Conference PaperPDF Available

Are there limits to growth in data traffic?: on time use, data generation and speed


Abstract and Figures

This discussion paper considers the nature of growth in data traffic across the Internet, as a basis for asking whether and how such growth might slow down or otherwise be limited. Over the last decade, data growth has been dramatic, and forecasts predict a similar ongoing pattern. Since this is associated with increasing electricity consumption, such a trend is significant to global efforts to reduce carbon emissions. In this paper, we selectively explore aspects of data growth that are linked to everyday practices and the way they draw upon and generate Internet data. We suggest that such growth does have some conceivable limits. However, the nature of 'Internet use' is changing and forms of growth are emerging that are more disconnected from human activity and time-use. This suggests that although there may well be limits, in principle, to some forms of growth, total data traffic seems likely to continue growing. This calls for careful attention to the nature of the trends involved, as a basis for intentionally building limits into this system before levels of Internet electricity demand becomes directly and more explicitly problematic.
Content may be subject to copyright.
Are there limits to growth in data traffic?:
On time use, data generation and speed
Mike Hazas, Janine Morley, Oliver Bates, Adrian Friday
Lancaster University
Bailrigg, LA1 4YW, United Kingdom
This discussion paper considers the nature of growth in data
traffic across the Internet, as a basis for asking whether and
how such growth might slow down or otherwise be limited.
Over the last decade, data growth has been dramatic, and
forecasts predict a similar ongoing pattern. Since this is
associated with increasing electricity consumption, such a
trend is significant to global efforts to reduce carbon emis-
sions. In this paper, we selectively explore aspects of data
growth that are linked to everyday practices and the way
they draw upon and generate Internet data. We suggest that
such growth does have some conceivable limits. However,
the nature of ‘Internet use’ is changing and forms of growth
are emerging that are more disconnected from human ac-
tivity and time-use. This suggests that although there may
well be limits, in principle, to some forms of growth, total
data traffic seems likely to continue growing. This calls for
careful attention to the nature of the trends involved, as a
basis for intentionally building limits into this system be-
fore levels of Internet electricity demand becomes directly
and more explicitly problematic.
CCS Concepts
Social and professional topics Sustainability; Human-
centered computing Interaction design;
Information infrastructures; social practice
In a recent article in Low-Tech Magazine, de Decker [5]
argues that “there are no limits to growth when it comes
to the Internet, except for the energy supply itself and so,
uniquely, “the energy use of the Internet can only stop grow-
ing when energy sources run out, unless we impose self-
chosen limits”. Other energy-demanding systems, he sug-
gests, encounter inherent constraints which limit their growth:
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for profit or commercial advantage and that copies bear this notice and the full cita-
tion on the first page. Copyrights for components of this work owned by others than
ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or re-
publish, to post on servers or to redistribute to lists, requires prior specific permission
and/or a fee. Request permissions from
LIMITS ’16, June 08-10, 2016, Irvine, CA, USA
2016 ACM. ISBN 978-1-4503-4260-5/16/06. . . $15.00
for example, the weight and size of cars can only increase
so much if they are still to remain compatible with infras-
tructures designed for them; infrastructures in which speed
limits are imposed for safety reasons. Similarly, he suggests
that some kind of ‘speed limit’ for the Internet is needed.
Putting to one side consideration of how such limits might
be achieved, it is intriguing to examine the claim that the
energy used by the Internet will continue to grow until the
availability of energy itself becomes problematic, that is, un-
less some other kind of checks or limits to growth are im-
posed first. This is a rather radical, fascinating and, in so
far as it is plausible, troubling claim.
Currently, the direct energy used to power the Internet
and to produce, transport and dispose of its components
(embodied energy) are relatively small contributors to global
emissions in comparison to, say, global transport [13]. Cur-
rent estimates suggest that operation of the Internet (power-
ing devices, networks and data centres) amounts to around
5% of global electricity use; yet this is growing faster (at 7%
per year) than total global electricity consumption (3% per
year) [19]. In other words, the Internet is consuming an in-
creasing portion of global electricity supply. In the context
of changing energy systems that include greater renewable
sources, and new forms of electricity demand such as electric
vehicles, the growing portion of global electricity required
to run the Internet may become increasingly significant in
efforts to balance supply and demand, reduce carbon emis-
sions and, as such, become potentially subject to energy-
related limits as de Decker suggests. Yet given the radically
distributed, and largely ‘invisible’ nature of this energy con-
sumption, how large a portion of global electricity could this
represent before such limits might be imposed? Some pre-
dictions suggest that production and use of information and
communication technologies might grow to around 20% of
global supply by 2030, or as much as 50% in a worst case
scenario [2].
In this short discussion paper, we consider trajectories of
growth in the energy that powers the Internet, not through
modelling, but through a more conceptual exploration of
some of the trends currently and historically associated with
such growth. Specifically, we ask whether or not these trends
may ultimately be limited in themselves, and thereby may
help to slow and limit growth in the future, at least in prin-
ciple. This is a complex question, so we frame and focus our
exploration in two ways.
First, we focus on volumes data traffic as an indicator of
operational energy use. This is by no means a direct or
simple relationship but in broad terms “the very substantial
gains from energy efficiency improvements have been more
than offset by increased consumption of services”[12, p. 583].
Thus, as flows of data traffic have increased so too has energy
use, albeit at a lesser rate.
Second, we focus on the link between the use and gener-
ation of data and activities in which people are more and
less directly involved. We explore the idea that, to date,
human time and attention has been related to the growth
in data traffic, in various ways. We ask whether this might
represent some limits to ongoing growth, and whether other
forms of growth are emerging and taking over.
We start by characterising the rate and composition of
data growth. We then consider aspects of this growth re-
lated to activities and time spent ‘online’, and ask whether
such trends will continue indefinitely or, at some point, rep-
resent some kind of check to growing data traffic. Whilst
growing levels of access to the Internet, a greater range of
‘online’ activities and services migrating and emerging, and
increases in time spent online continue to be relevant to the
growth in data, there is, in principle, a limit to these forms
of growth: the global population is finite (though clearly
growing), and the hours in the day available to each person
are also finite. We then turn to consider how the dynamics
of data growth are changing: the relationship to attention
and time-use is becoming less direct as data intensities of
certain services increase and as ‘background’ connectivity
and connected things become more prevalent.
Measures of actual Internet traffic volume and compo-
sition tend to be partial, limited to certain countries or
narrow periods of time. Based on a “representative cross-
section” of service providers, who volunteer to allow their
aggregate data to be reported, Sandvine [17] report that
the average monthly per-subscriber traffic volume on fixed
lines like broadband and fibre increased by about 50% in
North America, and 170% in Europe over 2013 and 2014
(Figure 1). According to Ofcom, the UK telecommunica-
tions regulator, UK home broadband data volumes also grew
markedly: monthly average traffic volume rose from 17 GB
in 2011, to 82 GB in 2015 [11].
Taking just one survey point, public aggregate input/output
data statistics published for January each year 2002–2016
from the Amsterdam Internet Exchange (Figure 2), we see
a clear pattern of year on year growth of between 20–140%
(20–40% each year in the last 5 years). Crudely, this data
demand has an equivalent direct energy cost as data is trans-
mitted and processed. Despite step changes in energy effi-
ciency as new technology is introduced, this could arguably
be offset by innovations in the marketplace such as increased
expectations around high-definition video (e.g. 1080p cam-
eras on smartphones, 3DTV and 4K video). Holistically, this
pattern of growth also causes increases in embodied energy
(emergy [13]) of the Internet as equipment needs to be re-
placed to cope with the continued increase in data demand,
and its associated demands for processing and storage.
For mobile access, data volumes are generally smaller, but
have consistently fast growth, doubling every few years ac-
cording to Sandvine (Figure 1). In fact, Ericsson [6] report
a 60% growth in global mobile cellular data in recent years,
and the forecasts are for 45% or more annual growth through
Figure 1: Growth in fixed and mobile traffic vol-
umes. Data source: Sandvine reports 2012–2014.
2021 [4, 6].
Several forms of growth are reflected in the overall in-
creases in mobile traffic volume: a) per-subscriber demand,
which is b) compounded by the increasing number of tablet
and smartphone handsets connected to mobile cellular data
services, which is c) related to, but not entirely explained
by, the proportion of the global population who have some
form of Internet access.
On the face of it, the latter form of growth has some lim-
its, even though these may be many years away: broad-
band services currently only reach around 30% of homes in
most developed countries, and only 43% of the world popu-
lation is using the Internet [8]. There are biophysical limits
to human perception and attention, which ostensibly could
place an upper bound on the fidelity of digital media we
deem ‘sufficient’. Some limits are indeed already visible: in
economically developed nations, mobile phones are reaching
saturation. For example, the number of mobile subscrip-
tions already exceeds the population in the US, Scandinavia
and Australia. And while by 2020 over 3.1 billion mobile
subscribers will have access to LTE and 4G communications
networks worldwide, this falls far short of world population.
Alongside questions of population and ‘saturation’, it is
important to consider what all this Internet traffic is for -
what activities it seems to be intertwined with. As Røpke
points out, “People are practitioners who indirectly, through
the performance of various practices, draw on resources” [15].
Figure 2: Monthly data from Amsterdam Internet
Exchange aggregated for each January 2002-2016.
Communications enabled devices increasingly augment ev-
eryday practices that previously would have not have re-
quired or even benefited from, Internet connectivity. In our
recent study of the communication impacts of applications
of mobile devices in everyday life [9], we found practices such
as exercise regimes to be augmented by social media appli-
cations; and migration of once offline or broadcast viewing
and listening of TV and radio to media rich communication
and streaming media platforms.
The personal ownership of these devices, coupled with the
convenience of accessing the Internet and their mobility, en-
able communication to fill and even expand small pockets of
‘dead time’. Along with increased multitasking supported
by digital technologies “enabled by the partial decoupling
of many practices from previous time and space constraints
through the use of ICT, contribute to a more densely packed
everyday life” [16, p. 356].
In other words, because Internet connectivity is being in-
tegrated into a huge variety of practices which may take
place throughout the day, time ‘spent online’ attending to
digital services is growing. Indeed, in 2005 in the UK an
average of 9.9 hours was spent online in a “typical week”
across home, work and elsewhere. By the end of 2014 this
average had grown to 20.5 hours per week [10, p. 28].
Yet, not all ‘online’ activities (those that require an In-
ternet connection) are equal. Firstly, significant differences
emerge between traffic types on domestic and mobile Inter-
net networks. Drawing on Sandvine, the category of “real-
time entertainment”, which is primarily video, accounts for a
large portion of total Internet traffic, at over a third of peak
period traffic. This is followed by web browsing, which might
support a wide variety of activities. Social networking is also
significant, particularly on mobile networks. Secondly, these
activities ‘take-up’ time in different ways, with some clearly
taking more time than others.
3.1 Watching
Some of the growth in time which is reportedly spent on-
line can be attributed to a daily increase in the number of
Internet users who watch TV or films online (10%–27% be-
tween 2007–2014) [10]. The number of Internet users who
watch short video content (i.e. video clips) has almost dou-
bled in the same amount of time, growing from 21% to 39%
Figure 3: Composition of peak period traffic in Eu-
rope in the second half of 2014. Redrawn from Sand-
in 2014.
Ericsson’s February 2016 report sums up why mobile video
is traffic is growing at such a rate, due to: larger screens en-
abling better quality streams; video content increasingly ap-
pearing as part of other applications (e.g. social networking,
news, advertising); growth of video streaming (50–70% of
video traffic for some mobile networks is YouTube); growth
in uptake of video on demand; changes in where and where
video is consumed and faster infrastructure.
3.2 Other online services
Whilst the number of people who are using online services
to consume media content is growing, it is worth noting
that the amount of time spent using these services is also
growing. In the US, the average time spent per day paying
attention to digital activities (including digital video, social
networks, digital radio, Facebook, Pandora) has increased
from 226–364 minutes per day between 2011–2015 [20]. The
time spent online watching digital video has grown from 39–
115 minutes, with growth also observed for social networks
(71–104 minutes), and digital radio (53–65 minutes) [20].
3.3 Concurrent uses of the Internet
In 2014 additional connected devices are being used to
go online compared to 2009, including Smart TVs, e-book
readers and wearable devices, with all technologies (com-
puters, smartphones, tablets, games consoles) other than
“portable media players” increasing in their use for going
online [10, Fig. 34]. It’s worth mentioning that multitask-
ing that takes place whilst watching TV is performed by
53% of UK adults [14, p. xi]. This multitasking is en-
couraged by “living room connected devices” (e.g. devices in
the living room that are connected to the Internet) seen to
“blur the line between passive and active entertainment”[14,
p. 4]. Whilst we are currently unaware of the overlap be-
tween watching on living room connected devices and pri-
mary viewing devices (e.g. Smart TVs, laptops, consoles) it
is worth acknowledging that living room connected devices
are likely to increase time spent online as they are predicted
to increase from 114 million to 267 million units shipped
worldwide by 2017 [14, p. 5].
So far we have talked about ‘attention-connected data de-
mand’, that is, exchanges in data that occur as the (more
or less) direct result of what people do by paying attention
to online services, reflecting the presence and significance
of media rich communication in everyday practices. As can
be appreciated from Figure 3, much of the data traffic dur-
ing peak times, currently seems to be associated with such
activities (e.g. entertainment, web browsing, social network-
ing). Although such traffic may continue to grow as a result
of increases in the data-intensity of these activities, such as
through higher definition video content or through new en-
tertainment services requiring complex forms of cloud com-
puting such as virtual reality and cloud gaming, there is
some limit to the time that is available to ‘invest’ in them,
and thus some ‘check’ to growth in these types of data de-
Yet not all traffic is so directly associated with attention-
linked services, and thus with patterns of time-use. One al-
ready automated form of demand on the Internet is through
software updates. These are currently about 6% of download
traffic, or perhaps up to 10% if computer game downloads
and updates are included (on marketplaces such as Steam,
the PlayStation Store, and Xbox Live) [17].
In addition, activities are at times accompanied by ‘unin-
tentional’ or ‘background’ data demand. In recent studies
of mobile device use, we found unexpectedly high levels of
communication between apps and the cloud when specific
applications were not in active use, for instance ensuring
that applications and operating system services were up to
date (900Mb or 5% of their overall traffic for one partici-
pant’s iPad), backup of application data and digital photos
to the cloud (2.25Gb/week or 71% of data demand cloud
syncing of photos and videos in one case). These levels
of data traffic, were both unobserved, uncoordinated, and
largely unmanaged by or even unmanageable for most of our
participants [3, ch. 6], [9]. While these studies are arguably
with small and isolated populations, the very unremarkable
nature of these findings coupled with the vast numbers of
similar devices in common use, suggests a very significant
data demand that is difficult to manage and limit at scale.
As the dominant paradigm for software development, tool
chains, and indeed the very business models that drive the
mobile eco-system push the design of software as thin clients
to powerful cloud backend services, we will doubtless witness
these kinds of data patterns repeated many-fold. Most sig-
nificant perhaps is the ongoing development of machine to
machine communication enabled by the so called ‘Internet
of Things’ (IoT). What about the data and energy impacts
of smart ‘things’, in homes, workplaces and civic infrastruc-
tures? This introduces another form of growth that is more
dissociated from the limits associated growth in direct forms
of Internet ‘use’ that have to date been so significant. Con-
necting ‘things’ or ‘machines’ to the Internet changes the
connection between what people do and exchanges of data.
This communication will occur transparently, without obser-
vation or interaction, and potentially without limit. At the
time of writing, the existing 6.4bn connected IoT devices is
only slightly less than world population (86%), but market
predictions suggest this will reach 21bn devices by 2020—
roughly three times world population estimates. Some pre-
dictions put machine to machine communication as 45% of
the whole Internet traffic by 2022 [7], with additional re-
liance on data upload [18] and processing in the cloud [1].
While IoT devices are presumably assumed to be low
communications footprint devices such as smart meters, or
smart home thermostats that communicate sporadic values
to backend services; others like self-driving cars, and re-
motely monitored wireless cameras and wearable medical
devices will be highly data intensive, and require match-
ing facilities in communications networks and data centres
to offer timely and responsive communication, computation
and storage. Some see IoT as a potentially ‘net negative’
contributor to energy demand due to the efficiency gains
made through better measurement and control. We merely
observe, that these new facilities have the potential to be
adopted into everyday practice, to raise expectations and
lower the burden of creating additional data and electricity
demand and have an initial and ongoing embodied energy
cost, by way of rebound effect—in exactly the same way
that smart devices, media and broadcast are already doing.
What seems clear, is that the impact of IoT is unknown or
unknowable at this point. Algorithms that create/consume
data can carry on doing so, regardless of who is paying at-
tention to them, and even less clear are the limits and brak-
ing functions that will ensure that such systems will operate
within sustainable limits.
This returns us to the question of whether there is an in-
herent limit to Internet traffic growth (other than energy)?
On the one hand, yes. Some of the reasons for current and
past growth will themselves be limited at some point in the
future; there is after all a finite, if growing, population on
the planet and only a certain number of devices that one
might explicitly use to access Internet services. Even if we
assume that the current access patterns in the most ‘con-
nected’ societies are replicated worldwide; the full panoply
of second screen, higher definition, and on-demand over 4G
and higher, there is plausibly a limit to the hours in the
day and content that can be actively engaged with. On the
other hand, ‘the data under the hood’ is growing unwatched
and unabated. The automated updates, cloud syncing, of-
floading of storage and computation to the cloud, that are
an increasing feature of the design of applications, and en-
demic to the tools and pervading technological culture that
is bringing these about.
Further, the Internet of Things is set to trigger a whirl-
wind of investment and connected infrastructure growth that
has the massive potential to grow operational electricity use
and emergy of the Internet. Despite sometimes question-
able benefits and motivations, the IoT is currently under
construction, in many different ways. This raises key ques-
tions as to whether, and what kinds, of limits there may
be to potentially self-generating cycles of data generation,
processing and circulation within such an Internet. If such
cycles are largely automated and operate at remove from
the time-limits associated with human activity then, poten-
tially, any ‘inherent’ limits to the growth in Internet traffic
will fall a long way short in the future.
Yet this Internet of Things is still in-the-making, and such
limits (or lack of them) are not yet ‘inherent’. Thus, as the
Internet continues to develop, de Decker’s proposal of some
kind of speed limit, that might be built into this system,
shaping its development for years to come, is an important
proposition to consider; especially in comparison to an alter-
native prospective of making retrospective reductions in In-
ternet traffic in the future. It is far from clear how such lim-
its could and should be formulated and enforced: should ‘un-
limited’ data tariffs be replaced with volume quotas or dif-
ferential pricing for services of various ‘importance’ ? Should
micro-payment schemes incentivise use of more bandwidth
frugal and offline media?
It is clear, however, that the dynamics of Internet traffic
growth are changing. In many ways the technical capac-
ities of data infrastructures, such as broadband networks,
continue to limit to the data flows and services that they
also make possible. Specifically, many broadband and mo-
bile contracts are limited and traffic is managed. In many
parts of the world, including areas in developed countries,
these infrastructural capacities are increasingly experienced
as tangible limitations to accessing a range of services that
are taken for granted elsewhere. To the extent that such lim-
its are experienced directly as limits to activity, then there
are important issues of social equity that cannot be ignored.
But where differences in data demand are less ‘visible’ and
less connected to explicit forms of ‘using the Internet’ other
possibilities for limiting data intensities might be explored
in ways that do not impact and consolidate existing inequal-
ities of access.
Our community is well placed to help shape these debates
and possible futures. By measuring and understanding the
holistic value and impacts of such systems, and understand-
ing the varied ways in which they change, can we help to
bring about a future where these systems operate within
similar and sufficient limits?
We would like to give our thanks to our colleague Kelly
Widdicks, for her compilation and analysis of the Sandvine
reports from 2012 to 2015.
[1] A. Al-Fuqaha, M. Guizani, M. Mohammadi,
M. Aledhari, and M. Ayyash. Internet of things: A
survey on enabling technologies, protocols, and
applications. Communications Surveys & Tutorials,
IEEE, 17(4):2347–2376, 2015.
[2] A. S. Andrae and T. Edler. On global electricity usage
of communication technology: trends to 2030.
Challenges, 6(1):117–157, 2015.
[3] O. E. G. Bates. The impacts of domestic media and
ICT: a study of digital technology, energy
consumption, data demand and everyday practice.
PhD thesis, 2015.
[4] Cisco. Cisco visual networking index: Forecast and
methodology, 2012–2017. Technical report, May 29,
[5] K. de Decker. Why we need a speed limit for the
internet. Technical report, Low-tech Magazine, 2015.
[6] Ericsson. Ericsson mobility report: On the pulse of the
networked society (mwc edition). Technical report,
Ericsson, Feb 2015.
[7] J. Gantz and D. Reinsel. The digital universe in 2020:
Big data, bigger digital shadows, and biggest growth
in the far east. IDC iView: IDC Analyze the future,
2007:1–16, 2012.
[8] ITU. The world in 2015. Technical report,
International Telecommunications Union, 2015.
[9] C. Lord, M. Hazas, A. K. Clear, O. Bates, J. Morley,
and A. Friday. Demand in my pocket: mobile devices
and the data connectivity marshalled in support of
everyday practice. In Proc. CHI, 2015.
[10] Ofcom. Adults’ media use and attitudes report.
Technical report, Ofcom, 2014.
[11] Ofcom. Communications market report. Technical
report, Ofcom, 2014.
[12] C. Preist and P. Shabajee. Energy use in the media
cloud: Behaviour change, or technofix? In Cloud
Computing Technology and Science (CloudCom), 2010
IEEE Second International Conference on, pages
581–586. IEEE, 2010.
[13] B. Raghavan and J. Ma. The energy and emergy of
the internet. In Proceedings of the 10th ACM
Workshop on Hot Topics in Networks, HotNets-X,
pages 9:1–9:6, New York, NY, USA, 2011. ACM.
[14] N. Robinson, J. Freeman, J. Gaspers, V. Horvath,
T. Hellgren, and A. Hull. Living room connected
devices. 2014.
[15] I. Røpke. Theories of practice —new inspiration for
ecological economic studies on consumption. Ecological
Economics, 68(10):2490–2497, 8 2009.
[16] I. Røpke and T. H. Christensen. Energy impacts of
ICT – insights from an everyday life perspective.
Telematics and Informatics, 29(4), 2012.
[17] Sandvine. Global internet phenomenon report.
Technical report, Sandvine Incorporated ULC, 1h
[18] M. Z. Shafiq, L. Ji, A. X. Liu, J. Pang, and J. Wang.
A first look at cellular machine-to-machine traffic:
large scale measurement and characterization. ACM
SIGMETRICS Performance Evaluation Review,
40(1):65–76, 2012.
[19] W. Van Heddeghem, S. Lambert, B. Lannoo, D. Colle,
M. Pickavet, and P. Demeester. Trends in worldwide
ict electricity consumption from 2007 to 2012.
Computer Communications, 50:64–76, 2014.
[20] A. Walgrove. The explosive growth of online video, in
5 charts. Technical report, 2015.
... An advertising ban on the Internet could further reduce the absolute Internet data volumes [123]. Other approaches are concerned with limiting broadband capacities [124] or limiting online video resolution [125]. ...
Full-text available
ICT hold significant potential to increase resource and energy efficiencies and contribute to a circular economy. Yet unresolved is whether the aggregated net effect of ICT overall mitigates or aggravates environmental burdens. While the savings potentials have been explored, drivers that prevent these and possible counter measures have not been researched thoroughly. The concept digital sufficiency constitutes a basis to understand how ICT can become part of the essential environmental transformation. Digital sufficiency consists of four dimensions, each suggesting a set of strategies and policy proposals: (a) hardware sufficiency, which aims for fewer devices needing to be produced and their absolute energy demand being kept to the lowest level possible to perform the desired tasks; (b) software sufficiency, which covers ensuring that data traffic and hardware utilization during application are kept as low as possible; (c) user sufficiency, which strives for users applying digital devices frugally and using ICT in a way that promotes sustainable lifestyles; and (d) economic sufficiency, which aspires to digitalization supporting a transition to an economy characterized not by economic growth as the primary goal but by sufficient production and consumption within planetary boundaries. The policies for hardware and software sufficiency are relatively easily conceivable and executable. Policies for user and economic sufficiency are politically more difficult to implement and relate strongly to policies for environmental transformation in general. This article argues for comprehensive policies for digital sufficiency, which are indispensible if ICT are to play a beneficial role in overall environmental transformation.
... To get some understanding, we need to look at independent studies. Support for the projection that AI will contribute substantially to climate change can be found in studies of the actual and expected carbon footprint of ICT in general (Andrae and Edler, 2015;Hazas et al., 2016;International Energy Agency (IEA), 2017;Belkhir and Elmeligi, 2018;Jones, 2018;Unwin, 2020). It can also be found in studies of the footprint of individual AI applications (Strubell et al., 2019;Anthony et al., 2020). ...
Full-text available
Purpose The purpose of this paper is to pinpoint and analyse ethical issues raised by the dual role of artificial intelligence (AI) in relation to climate change, that is, AI as a contributor to climate change and AI as a contributor to fighting climate change. Design/methodology/approach This paper consists of three main parts. The first part provides a short background on AI and climate change respectively, followed by a presentation of empirical findings on the contribution of AI to climate change. The second part presents proposals by various AI researchers and commentators on how AI companies may contribute to fighting climate change by reducing greenhouse gas emissions from training and use of AI and by providing AI assistance to various mitigation and adaptation measures. The final part investigates ethical issues raised by some of the options presented in the second part. Findings AI applications may lead to substantial emissions but may also play an important role in mitigation and adaptation. Given this dual role of AI, ethical considerations by AI companies and governments are of vital importance. Practical implications This paper pinpoints practical ethical issues that AI companies and governments should take into account. Social implications Given the potential impact of AI on society, it is vital that AI companies and governments take seriously the ethical issues raised by the dual role of AI in relation to climate change. Originality/value AI has been the subject of substantial ethical investigation, and even more so has climate change. However, the relationship between AI and climate change has received only limited attention from an ethical perspective. This paper provides such considerations.
... the Internet is evident (Hazas et al., 2016;Vassakis et al., 2018). This growth leads to the need for efficient search engines that provide Internet users with a friendly and agile way to find relevant information according to some criteria. ...
In this paper, we address a problem arising in information retrieval (IR) called composite retrieval problem (CRP) of diverse and complementary bundles. The CRP aims to group items into bundles and then select a subset of such bundles, so that we can maximise the similarity of the items within a bundle and, simultaneously, can maximise the complementarity of the selected bundles. To this end, the CRP approach considers the existing relations among items' attributes, leading to the selection of bundles that satisfy users' expectations without the needing for any refining query and, thus, improving the searching experience, with respect to traditional IR approaches. In this study, we propose three efficient yet straightforward algorithms, namely Local Search, Iterative Local Search and Variable Neighbourhood Search. Further, two different neighbourhood moves are evaluated at each algorithm. Although the first neighbourhood move is focused on the exploitation of the nearby search space, the second one is focused on the exploration of larger portions of the search space. All these algorithms are applied to two real-world publicly available instances and compared to the state-of-the-art algorithms in CRP. Obtained results suggest that combining both neighbourhood moves leads to better results in terms of both, the complementarity of the bundles and similarity of the items. Further, results show that, after statistical analysis, the proposed algorithms are significantly better, for the vast majority of the experiments performed in this study, when compared to the state-of-the-art algorithms in CRP.
... And Northern Virginia, being the base for the largest number of data centers on the planet, is operated by a utility company with only 1% of its electricity coming from renewable sources (Cook et al., 2017). With the appearance of wasteful cryptocurrency mining (Hern, 2018) and 5G networks forced on realizing the Internet of Things, data and traffic collection is already accelerating (Hazas et al., 2016). Moreover, production of electrical devices requires not only big energy expenditure but also intensive mining of raw materials, the same as plastic used for producing devices and its packaging. ...
Full-text available
The aim of the article is to study the role of artificial intelligence (AI) in solving current issues of climate change, environmental protection and natural resources management. The advantages and threats of using AI for the development of political and legal parameters for ensuring the safe and effective implementation of technological system, as well as ensuring sustainable control over its functioning and development trends, are analyzed. The relevance of the topic is substantiated by the fact that the legislative basis in this area is at the early stage of formation, while the scale of the impact of AI on all the aspects of social life may be impossible to accurately foresee. A special attention is paid to the analysis of the legal regulation of these issues in the context of European Union and Ukraine. The present work is one of the few that addresses three issues: climate change, the growing influence of artificial intelligence, and the possibility of legal regulation of the use of AI to solve urgent environmental problems without threatening the fundamental human rights and freedoms.
... With the emergence of 5G networks aiming to realize the "internet of things," the increased acceleration of data collection and traffic is already underway. 592 In addition to 5G antennas consuming far more energy than their 4G predecessors, 593 the introduction of 5G is poised to fuel a proliferation of carbon-intensive AI technologies, including autonomous driving 594 and telerobotic surgery. 595 A core contributor to the AI field's growing carbon footprint is a dominant belief that "bigger is better." ...
Full-text available
There is mounting public concern over the influence that AI based systems has in our society. Coalitions in all sectors are acting worldwide to resist hamful applications of AI. From indigenous people addressing the lack of reliable data, to smart city stakeholders, to students protesting the academic relationships with sex trafficker and MIT donor Jeffery Epstein, the questionable ethics and values of those heavily investing in and profiting from AI are under global scrutiny. There are biased, wrongful, and disturbing assumptions embedded in AI algorithms that could get locked in without intervention. Our best human judgment is needed to contain AI's harmful impact. Perhaps one of the greatest contributions of AI will be to make us ultimately understand how important human wisdom truly is in life on earth.
... With the emergence of 5G networks aiming to realize the "internet of things," the increased acceleration of data collection and traffic is already underway. 593 In addition to 5G antennas consuming far more energy than their 4G predecessors, 594 the introduction of 5G is poised to fuel a proliferation of carbon-intensive AI technologies, including autonomous driving 595 and telerobotic surgery. 596 A core contributor to the AI field's growing carbon footprint is a dominant belief that "bigger is better." ...
Full-text available
The rapid spread of artificial intelligence (AI) systems has precipitated a rise in ethical and rights-based frameworks intended to guide the development and use of these technologies. Despite the proliferation of these "AI principles", there is mounting public concern over the influence that the AI systems have in our society, and coalitions in all sectors are organizing to resist harmful applications of AI worldwide. Responses from peoples everywhere, from workers protesting unethical conduct and applications of AI, to student's protesting MIT's relationships with donor, sex trafficker and pedophile Jeffery Epstein, to the healthcare community, to indigenous people addressing “the twin problems of a lack of reliable data and information on indigenous peoples and biopiracy and misuse of their traditional knowledge and cultural heritage”, to smart cities stakeholders, and many others. Like corporations, governments around the world have adopted strategies for becoming leaders in the development and use of Artificial Intelligence, fostering environments congenial to AI innovators. However, in most cases, neither corporations nor policymakers have sufficiently addressed how the rights of children fit into their AI strategies or products. The role of artificial intelligence in children’s lives—from how children play, to how they are educated, to how they consume information and learn about the world—is expected to increase exponentially over the coming years. Thus, it’s imperative that stakeholders evaluate the risks and assess opportunities to use artificial intelligence to maximize children’s wellbeing in a thoughtful and systematic manner. This paper discusses AI and children's rights in the context of social media platforms such as YouTube, smart toys, and AI education applications. The Hello Barbie, Cloud Pets, and Cayla smart toys case studies are analyzed, as well as the ElsaGate social media hacks and education's new Intelligent Tutoring Systems and surveillance of students apps. Though AI has valuable benefits for children, it presents some particular challenges around important issues including child safety, privacy, data privacy, device security and consent. The article maps together the potential positive and negative uses of AI on children’s lives, in hopes to contribute to the conversation on developing a child rights-based framework for artificial intelligence that delineates rights and corresponding duties for governments, educators, developers, corporations, parents, and children around the world. The article concludes with some recommendations for corporation, parents, governments, and educators on Responsible AI development for children. Technology giants, all of whom are heavily investing in and profiting from AI, must not dominate the public discourse on responsible use of AI, we all need to shape the future of our core values and democratic institutions. As artificial intelligence continues to find its way into our daily lives, its propensity to interfere with our rights only gets more severe. Many of the issues mentioned in this examination of harmful AI are not new, but they are greatly exacerbated and threatened by the scale, proliferation, and real-life impact that artificial intelligence facilitates. The potential of artificial intelligence to both help and harm people is much greater than earlier technologies. Continuing to examine what safeguards and structures can address AI’s problems and harms, including those that disproportionately impact marginalized people.
... Several researchers [1][2][3] worked on reducing bandwidth requirements, whereas others [4] worked on optimizing storage utilization. In today's modern world, a huge amount of data is being generated [5], therefore, extracting the required information from large data volumes is a difficult task, while the storage of large data volumes is an issue [6]. Multimedia mining is a subfield of data-mining that addresses the issues of extracting information such as audio, text, links, images, or digital writing from multimedia files. ...
Full-text available
Data storage is always an issue, especially for video data from CCTV cameras that require huge amounts of storage. Moreover, monitoring past events is a laborious task. This paper proposes a motion detection method that requires fewer calculations and reduces the required data storage up to 70%, as it stores only the informative frames, enabling the security personnel to retrieve the required information more quickly. The proposed method utilized a histogram-based adaptive threshold for motion detection, and therefore it can work in variable luminance conditions. The proposed method can be applied to streamed frames of any CCTV camera to efficiently store and retrieve informative frames.
... Datafication: Datafication refers to the acceleration of data collection, transfer and storage (Hazas et al., 2016;Sadowski, 2019). Discussions of data typically focus on user data from apps and social media; however, the industrial use of data from machinery, transportation, commerce and other activities has become the norm. ...
Full-text available
This report addresses the nature, scope and possible effects of digital automation. It reviews relevant literature and situate s modern debates on technological change in historical context. It identifies threats to job quality and an unequal distribution of the risks and benefits associated with digital automation. It also offers some policy options that, if implemented, would help to harness technology for positive economic and social ends. The policy options range from industry and sectoral skills alliances that focus on facilitating transitions for workers in 'at risk' jobs, to proposals for the reduction in work time. The suggested policies derive from the view that digital automation must be managed on the basis of principles of industrial democracy and social partnership. The report argues for a new Digital Social Contract. At a time of crisis, the policy options set out in the report aim to offer hope for a digital future that works for all.
Full-text available
Le métabolisme territorial offre un paradigme pour étudier les flux de matières et d’énergie sur un territoire. Il vise à mieux qualifier et quantifier les ressources mobilisées et rejetées dans l’environnement. Néanmoins, l’étude du métabolisme reste complexe par la quantité de données à mobiliser et à traiter. Dans cette thèse, nous abordons directement cette problématique. Pour commencer, nous formalisons les notions et approches à mobiliser autour du traitement des données et du métabolisme. Nous concevons ensuite un Système d’Information pour l’Analyse du Métabolisme des Territoires (SINAMET). Enfin, nous mettons en application ces outils sur quatre cas d’études : la consommation d’énergie du patrimoine de l’Eurométropole de Strasbourg, les marchandises transportées par voie navigable dans le port de Strasbourg, la sensibilité à l’échelle des indicateurs d’importation et d’exportation, et les flux de matières alimentaire à l’échelle de l’Eurométropole.
Full-text available
This paper provides an overview of the Internet of Things (IoT) with emphasis on enabling technologies, protocols, and application issues. The IoT is enabled by the latest developments in RFID, smart sensors, communication technologies, and Internet protocols. The basic premise is to have smart sensors collaborate directly without human involvement to deliver a new class of applications. The current revolution in Internet, mobile, and machine-to-machine (M2M) technologies can be seen as the first phase of the IoT. In the coming years, the IoT is expected to bridge diverse technologies to enable new applications by connecting physical objects together in support of intelligent decision making. This paper starts by providing a horizontal overview of the IoT. Then, we give an overview of some technical details that pertain to the IoT enabling technologies, protocols, and applications. Compared to other survey papers in the field, our objective is to provide a more thorough summary of the most relevant protocols and application issues to enable researchers and application developers to get up to speed quickly on how the different protocols fit together to deliver desired functionalities without having to go through RFCs and the standards specifications. We also provide an overview of some of the key IoT challenges presented in the recent literature and provide a summary of related research work. Moreover, we explore the relation between the IoT and other emerging technologies including big data analytics and cloud and fog computing. We also present the need for better horizontal integration among IoT services. Finally, we present detailed service use-cases to illustrate how the different protocols presented in the paper fit together to deliver desired IoT services.
Full-text available
This work presents an estimation of the global electricity usage that can be ascribed to Communication Technology (CT) between 2010 and 2030. The scope is three scenarios for use and production of consumer devices, communication networks and data centers. Three different scenarios, best, expected, and worst, are set up, which include annual numbers of sold devices, data traffic and electricity intensities/efficiencies. The most significant trend, regardless of scenario, is that the proportion of use-stage electricity by consumer devices will decrease and will be transferred to the networks and data centers. Still, it seems like wireless access networks will not be the main driver for electricity use. The analysis shows that for the worst-case scenario, CT could use as much as 51% of global electricity in 2030. This will happen if not enough improvement in electricity efficiency of wireless access networks and fixed access networks/data centers is possible. However, until 2030, globally-generated renewable electricity is likely to exceed the electricity demand of all networks and data centers. Nevertheless, the present investigation suggests, for the worst-case scenario, that CT electricity usage could contribute up to 23% of the globally released greenhouse gas emissions in 2030.
Conference Paper
Full-text available
This paper empirically explores the role that mobile devices have come to play in everyday practice, and how this links to demand for network connectivity and online services. After a preliminary device-logging period, thirteen participants were interviewed about how they use their iPhones or iPads. Our findings build a picture of how, through use of such devices, a variety of daily practices have come to depend upon a working data connection, which sometimes surges, but is at least always a trickle. This aims to inform the sustainable design of applications, services and infrastructures for smartphones and tablets. By focusing our analysis in this way, we highlight a little-explored challenge for sustainable HCI and discuss ideas for (re)designing around the principle of ‘light-weight’ data ‘needs’.
Full-text available
Information and Communication Technology (ICT) devices and services are becoming more and more widespread in all aspects of human life. Following an increased worldwide focus on the environmental impacts of energy consumption in general, there is also a growing attention to the electricity consumption associated with ICT equipment. In this paper we assess how ICT electricity consumption in the use phase has evolved from 2007 to 2012 based on three main ICT categories: communication networks, personal computers, and data centers. We provide a detailed description of how we calculate the electricity use and evolution in these three categories. Our estimates show that the yearly growth of all three individual ICT categories (10%, 5%, and 4% respectively) is higher than the growth of worldwide electricity consumption in the same time frame (3%). The relative share of this subset of ICT products and services in the total worldwide electricity consumption has increased from about 3.9% in 2007 to 4.6% in 2012. We find that the absolute electricity consumption of each of the three categories is still roughly equal. This highlights the need for energy-efficiency research across all these domains, rather than focusing on a single one.
Cellular network based Machine-to-Machine (M2M) communication is fast becoming a market-changing force for a wide spectrum of businesses and applications such as telematics, smart metering, point-of-sale terminals, and home security and automation systems. In this paper, we aim to answer the following important question: Does traffic generated by M2M devices impose new requirements and challenges for cellular network design and management? To answer this question, we take a first look at the characteristics of M2M traffic and compare it with traditional smartphone traffic. We have conducted our measurement analysis using a week-long traffic trace collected from a tier-1 cellular network in the United States. We characterize M2M traffic from a wide range of perspectives, including temporal dynamics, device mobility, application usage, and network performance. Our experimental results show that M2M traffic exhibits significantly different patterns than smartphone traffic in multiple aspects. For instance, M2M devices have a much larger ratio of uplink to downlink traffic volume, their traffic typically exhibits different diurnal patterns, they are more likely to generate synchronized traffic resulting in bursty aggregate traffic volumes, and are less mobile compared to smartphones. On the other hand, we also find that M2M devices are generally competing with smartphones for network resources in co-located geographical regions. These and other findings suggest that better protocol design, more careful spectrum allocation, and modified pricing schemes may be needed to accommodate the rise of M2M devices.
The environmental implications of information and communication technology (ICT) have been the subject of study since the early 1990s. Although previous research covers energy issues quite extensively, the treatment of the energy impacts of ICT integration in everyday life is still inadequate. The purpose of this paper is to complement the existing research by applying a perspective from which everyday life takes centre stage. A theoretical framework for describing and analysing the energy impacts of everyday life is outlined, based on a combination of practice theory and time geography. The framework is applied to a discussion of how ICT co-develops with changing everyday practices and energy-demanding features of everyday life. Based on empirical findings, it is explored how the use of ICT affects practices in relation to time and space, and it is argued that the changes may increase energy consumption considerably. The findings do not suggest that the integration of ICT in everyday practices inherently results in a more energy-intensive everyday life. ICTs have a great potential for reducing energy consumption, but the realisation of this depends on the wider economic and political conditions.
Recent years have seen a flurry of energy-efficient networking research. But does decreasing the energy used by the Internet actually save society much energy? To answer this question, we estimate the Internet's energy consumption. We include embodied energy (emergy)---the energy required to construct the Internet---a quantity that has often been ignored in previous work. We find that while in absolute terms the Internet uses significant energy, this quantity is negligible when compared with society's colossal energy use.