Investing in Ephemeral Virtual Worlds
An Educational Perspective
Athanasios Christopoulos and Marc Conrad
Department of Computer Science & Technology, University of Bedfordshire, Park Square, Luton, U.K.
Keywords: Virtual World, Virtual Reality, Virtual Learning, Second Life, Opensim, Persistence.
Abstract: The increased demand for the use of virtual worlds in higher education has led many educators and
researchers in in-depth analysis and evaluation of a number of different virtual environments, aiming to
highlight their potentials. Until recently, Second Life was one of the most widely used virtual worlds for
educational purposes. However, the decision of Linden Lab to stop offering the educational discount, the
rumours around its future and the emergence of a novel technology called OpenSim challenged institutions’
decisions to keep using Second Life. In a try to identify the way institutions make their decision to use a
virtual world, 34 interviews have been conducted with university educators. The results of this study reveal
that both the cost and the persistence of a virtual world play an important role on this decision. However,
there are still some unique benefits offered by each world affecting to a great extent the educators’ decision.
We conclude the paper by advocating the use of a cross-institutional hypergrid.
While much of research on virtual worlds (VWs),
and in particular Second Life (SL) – the possibly
most prominent of these – has been performed about
projects within the VW itself (Bredl et al., 2012;
Childs, 2010; Miller et al., 2010; Vosinakis et al.,
2011) it would be naive to restrict oneself only to the
“inside” of these worlds and ignore the “real world”
environment in which these exist. The relevance of
this aspect has been already highlighted by Shukla
and Conrad (2011) who identify such experience
external to the VW via the notion of “broad
environment” and “direct environment”. Specifically
for the educational use of VWs the concurrent
consideration of both an “intrinsic” and “extrinsic”
view led to the development of an evaluation
framework (Conrad, 2011) that, in further
discriminating into an “individual” and a “world”
aspect identifies the four dimensions: cost,
persistence, context and immersion.
Since the authors have already examined the
“intrinsic” perceptions of the use of VWs in
previous works (Christopoulos and Conrad, 2012;
Christopoulos and Conrad, 2013), this paper aims to
enlighten its extrinsic view of persistence, focusing
on three specific VW paradigms: (i) SL, (ii) non-
isolated VWs based on the OpenSim (OS)
technology and hosted by Dedicated Providers
(OSDP), and (iii) isolated and closed VWs based on
the OS technology and hosted Internally (OSIH). At
this point the fact that OSIH can also be open and
interconnected through hypergriding is essential to
be pointed out. However, this case is not in depth
analysed in the following sections.
In this paper first we provide a short and
summative account of our results concerning
Immersion and Context in so far as they are relevant
for the further discussion on the extrinsic dimension
of Cost. The main sections on Persistence then
contrasts the view found in the literature (Section 3)
with educators’ opinions on these themes derived
from our interviews (Section 4) and analysed via
Grounded Theory (Strauss and Corbin, 1998). The
findings of this paper (Section 5) are based on data
collected and analysed within the wider context of a
Masters by Research thesis of one of the authors
(Christopoulos, 2013). We conclude the paper by
highlighting the authors’ position on how to move
Many attempts have been made to evaluate the
context of SL and OS based VW’s (Diener, 2009;
Christopoulos A. and Conrad M..
Investing in Ephemeral Virtual Worlds - An Educational Perspective.
DOI: 10.5220/0004957801180123
In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 118-123
ISBN: 978-989-758-022-2
2014 SCITEPRESS (Science and Technology Publications, Lda.)
Miller et al., 2010; Vilela et al., 2010). In one of our
previous works (Christopoulos and Conrad, 2013)
we comparatively examined these and concluded
that both SL and OS have many positive and
negative features in common as far as their contexts
are concerned, but at the same time, one differs from
the other, each having its own separate positive and
negative characteristics. However, the negative
elements of these worlds are not powerful enough to
discourage academics from exploiting them in
education. It became apparent that both SL and OS
can cover various needs that are difficult to be
covered or may not be covered effectively through
the use of the educational tools of the physical
world. As a result, educators consider each one of
them suitable for different types of educational
In addition to context, several researches have
been pursued to quantify immersion (Bredl et al.,
2012; Childs, 2010; Vosinakis et al., 2011). Indeed,
VWs are not immersive by definition (Christopoulos
and Conrad, 2012). Taking into account the features
of SL, it seems to be coming first in preference,
however slight it may be, over OS. These two VWs
were judged by educators as almost equivalent in
developing a sense of presence to students, but the
broader and richer network of interactions that exist
in SL gave it the lead. Second in line come the
OSDPs, and last of all come the OSIHs.
Unlike context and immersion, very limited
studies have been conducted regarding the cost and
the persistence of such VWs. In the following two
sections we therefore attempt to fill this gap and
establish how educators view these two extrinsic
aspects of virtual worlds. The findings are based on
semi-structured interviews that are contextualized
within the available literature.
Undoubtedly, the future of a VW and its persistence
over time cannot be predicted with certainty.
Nevertheless, the possibility of a VW stop operating
is certainly not a pleasant prospect, considering the
fact that educators and universities are based on it
for the implementation of successful projects,
investing time, effort and money on it.
Up until December 2010 Linden Lab was
offering a 50% discount to non-profit and
educational institutions for the acquisition and
maintenance of land, a fact that encouraged the
educational community to engage in SL. As from
January 2011 that discount stopped being offered
and that caused great inconvenience to the
universities maintaining their virtual land in SL,
since the cost became unbearable, and great
displeasure to many educators using SL
(Christopoulos, 2013). Even though a new discount
came to replace the previous one, very few
universities were able to be benefited from it
(Harrison, 2010).As a result, some universities
stopped using SL, some moved to other, cheaper
VWs such as OS, while others opted to coexist in a
shared piece of virtual land (Christopoulos 2013).
The universities, however, were not the only
ones that left their spaces in SL. Even the private
estates, the fees of which are the main source of
income from SL for Linden Lab, decreased
considerably during the previous years (Au, 2012).
This obviously implies an income reduction for
Linden Lab, which, according to estimates, will have
to face serious economic problems (Llewelyn, 2012;
Au, 2012), if this issue is not addressed to soon.
The “workspace sharing” practice of many
universities in SL that aim to reduce the cost of
using the world, without, however, losing the
multiple benefits it offers, is not just a practice
which only universities follow. It is a general trend
of many individuals, businesses and educational
institutions to opt to share a common virtual space,
as well as its fees (Llewelyn, 2012).
Thus, the future of SL looks uncertain. Even
more so, given the situation in SL and Linden Lab’s
attitude towards educational institutions, predictions
like this of Rogate (2012) should not be taken
lightly: “SL as a product for educators is actually
dead, unless something dramatically changes with
the strategy of Linden Labs—which always remains
On the other hand, although OS technology had
several glitches and instability issues at its first
steps, it has become considerably stable over time.
The qualitative improvement of OS, in conjunction
with its low economic cost of use makes it attract
new users, whereas SL keeps losing them (Gracious,
2012). Therefore, OS has lately become a very
worthy competitor of SL, since it has evolved into a
VW almost as functional as SL (Reeve, 2012).
Moreover, the features of keeping backups of the
world and hypergriding, i.e. the teleportation from
one grid to another, are exclusive advantages of OS,
which enhance its persistence over time. OS,
essentially, is not a VW, but a technology open to
anyone who wishes to develop a VW. This world
can be backed up along with all its content at any
time and reused whenever necessary, by anyone
holding the backup files (Miller et al., 2010). This
means that each VW persists for as long as its
backup files exist and independently of the operation
of the others (Fishwick, 2009).
For the needs of this study 34 educators were
interviewed. The interviews took place through
skype or within SL or OS. The questions asked were
the following:
1. If Second Life were to close many
educational institutions would be left
“homeless”.Have you taken this issue into
account?What is your opinion?
2. Are you concerned about Second Life’s
closure? Does this possibility affect your decision to
use Second Life?
3. If eventually Second Life terminates, will you
attempt to replace it with another virtual
environment? Can you, please, name this alternative
4. OpenSim is a new technology used for the
development of virtual environments. How stable do
you expect this technology to be?
5. A major advantage of OpenSim technology is
the opportunity given to its users to keep backups.
How useful do you consider it?
6. OpenSim grids have the potential of “hyper
gridding” (teleportation of avatars and items from
one grid to another). How useful do you consider
this fact?
7. OpenSim technology faces stiff competition
from other well established virtual worlds such as
Second Life. Thus, do you consider that this
competition will affect negatively its persistence?
4.1 The Future of Education in SL
The spreading rumours about the future of SL and its
potential closure raise interviewees’ concerns about
the future of their projects running in-world. They
state that they worry less about their educational
projects, which anyway may find shelter in other
VWs, but more for research projects on SL which
cannot be carried out into another world. They are
also concerned about the resources spent for the
needs of these projects that will be lost if SL
terminates. Furthermore, the concern that, if SL
terminates, its community and the thematic groups
will be lost, was also expressed. Then, these groups
will no longer be able to organize in-world
professional events, which are considered to be very
useful and constructive for professionals in any
industry operating in SL.
However, some educators indicated that they will
continue using SL for educational purposes. Some of
them stated that their stay in SL will last until the
expiration of their contract with Linden Lab, or until
their projects stop being funded. Others stated that
they are not intending to stop using SL, either
because it is money, time and effort consuming to
create their workspace from scratch within another
VW, or because they have not yet found another
VW as worthy as to replace SL. Contrary to them,
some other educators stated that they are intending
to replace or have already replaced SL with another
VW or technology such as OS, OpenWonderland,
Unity3D, Blue Mars and Active Worlds. Finally, the
view that if Linden Labs keeps following their
strategy of not supporting education, educators will
opt to continue without the use of VWs, was
expressed, as well.
Figure 1: Educators’ views for the potential closure of
Second Life.
Nonetheless, at this point it should be noted that
it is rather the high cost of using SL that affected
educators in making these decisions.
4.2 The Educational Potentials in OS
The participants stressed that the OS technology has
been significantly improved in the recent years. The
competition with other well-established VWs had a
positive effect on improving its stability, its
reliability, and its interoperability. Nevertheless, this
competition has a negative effect on its evolution
too. The OS worlds have online communities
narrower than SL, and given that it is open-source
software, its upgrading may be slow, since it
depends on the involvement of its own community,
rather than a company’s. It was also suggested that
competition has no impact on the evolution of OS,
as it provides services very different from other
During the interviews, the importance of backups
for the persistence of OS worlds was also
highlighted. More precisely, it was stated that it is
very useful for cases where the workspace retrieval
is considered necessary. In these cases, the educator
keeps a backup of the world when it is in a desired
state, introduces students to the world so as to carry
out their activities, and then uses the backup file, in
order to “regularize” the world and bring it back to
its previous state. The same technique can be used in
cases where technical issues that affect the smooth
conduct of activities arise. Other educators reported
that they use the backups in order to transfer to other
servers and share with other educators objects, tools
or even their entire workspace.
Figure 2: Educators' views about the future of OpenSim.
Finally, the interviewees’ view of the
hypergriding potential was very positive, even
though some of them seemed never to have used it.
It is thought that hypergriding contributes to the
overcoming of the isolation that is likely to occur in
OSIHs or in worlds with a very small community.
Consequently, interuniversity communication and
collaboration can be achieved. Students have the
opportunity to see the creations of others, and this
may be an inspiration for them, making that way the
lesson more interesting. Beyond these, it was
reported that migration options are given to
universities, and therefore a training group or the
entire university can carry out their activities in
another OS world, any time and for any reason.
4.3 Discussion
It seems, overall, that whether and for how long SL
will persist depends purely on Linden Lab.
Therefore, educators have to accept the decisions of
the enterprise and then decide about their future
plans. On the contrary, the persistence of the OS
worlds depends on the aims and plans of the
In conclusion, each VW has different advantages in
terms of utility costs, while each can cover different
needs. Specifically, SL is the ideal choice for those
educators seeking a more time and effort effective
option, or for educators who cannot devote much of
their time for the preparation of the educational
activities, or even for those who do not have enough
knowledge on building and scripting to create the
virtual space in accordance with their teaching
needs. This is actually the case because of the
existing wide marketplace in SL where educators
can buy various items ready for use. Furthermore,
there are many builders and scripters working in SL,
able to offer their services to anyone upon payment.
Additionally, educators often resort to the solution
of “workspace sharing” or “items sharing” within
the VWs, and are greatly facilitated by the fact that
SL has a very wide community. Nonetheless,
universities that face a decision to use SL for
educational purposes should be prepared to pay high
enough monthly fees for the rental of the virtual
land, and they should also be aware that additional
charges apply on the uploading of files and the use
of more primitives than those granted along with the
purchase of the land.
Exactly the opposite applies in the case of
OSIHs. This choice is ideal for the universities
which seek the most cost-effective option, but a
basic prerequisite for this is the existence of proper
infrastructure and qualified personnel which is able
to spend time and effort to set up, maintain and
ensure the server’s proper operation, and which is
also able to build and script for the creation of the
needed in-world facilities for the educational
activities. Therefore, even though the economic cost
of this option for the university is minimal, it cannot
be considered as time and effort effective.
Nevertheless, the required effort by the university
staff can be significantly reduced, if these actions are
assigned to students as part of their internship. It is
also worth mentioning that the OSIHs are an ideal
choice in cases where the main purpose of the in-
world sessions is to allow students to build and
script. In these cases, on one hand both the effort
and time which has to be devoted by the university
staff for the preparation of the in-world spaces is
reduced, while on the other hand students can freely
“play around” with the space, since there are no
restrictions similar to those that occur in SL
regarding the amount of objects and scripts.
Moreover, the “workspace sharing” and “objects
sharing” mentioned in SL also apply in the OSIHs,
even in a different way than that of SL, since the
backup files which are parts of the OS technology
can be run in any OS server. Thus, the workspace
created by a university can be copied and given for
use to another university. Obviously, this is a highly
money, time and effort effective practice.
Finally, choosing an OSDP is the “middle
ground” between SL and OSIHs. Even though the
university has to pay monthly fees for the provider’s
services, the land fees of dedicated providers are
considerably lower compared to the fees charged by
Linden Lab for SL. Furthermore, unlike Linden Lab,
the providers offer, from the beginning, the
maximum number of primitives which can be used
in each piece of land, while their cost is included in
the monthly land fees. Therefore, similarly to the
OSIHs, this option is also very suitable for building
and scripting activities.
An apparent disadvantage of OSDPs, compared
to SL, is the lack of a marketplace which helps
educators to save effort and time. However, the
“workspace and objects sharing” practice applies in
OSDPs as well, as described both in the case of SL,
i.e. the temporary use of the in-world facilities from
other universities, and in the case of OSIHs, i.e. the
backup files exchanging. Besides, there are no
additional charges for files uploading.
Therefore, it seems that the case of OSDPs
gathers many of the advantages –in terms of cost–
that the other two solutions have, but it also shares
few of their disadvantages, as well. It seems to have
the lowest cost in terms of money, time and effort, if
seen in total, but which one is the most “cost-
effective” choice clearly depends on the needs and
capabilities of each university.
Regarding persistence, the educational
community appears to be very disappointed with the
overall current situation in SL and some educators
have already dropped out of it, choosing to use other
VWs, even though it keeps operating. Therefore, the
question is “for how long the educational
community will be present within SL?”, or, in other
words, “for how long will SL worth being used as an
educational tool?”, rather than “for how long will SL
persist?”. Regarding these questions, the views of
the educational community members vary. A part of
them still sees SL as a very convenient educational
tool. Some consider it appropriate under certain
circumstances, while others believe that education in
SL has no future.
The advantage of OS regarding its persistence is
that it is a technology for the creation of VWs,
which is not supported by an enterprise, but by the
open source community. Therefore, the persistence
of OS depends on the choices of individual
educators or universities. However, even if an
educator opts to use the services of an OS provider,
there is always the potential of transferring the
activities to the server of another provider, or a
private server, if it is deemed necessary, using
backups and the hypergrid architecture.
This leads to the conclusion that educators who
wish to obtain the widest possible control of the
persistence of their world should opt to use either an
OSIH or an OSDP. In cases where the long-term
persistence of the world is not a major concern for
educators, they may use SL, if they think that its
benefits are essential for their projects.
In summary it can be seen that the extrinsic view
is a matter of concern of educators using SL and OS.
Given the SL issues concerning a possible closure
(persistence) or increased costs a move towards OS
based solutions is tempting and in many cases
indeed has happened. Our findings seem to suggest
that – unless Linden Lab positions itself clearly
concerning their long-term SL strategy, in particular
towards educators – this shift towards OS will
continue as educators are more and more willing to
accept a loss of the intrinsic dimensions of context
and (possibly) immersion in order to get reassurance
and a perspective concerning the extrinsic
dimensions cost and persistence.
Following from the above the University of
Bedfordshire is hosting their activities now on their
own OSIH (after extensive experience with SL and
OSDPs in previous years). The virtual world is used
to teach LSL as an event driven programming
language and to foster activities of students learning
Project Management.
The control provided within an OSIH is an
advantage not only in providing an environment for
students but also to analyse their activities as part of
ongoing research.
Nevertheless to create the ‘look and feel’ of a
true virtual ‘world’ we are now actively seeking
collaboration to join educational virtual worlds as
part of a hypergrid. The technology is readily
available and it is our belief that seamless utilisation
of virtual worlds across educational institutions will
create a persistent and cost efficient virtual
environment in which educational activities can be
made available.
Au, W. J. (2012). “SL Forecast to Lose 10% of World’s
Private Sims This Year”. New World Notes [blog],
June 6, 2012, Available at:
/2012/06/second-life-losing-10-private-land.html [last
accessed: August 26, 2012].
Bredl, K., Groß, A., Hünniger, J. and Fleischer, J.
(2012).“The Avatar as a Knowledge Worker? How
Immersive 3D Virtual Environments may Foster
Knowledge Acquisition”. In: The Electronic Journal
of Knowledge Management, Volume 10 Issue 1 (pp.
Childs, M. (2010).“Learners’ Experience of Presence in
Virtual Worlds”.Ph.D. Thesis, University of Warwick,
Coventry, UK.
Christopoulos, A. (2013). “Higher Education in Virtual
Worlds: The Use of Second Life and OpenSim for
Educational Practices”. M.Sc. Thesis, University of
Bedfordshire, Luton, UK. Available from:
Christopoulos, A., & Conrad, M. (2012).“Views of
Educators on Immersion in Virtual worlds from
Second Life to OpenSim”. In: M Gardner, F Garnier&
CD Kloos (eds), Proceedings of the 2nd European
Immersive Education Summit: EiED 2012. E-iED,
Universidad Carlos III de Madrid, Departamento de
IngenierÃaTelemÃitica, Madrid, Spain, pp. 48-59, 2nd
European Immersive Education Summit, Paris,
France, 26-27 November, 2012.
Christopoulos, A. & Conrad, M. (2013).“Maintaining
Context in a Changing (Virtual) World – Educators’
Perspectives for OpenSim and Second Life”.In: 5
International Conference on Computer Supported
Education (CSEDU 2013). 6-8 May, 2013, Aachen,
Conrad, M. (2011). “Leaving the Lindens: Teaching in
Virtual Worlds of Other Providers”. Proceedings of
the Researching Learning in Immersive Virtual
Environments Conference (ReLIVE11), (pp. 28–36).
21-22 September, 2011, Milton Keynes, UK.
Diener, S. (2009). “This Will Change Everything. Virtual
Worlds in Education”. In: Same places, different
spaces Proceedings of ascilite Auckland 2009 - 26th
Annual ascilite International Conference, (pp. 1-14).
6-9 December 2009, Auckland, New Zealand:
Auckland University of Technology, and Australasian
Society for Computers in Learning in Tertiary
Education (ascilite).
Fishwick, P. A. (2009), “An Introduction to
OpenSimulator and Virtual Environment agent-based
M&S Applications”. In: Dunkin, Ann, Ingalls, Ricki
G., Yücesan, Enver, Rossetti, Manuel D., Hill, Ray
and Johansson, Björn (Eds.) Proceedings of the 2009
Winter Simulation Conference - WSC 2009, (pp. 177-
183). 13-16 December, 2009, Austin, TX, USA.
Harrison, D. (2010). “Linden Lab To End Second Life
Educational Discounts”. THE Journal. 1105 Media.,
Wed November 3rd, 2010. Available at:
to-end-second-life-educational-discounts.aspx [last
accessed: 26th Aug 2012].
Miller, A., Allison, C., McCaffery, J., Sturgeon, T., Nicoll,
J., Getchell, K., Perera, G. I. U. S. and Oliver, I.
(2010).“Virtual worlds for Computer Science
Education”.Proceedings of the 11
Annual Conference
of the Higher Education Academy Subject Centre for
Information and Computer Sciences, (pp. 239–244).
24-26 August, 2010, Durham, UK. HEA ICS.
Reeve, J. (2012).“Making OpenSim safe for students”.
Hypergrid Business, [blog], March 20, 2012.
Available at:
2012/03/making-opensim-safe-for-students/ [last
accessed: August 26, 2012].
Rogate, J. (2012). “Prof: Schools moving to OpenSim
should pay for hosting”. Hypergrid Business, [blog],
January 2, 2012. Available at: http://
moving-to-opensim-should-pay-for-hosting/ [last
accessed: August 26, 2012].
Shukla, M., Conrad, M. (2011). “Second Life is not an
Island”. Proceedings of the IADIS International
Conference on e-Society.10-13 March 2011, Avila,
Strauss, A., & Corbin, J. (1998).Basics of Qualitative
Research: Techniques and Procedures for Developing
Grounded Theory. Second Edition, Sage Publications,
Thousand Oaks, California, USA. ISBN: 0-8039-
Llewelyn, G. (2012). “Second Life: Towards
Consolidation and Cloud Computing”., [blog], July 12, 2012, Available
towards-consolidation-and-cloud-computing/ [last
accessed: August 26, 2012].
Gracious, G. (2012). “Linden Labs At War With
Opensim”.Metaverse Traveller, [blog], August 7,
2012. Available at: http://metaverse-
with-opensim.html [last accessed: August 26, 2012].
Vosinakis, S., Koutsabasis, P. and Zaharias, P. (2011).
“An Exploratory Study of Problem-Based Learning in
Virtual Worlds”.Proceedings of the 3rd International
Conference in Games and Virtual Worlds for Serious
Applications (VS-Games 2011), (pp. 112 – 119). 4-6
May, 2011, Technical University of Athens (NTUA),
Athens, Greece.
Vilela, A., Cardoso, M., Martins, D., Santos, A., Moreira,
L., Paredes, H., Martins, P., Morgado, L.
(2010).“Privacy challenges and methods for virtual
classrooms in Second Life grid and opensimulator”.
In: IEEE Proceedings of the Second International
Conference in Games and Virtual Worlds for Serious
Applications (VS-Games), (pp. 167 – 174). 25-26
March, 2010.