High- vs. Low-quality Video Lectures

Don't Worry, Just Put them Online!

Robert Schleicher, Tilo Westermann and Sebastian Möller

Quality & Usability Lab, Telekom Innovation Laboratories, TU Berlin, Ernst-Reuter-Platz 7, 10587 Berlin, Germany

Keywords: Video Lectures, Video Quality, Real-time Annotations, Crowd Sourcing, Teaching.

Abstract: This position paper claims that a major obstacle of offering video lectures for public universities appears to

be the fact that they intend to compete with prestigious private universities regarding quality of the videos

and complexity of the installed platform without being able to provide the additional resources required to

do so. We argue that in other areas of teaching this issue has been acknowledged for a long time, and lack-

ing resources are usually compensated for by primarily two means: individually offering provisory course

material (manuscripts), and active participation of the student body in administering those. Based on this, a

simple system is proposed that mostly draws on existing platforms and tools, and refrains from extensive

video editing prior to publishing. We discuss technical and non-technical requirements and possible re-

search directions that result from establishing such low-fidelity video lectures.

1 MOTIVATION

More and more private universities provide video

recordings of their courses, and some even special-

ize on online lectures targeting a huge audience,

therefore called massive open online courses

(MOOC). Examples are Udacity, Coursera, edX.

The format switches from 90 minute lecture record-

ings to short clips enriched with visualisations or

augmented reality to attract a worldwide audience

beyond the campus.

Public universities try to copy this approach and

will thus have to compete with these offers, while at

the same time frequently lacking comparable re-

sources, e.g. time of already employed lecturers,

dedicated technical staff etc. Exemplary recordings

in prestigious private universities are achieved with

the help of a large group of dedicated experts – the

MIT Open Courseware team for example includes

more than 20 people

, while at many public German

universities the basic idea appears to be that the

lecturers accomplish recording, editing, and publish-

ing themselves. Supplying additional staff is usually

limited to pilot projects which are then difficult to

maintain once funding ends. Of course there are

exceptions, but the aforementioned statement sum-

http://ocw.mit.edu/about/ocw-team

marizes the situation we were confronted with when

we started to inquire ways to provide video record-

ings ourselves. Our experience motivated us to con-

sider alternative ways to tackle this issue, which we

will describe in the following.

1.1 General Rationale

The first step was to bring to mind the assets many

public German universities have in this case:

 A large audience that will be attending the lec-

tures in any case as it is required by their

study regulations, i.e. the predominance of on-

site learning with supplementary online mate-

rial over mere online classes.

 a long tradition of providing students some-

times ill-formatted, still highly informative

provisory course material, i.e. manuscripts of

the basic course content, which cannot com-

pete with published books in terms of layout,

but are at least as valuable with regard to con-

tent, partly due to the fact that they do not

have to consider copyright issues to the same

extent as an 'official' publication. Unlike in

universities with high tuition fees, students

expect much less professionally edited materi-

al as long as it is free or inexpensive, e.g. the

master copy template residing in a library for

duplication. Concomitant with that, there is

204

Schleicher R., Westermann T. and Möller S..

High- vs. Low-quality Video Lectures - Don’t Worry, Just Put them Online!.

DOI: 10.5220/0004414802040209

In Proceedings of the 5th International Conference on Computer Supported Education (CSEDU-2013), pages 204-209

ISBN: 978-989-8565-53-2

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

usually no centralized supervision of these

manuscripts, but they are rather issued single-

handedly by the lecturer.

 A mentality of active participation in the

courseware preparation process among the

student body, manifested in student associa-

tions (in German: 'Fachschaft'), and less the

expectation of being treated as paying educa-

tional customers. It is common that these stu-

dent associations offer access to self-written

summaries of textbooks or exams, usually un-

der the premise that the borrower contributes

to the available corpus e.g. by writing exami-

nation minutes him/herself.

Web 2.0 communities also rely to a large extent on

user-generated content. Instead of having a central-

ized professional editorial team, quality control is

usually achieved by letting other users rate and

comment the contributions, which appears to work

stunningly well (Giles, 2005). A key aspect is that

user involvement in the preparation of the material is

quite high, which might be desirable also for the

preparation of course material.

Combining these two approaches, offering edu-

cational 'raw material', e.g. mostly un-edited film

footage via existing platforms like YouTube plus a

couple of low-fidelity tools to annotate and extend

this material might be a feasible and perhaps even

desirable way to proceed for public universities with

too little resources for preparing online material with

high technical quality.

In the following, we will first describe a couple

of such desired functionalities and outline their pos-

sible implementation to provide the reader with a

more specific idea of what we have in mind, and

also make clear where it extends previous similar

proposals, e.g. Copley (2007). Subsequently, we will

discuss non-technical requirements which will also

clarify what makes us favour such a system, and

mention research questions that might arise. Finally

we will summarize the idea and our motivation in

the conclusion.

2 EXEMPLARY SYSTEM

A combination of a couple of basic functionalities

that are in part already available on popular web

sites might suffice to allow for simple processing of

lecture recordings. By processing we do not refer to

video-editing as it is done with software like Adobe

Premiere®, Camtasia® and the like, but the attempt

to enrich the content of a lecture recording in order

to facilitate understanding of it. In particular, these

functionalities might be:

 A simple way to make videos available online

for a larger audience, i.e. upload them some-

where.

 The possibility to add text comments includ-

ing links to slides or other web documents at

certain points of time in the video. We will

call these annotations, and as far as they are

done subsequent to the lecture, subsequent

annotations.

 A listing of these annotations that can serve as

a rough table of content or index for the video.

 The possibility to give simple ratings via a 5-

star or thumbs up/down scale of added com-

ments to indicate their usefulness

 A way to perform annotations during the lec-

ture, e.g. to mark important or less understood

parts. We will call this real-time annotations.

Most of these functionalities are included in pro-

fessional video editing software, however, next to

their price, they also require considerable training

and a deeper understanding of the underlying data

organization in terms of projects, audio- vs. video

track, codecs, and so on. Similarly, the most prom-

inent open-source platform to manage audio- and

video lectures, Opencast Matterhorn, at least re-

quires the setup of a server prior to working with

it, a task that is envisaged for a dedicated campus

administrator

. Evidently, the complexity of these

programs is due to the fact that editing and pub-

lishing videos on a professional level is complex.

We would like to keep all this to a minimum as it

might scare off the user. To specify our proposal, a

possible implementation is drafted below.

2.1 Possible Realization

For the case of simplicity, we will restrict this de-

scription to the most popular web site to publish

videos, YouTube. So let's assume a simple camcord-

er recording of a lecture has been uploaded to this

platform. The first task would be to add simple

comments to certain moments in the video once it is

recorded, and to make these annotations available in

a way that they can be searched and serve as a sim-

ple table of contents.

2.1.1 Subsequent Annotations

YouTube already offers the possibility to add com-

http://opencast.org/matterhorn/feature-tour

High-vs.Low-qualityVideoLectures-Don'tWorry,JustPutthemOnline

205

ments in the video that then appear at the defined

point of time inside the video once the video is

played back. However, for using them as an index, it

would be desirable to have access to them as a per-

manent text outside of the video, which also persists

once the video has stopped playing. Apparently, the

added notes are stored as an XML file by YouTube

that can be downloaded and added to other clips

. In

a similar way, this XML-file could be parsed and the

notes including their time stamp (with regards to

video clip time) extracted as text.

2.1.2 Index / TOC

The aforementioned XML comments are then used

on a new web page with the video embedded, where

all available annotations from the XML-file are

listed in the order of their appearance in the video,

desirably with the possibility to jump to the moment

in the video by clicking on the corresponding time

stamp. To illustrate the layout, we refer to the ap-

pearance of comments on SoundCloud

(see Figure

1), a popular web site to listen to uploaded music,

predominately DJ sets. The search function of the

web browser allows finding keywords in the com-

ments.

2.1.3 Review

It is likely that, if done anonymously, not all annota-

tions made by users are on the same level of appro-

priateness. Therefore, a simple rating system would

be helpful to indicate valuable annotations. These

systems are widely available as open source soft-

ware (e.g. MooTools MooStarRating

, for a discus-

sion of various rating interfaces see (Nobarany et al.,

2012), and comments with low rating can later be

filtered out or deleted at all.

2.1.4 Real-time Annotations

So far, the described elements were all adopted from

other sites that present user-generated content. How-

ever, one big difference to these sites is that for

lecture recordings a large group of later users were

already present during the time of recording. Thus it

might be useful to offer them a way to start annotat-

ing in real time, i.e. while sitting in the lecture to

facilitate blended learning, the desired combination

http://stefansundin.com/stuff/youtube/

youtube-copy-annotations.html

http://www.soundcloud.com

http://mootools.net/forge/p/moostarrating

of face-to-face and electronic lectures (Wieling &

Hofman, 2010). Here, our proposal is a simple app

that synchronizes with the first slide via a QR code

(containing the title of the lecture or the later video

file name) and then offers a GUI to immediately

mark critical moments and stores them in the same

XML format as the subsequent annotations so that

they can later be loaded together with the actual

video file. As annotation should not distract too

much from attending, few, easy-to-reach functionali-

ties would be desired (Schleicher, Sahami, Rohs,

Kratz, & Schmidt, 2011). An exemplary GUI is

depicted in figure 2.

The available tags or markers are limited to four

types, each represented with an icon: indicating

moments where the listener did not completely un-

derstand what was explained, moments that ap-

peared important to listener or included a good ex-

ample, and finally moments where the lecturer

pointed out that the current statement might be rele-

vant for the exam. This one-click-tagging might

reduce cognitive overload during the lecture as it

reduces the need to write down extensive notes and

already facilitate later processing of the lecture re-

cording (Mayer & Moreno, 2003).

The individual real-time annotations can be up-

loaded to the joint web page serving as a starting

point for subsequent review by the individual stu-

dent, while at the same time serving as non-

personalized clusters of short tags to see where other

students struggled or noted important points. A simi-

lar idea is pursued by myTU

, an app for the Tech-

nical University Bergakademie Freiberg, however,

their emphasis is on providing real-time feedback to

lecturer in order to slow down the pace, not on sub-

sequent review.

We are aware that several aspects of the pro-

posed system are not completely specified on a

technical level, and others may be disputable. For

example the advantage of just having one common,

anonymized set of annotations online and thus need-

ing no additional user management for administrat-

ing sets of comments comes with the disadvantage

that some users may not want to share their personal

annotations, or cannot be contacted individually.

Here, modifications are easily conceivable. The

main purpose of the above given outline is to pro-

vide the reader with an idea of the system we have in

mind when we now describe the non-technical re-

quirements we see for it to work.

http://mytu.tu-freiberg.de

CSEDU2013-5thInternationalConferenceonComputerSupportedEducation

206

Figure 1: Layout of SoundCloud, where user comments to an audio track are listed below including the time in the track

they refer to ('ID' in the comments refers to 'track ID', i.e. the specific title of the piece of music playing at that time).

Figure 2: Exemplary GUI of an app that allows for basic

annotations during the lecture, containing only four types

of labels/tags. Lecture time and title is given in the header

of the GUI.

3 NON-TECHNICAL

REQUIREMENTS

There are certain non-technical requirements to

establish a comparable system, the first surely being

a change of mind to move away from offering pol-

ished videos under centralized supervision, and

rather go for uncoordinated low-fidelity versions, as

it is the case with written material: if professionally

edited video material corresponds to published text-

books, the proposal made here is to go for the video

equivalent of manuscripts. As pointed out above, we

did not include any video editing prior to publishing

it online, because we have the impression that this

constitutes a major obstacle to most lecturers inter-

ested in offering video material.

The time lecturers save editing the footage can

be invested to supervise the correctness of the sub-

sequent annotations addressing the content of the

lectures, their actual area of expertise, rather than

urging them to become semi-professional video

editors. There are various examples of amateurish

recorded lectures online (e.g. YouTube), which still

convey valuable information to the viewer.

High-vs.Low-qualityVideoLectures-Don'tWorry,JustPutthemOnline

207

The second major obstacle we notice is a general

uncertainty regarding legal aspects of offering re-

cordings that may include pictures or other copy-

righted material. This uncertainty may differ be-

tween countries (c.f. Deimann & Bastiaens, 2010)

for a discussion for German institutions), as copy-

right laws may include a fair use

doctrine like in the

US. Of course, publishing lectures as proposed here

requires a certain ambiguity tolerance to reside in a

legal grey zone, but we are not aware that this dis-

cussion was that prevalent for the master template of

a manuscript residing in the library for every student

to make a copy of it. These scripts certainly con-

tained copyright-protected material. Here, apparent-

ly no one cared, probably because the library was

not that easily accessible as content in the internet.

Offering the videos only within the intranet of a

university or their online learning management sys-

tem (e.g. Moodle

) might be a compromise to estab-

lish similar conditions for digital material. However,

we rather think the main reason for this previous

indifference was the implicit agreement amongst all

involved parties that the provisory copy cannot

compete with the high-quality original. In a similar

vein, the sensitivity towards copyright issues might

be attenuated in the right holder if the video depict-

ing protected items is of obviously lower quality

than the officially published version, and not a loss-

less copy. Our intention is not to dry out commer-

cially produced and distributed lecture material, but

to complement it.

The availability of manuscripts did not keep au-

thors off from publishing textbooks, in some cases

the previously published script was offered as a beta

version of the actual book. The ratings collected via

'informal' videos may help to decide which lecture

should be edited and released, then in agreement

with the publishers whose material is involved.

The third requirement is the willingness to

switch from complete control over all content in-

cluding annotations to 'moderated' control by stu-

dents. In our opinion, this is the least difficult part

because considering student-generated material to

complement teaching is quite common at our univer-

sities, and the experiences have been clearly positive

(e.g. 'informal' solutions published by a student

being declared the 'official' sample solutions later on

as experienced by one of the authors). To establish

low-fidelity online courses as additional material, it

might be necessary to provide student organizations

http://www.copyright.gov/fls/fl102.html

https://moodle.org/

with a couple of annotated lecture videos as some

kind of initial seed. Shifting processing partly to

students will encourage active learning instead of

passive consumption of information, which increas-

es both, learning outcome as well as satisfaction

(Zhang, Zhou, Briggs, & Nunamaker, 2006).

The internal discussion of the approach revealed

several research questions that arise from using low-

fidelity video material for educational purposes.

3.1 Open Questions & Research

Directions

The main questions are whether such low-fidelity

videos will first be accepted by the students and

lecturers, and to what extend it will actually support

the learning process.

We think that using platforms and interaction

concepts instructors and students are familiar with

from their daily internet browsing (watching

YouTube videos, rating content, and adding com-

ments) will be less time-consuming than getting

used to completely new tools. We are aware of the

impact technical quality of audiovisual material has

on the recipient (Möller, 2010)(Arndt, Antons,

Schleicher, Möller, & G., 2012), although the issue

might not be as important as reported in (Lauer,

Müller, & Trahasch, 2004) due to a general increase

in available bandwidth since then. The Opencast

Matterhorn app Matterhorn 2 go

for example offers

searching and watching video lectures on the mobile

phone. Nevertheless, the lower quality as compared

to MOOC clips will of course be obvious, and may

in some cases even lead to ambiguous or non-

understandable sections. The euphemistic reply

would be that this emphasizes the 'authentic' charac-

ter of the material like jittery mobile phone clips

presented in news shows, where the unedited nature

of the clips almost increases their credibility.

However, this may be too optimistic, so let's as-

sume that the quality impairments simply prevent

understanding of certain sections. Here, a look in the

other available course material may be necessary,

probably based on the recommendations of fellow

students. The fact that processing a lecture cannot be

achieved without supplementary material might also

help to attenuate one objection we repeatedly heard

from lecturers, namely the worry that offering online

versions will discourage students to attend the class-

room. The more obvious it is that working with the

video material already starts in the lecture (by anno-

http://vm193.rz.uni-osnabrueck.de/matterhorn2go

CSEDU2013-5thInternationalConferenceonComputerSupportedEducation

208

tating it in real time) and that attending it will have

benefits in terms of acoustic and visual quality (as

compared to the low-fidelity video), the less an atti-

tude of 'I can attend it later/at home' will arise.

4 SUMMARY & CONCLUSIONS

In this position paper, we outlined a simple way to

offer video recording of lectures with low technical

quality to students and enable them to use this as

supplementary learning material. Unlike most avail-

able systems, the approach aims at utilizing existing

platforms and interaction paradigms as much as

possible, namely the possibility to watch videos

online via e.g. YouTube, add comments, and rate

those comments. Instead of editing the videos exten-

sively prior to uploading, the idea is that the main

focus should be on content-related annotating, which

can to a large extend be achieved by the students. To

facilitate this, they should be enabled to already start

with annotating while attending the lecture.

Shortcomings due to limited annotation func-

tionalities or arguably low technical quality of the

video footage are acknowledged and accounted for

by explicitly stating that the videos are just an addi-

tional teaching supplement without the intention to

replace other material or even lecture attendance.

This proposal is based on our experience that the

attempt to compete with platforms that offer profes-

sionally produced video lectures might fail without

providing substantial additional resources regarding

technical as well as legal expertise. At the same

time, the imbalance in resources has been dealt with

for a long time in other areas of teaching at public

universities both by students and lecturers alike, who

usually compensate for it by individually providing

material with low technical quality and increased

participation of the student body. We tried to show

how the same principle might be applied to video

lectures. The intention is to encourage all involved

parties, lecturers and students as well as experts on

e-learning to further develop this idea.

ACKNOWLEDGEMENTS

We would like to thank the Zentraleinrichtung Wis-

senschaftliche Weiterbildung und Kooperation

(ZEWK) TU Berlin, whose courses and support

motivated this manuscript.

REFERENCES

Arndt, S., Antons, J. N., Schleicher, R., Möller, S., & G.,

C. (2012). Perception of Low-Quality Video Analyzed

by Means of Electroencephalography. 4th Internation-

al Workshop on Quality of Multimedia Experience. Ju-

ly 5-7, Yarra Valley, Australia.

Copley, J. (2007). Audio and video podcasts of lectures

for campusbased students: production and evaluation

of student use. Innovations in Education and Teaching

International, 44(4), 387–399.

Deimann, M., & Bastiaens, T. (2010). Potenziale und

Hemmnisse freier digitaler Bildungsressourcen - eine

Delphi-Studie. Zeitschrift für E-Learning, 5(3), 7–18.

Giles, J. (2005). Special Report Internet encyclopaedias go

head to head. Nature, 438, 900–901.

Lauer, T., Müller, R., & Trahasch, S. (2004). Learning

with Lecture Recordings : Key Issues for End-Users

Rainer Müller. Proceedings of the IEEE International

Conference on Advanced Learning Technologies

(ICALT’04).

Mayer, R. E., & Moreno, R. (2003). Nine Ways to Reduce

Cognitive Load in Multimedia Learning. Educational

Psychologist, 38(1), 43–52.

Möller, S. (2010). Quality Engineering. Qualität

kommunikationstechnischer Systeme. Berlin: Springer.

Nobarany, S., Oram, L., Rajendran, V. K., Chen, C.,

Mcgrenere, J., & Munzner, T. (2012). The Design

Space of Opinion Measurement Interfaces : Exploring

Recall Support for Rating and Ranking. CHI 2012 (pp.

2035–2044). May 5–10, 2012, Austin, Texas, USA.

Schleicher, R., Sahami, A., Rohs, M., Kratz, S., &

Schmidt, A. (2011). WorldCupinion: Experiences with

an Android App for Real-Time Opinion Sharing dur-

ing Soccer World Cup Games. International Journal

of Mobile Human Computer Interaction, 3(4), 18–35.

Wieling, M. B., & Hofman, W. H. A. (2010). The impact

of online video lecture recordings and automated

feedback on student performance. Computers & Edu-

cation, 54(4), 992–998.

Zhang, D., Zhou, L., Briggs, R. O., & Nunamaker, J. F.

(2006). Instructional video in e-learning: Assessing the

impact of interactive video on learning effectiveness.

Information & Management, 43(1), 15–27.

High-vs.Low-qualityVideoLectures-Don'tWorry,JustPutthemOnline

209