Philosophy, Tufts University Description: The emergence (in history and, in the individual, today) of human conciousness with all its powers and idiosyncrasies, depends on the invention and proliferation, through human culture, of a variety of artifacts. Professor Dennett presented his (technology oriented) views upon this topic. This seminar was originally delivered and recorded 10th December 1997. Four participants followed the multicast retransmission of the lecture. The quality was fairly good, and reception went well. Snapshots of this seminar are shown in figure 4.


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4: Screen shots from “Tools for The Brain”

2.2.4 Tangles, Bangles and Knots

Date: 6th September 1999

Speaker: Professor John Conway FRS, Professor of Mathematics, University of Princeton

Description: The mathematical study of knots used to be centered on the study of knot invariants: mathematical devices which can tell – sometimes - whether a knot is “knotted” or not. It has now become so easy to construct knot invariants that mathematicians have started looking more closely at the geometrical structure of knots. In his talk, Dr. Conway showed how lots of knots can be easily described as “bangles” made out of “tangles”, which temselves can be simply described by numbers. The talk was originally delivered and recorded 29th April 1998.

 

During the multicast retransmission of the seminar, some 6 people participated. Due to problems with Mbone connectivity, the visibility varied somewhat during the transmission and some participants lost connection. A snapshot taken at ACC is shown in figure 5 where only 4 participants are visible in the RAT window, but, at the time, all 6 participants could see the actual transmission.

 

Figure 5: Snapshot of the seminar on September 6th 1999 recorded at ACC

2.3 The dlb seminar

Date: January 24th 2000.

Speaker: Werner Geyer.

Description: The digital lecture board seminar took place on January 24th 2000. It was a joint seminar of the MECCANO project and the COST264 project. Its main aim was to demonstrate the capabilities of the digital lecture board to the participants. 13 participants from 6 European countries joined the session. The presentation was given by Werner Geyer of the University of Mannheim.

 

The Mbone connectivity was symptomatic of the problems that we experienced throughout the whole MECCANO project. Some sites could connect to the session using the Mbone while the majority of participants used a packet reflector that was set up at the University of Mannheim. After the session all participants reported that they received the transmission with good quality (either through the Mbone or via the reflector). Two participants from UCL joined the session via multicast but were never seen by anyone outside of the UK. However, they reported that they received the seminar and that the quality was good.

 

The session itself was successful. We took several screen shots of all the tools involved. Figure 6 shows the digital lecture board in action. It illustrates how a screenshot of the dlb was loaded into dlb, so that the speaker could discuss and annotate the user interface.


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6: Screen shot of the Digital Lecture Board.

The chat tool in the dlb helped us to administer the session without disturbing speaker or audience; RAT was used for the audio transmission (see figure 7)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 7: Using Chat and RAT in the dlb session   

The UCL version of VIC was used for the transmission as shown in figure 8.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Figure 8: VIC video transmission in dlb seminar.

2.4 Non-secret encryption and public-key cryptography.

Date: May 16th 2000

Speaker: Dr. Whitfield Diffie

Description: In a remarkable case of “scientific parallelism”, a secret British group and a public American group, working entirely independently, made very similar discoveries during the early 1970s. The discovery was a revolutionary new form of cryptography. What was “Non-secret Encryption” to the British and “Public Key Cryptography” to the Americans, is at the heart of internet commerce and is achieving wide use throughout telecommunication.

 

At a UCL public lecture on 29 April 1999, Dr. Diffie ranged from reflections on the personal experience of discovery to an examination of the techniques developed. This seminar, a replay of a video recording of the event, was multicast only, and was meant to be the final MECCANO test of the European Mbone. Once again, the test failed. Only 2 to 4 of a total of 7 participants managed to establish Mbone interconnection and, for most of the time, the Mbone was down. No screen shots were taken.

3.    ISS UW lectures to SSEES

This series of lectures from the Institute for Social Studies, Warsaw University, Poland (ISS UW) was transmitted over the Mbone using MECCANO videoconferencing tools. Listeners to the seminars were, in the main, from the School of Slavonic and East European Studies (SSEES) in London, but MECCANO project participants also frequently listened in. In addition to the interesting content, the weekly multicast has provided another very good test of the European Mbone.

 

The series of ten lectures, transmitted from the Center of Open Multimedia Education each week from January 12th to March 15th, was entitled: “Social Change: Adaptation and Resistance”. Detailed information on the contents of the specific lectures given during this seminar series is available on the ACC web site <3>.

 

The seminars were announced on the Mbone using SDR in a session named “UW Lectures”. However, due to the persistant problems with European Mbone connectivity, the lectures were often joined using a unicast reflector at ACC. Screen shots from the lectures are shown in figures 9 to 11.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 9: SSEES students in London listening to UW lecturer in Warsaw.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 10: Participants in VIC and RAT during the UW lecturer in Warsaw.

 

 


 


Figure 11: Technical discussions in NTE

4.    Demonstrations at the TERENA-NORDUnet conference.

The conference took place at the beginning of June 1999 in Lund, Sweden. The University of Oslo (UiO) and New Learning A/S set up a stand representing the MECCANO project. Two demonstrations were performed and a number of handouts and posters were made available.

 

One demonstration showed Mbone over satellite, demonstrating how high quality Mbone can be received using a cheap satellite dish and a PC card. The other demonstration showed the distributed electronic class room through a PC connected to the class room at UiO.

 

Many people visited the stand and the MECCANO people had good discussions with conference participants. It seemed that many people were concerned about the poor quality of the European multicast infrastructure. Figure 12 shows the MECCANO stand.

 


 

Figure 12: The MECCANO stand at the TERENA-NORDUnet conference.

5.    Seminar announcement and assessment support

5.1 Announcement and assessment system

A calendar with an overview of planned seminars was provided on the Web. Days with seminars are highlighted, providing a quick link to the description. It is also possible to view the seminar descriptions sequentially. A screen shot of the calendar web page is shown in figure 13.

 

A questionnaire was developed, aimed at investigating the experiences felt by the participating parties in a MECCANO multicast seminar. All seminar attendees were encouraged to fill in this assessment form. Part of the form is shown in figure 14.

 

Further details of the system are available from the seminar announcements and questionnaire web page at UiO <4>.


Figure 13: Seminar calendar

 


 

 

Figure 14: Section of the MECCANO assessment questionnaire

 

5.2 Assessment results

Date

Seminar

title

SDR/

SPAR

Comm

.line

Audio

quality

Video

quality

Mbone
Reflec-
tor

Satellite

TEN155

Loss/

duplic.

No

connect

990517

UDLR

 

11

good

good

3

5

3

some

dupl.

 

990621

Not just a

pretty molec.

9

 

good

good

3

 

6

20-30%

loss

 

990705

Topology &

Quant.phys.

 

 

 

 

 

 

 

 

all parties

990906990815

Tools for the

brains

1

3

fair

fair

1

 

3

5-10%

loss

 

990906

Tangles, ban

gels & knots

2

4

fairly

good

not too

good

2-4

2

 

5-20%

loss

1-3

000124

DLB

 

13

good

good

5

8

 

5% loss

 

000516

Non-secret

encryption

3

1

OK if

seen

OK if seen

2-4

 

 

5-20%

loss

4

 

Table 1: Summary of entries in MECCANO assessment questionnaire

Table 1 gives a summary of the answers received via the assessment questionnaire. It shows that the quality of the video and audio reception was usually rather good when the connectivity of the network was adequate. Usually, however, the Mbone did not provide a sufficient degree of interconnectivity to enable the participants to see the announcements in SDR or SPAR, so that one or more reflectors had to be used instead of multicast. The connectivity situation seems to have become worse during the period autumn 1999 to spring 2000; trying to use the Mbone multicast during the final MECCANO seminar on May 16th 2000, no participant actually managed to connect.

 

Similar problems were experienced during the MECCANO Management meetings, during which we expected to do most of our trials of the MECCANO tools in multicast mode. The private MECCANO web page containing the meeting minutes shows that there were problems at all of the ten on-line management meetings held between June 1999 and May 2000, with one being abandoned and the final four relying on reflectors.

6.    Evaluation results

Preliminary evaluation has indicated the following conclusions can be drawn:

 

·        provided that the network infrastructure is working satisfactorily, the tools used are sufficiently stable. It is therefore feasible to hold regular lectures and even examinations with the MECCANO toolkit.

 

·        if the technical quality of the transmission is good enough to follow the session, then the quality of preparation of the content presented is a far more important factor than improved audio and video quality. In particular, additional video feed of the audience at the source is unimportant.

 

·        the local setup of the audio and video hardware is of extreme importance. Adjusting the microphones to the right level, correct placement of microphones and providing adequate lighting for the video are much more important than enhanced tools for audio and video transmissions.

 

·        a key problem is the user-friendliness of the tools. It would be desirable to have a simple user interface that does not overwhelm non-computer scientists. Currently some of the MECCANO tools are not well suited to use by non-technical users.

 

·        the concept of displaying the user interfaces of multiple tools is confusing. It would be better to have a single user interface, such as the ReLaTe interface, which hides everything but the shared workspace and the video of the lecturer.

 

·        anything required to setup the session should be invisible, once the lecture starts.

 

·        the software should run under Windows95/98/2000.

 

·        the import format for shared workspaces should be extended to allow PowerPoint and html presentations. Postscript or plain text are not appropriate for most lecturers.

 

·        students regard the ability to access the recording of lectures as extremely useful, especially for the preparation of examinations. The first press run of CDs of a lecture series given in the Viror project <5> sold out within a couple of weeks and the web based replay service experienced a peak access rate of 1000 hits per month for the recording of lectures. This is a very high number for the 100 or so students who normally attended the lectures.

7.    Acknowledgements.

UCL, London:      Roy Bennett (screen snapshots, demos)

RUS, Stuttgart:     Andreas Rozek (snapshots, network performance, demos)

8.    Links

<1>   More info on the UDML tool on web page:

         http//:www-sop.inria.fr/rodeo/udlr/

<2>   Description of the HP/BRIMS seminar series on web page:

         http//:www-uk.hpl.hp.com/brims/

<3>   Overview of the MECCANO UW Lectures on web page:

         http//:www.ics.agh.edu.pl/Meccano/

<4>   MECCANO seminar announcements and questionnair web page:

         http://mbone.ifi.uio.no/meccano/seminars

<5>   The virtual university project (VIROR)

          http://www.viror.de/