Drowning in data – digital pollution, green IT, and sustainable access – WS 11 2017

From EuroDIG Wiki
Jump to: navigation, search

7 June 2017 | 14:00 - 15:30 | Room Tornimäe II, Swissotel, Tallinn, Estonia | video record
Programme overview wiki | Programme overview EuroDIG web site

Session information

Date: Wednesday, June 7, 2017

Time: 14:00 - 15:30 (GMT+3)

Room: Tornimäe II

Venue: Swissotel, Tallinn, Estonia

Session teaser

We cannot legitimately discuss Internet access without addressing sustainability.

Keywords

Access, Carbon footprint, Circular economy, Climate change, Digital pollution, E-waste, Energy, Greenhouse gas emissions, Green IT, ICT4D, ICT4S, Sustainability, Sustainable development

Session description

Largely absent from discussions surrounding Internet governance and Internet access is the concept of sustainability. It touches on much more than merely the kind of energy supplying critical Internet infrastructure or cooling servers; it also relates to the sourcing, manufacturing, and recyclability of Internet-connected devices, the amount of power that is being consumed by information and communications technologies (ICTs), the energy needed to power our ever-growing data consumption, and even "digital pollution," such as unused spectrum or Internet Protocol (IP)/Autonomous System (AS) addresses.

Europe is a global leader in promoting sustainability and renewable energy, and considering that environmental challenges such as the proliferation of e-waste and worsening climate change, the EuroDIG community is in a prime position to promote sustainable access and concepts like the circular economy as it relates to ICTs. Given the Internet governance community's endeavor to [www.intgovforum.org/multilingual/content/policy-options-for-connecting-and-enabling-the-next-billions-phase-ii connect the next billion] Internet users as well as the exponential growth of Internet of Things (IoT) devices, it is critical that we consider sustainability when addressing Internet access.

Working with the RIPE NCC, this session aims to bring the concept of sustainability to the fore, and explore how the Internet governance community, specifically in Europe, can work in a multistakeholder fashion to promote sustainability within the Internet ecosystem.

Our intended aims & outcomes for the session include:

  1. Link the work of the ICT for sustainability community with the Internet governance community, and highlight the interconnection
  2. Promote sustainable access and concepts such as green IT and the circular economy
  3. Address why sustainability and sustainable access are relevant to Europe and the European context

We are also partnering with a local e-waste recycler (ESS – Ringlus) to provide an on-site, e-waste collection point for those attending EuroDIG to leave their e-waste

Format

The session will be conducted in a manner that maximizes interaction using a roundtable discussion format. It will include key participants as well who can better inform the discussion about various perspectives on the topic. The agenda is as follows:

10' – Introduction to the session and of the key participants, and an overview of why the subject is relevant and timely.

15' – Keynote presentation by Dr. Mike Hazas (University of Lancaster)

60' – Moderated discussion with guiding questions (to allow for open and organic discussion relevant to the roundtable participants' questions and interest)

5' – Wrap-up and conclusion

Further reading

Select resources and links reflecting perspectives from stakeholders are listed below:

  1. Digiconomist (2017) – Bitcoin Energy Consumption Index
  2. Marc Bevand (2017) – Serious faults in Digiconomist's Bitcoin Energy Consumption Index
  3. David Franquesa and Leandro Navarro (2017) – Sustainability and Participation in the Digital Commons
  4. VOX (2017) – Why your old phones collect in a junk drawer of sadness
  5. Energy Transitions Comission (2017) – Better energy, greater prosperity: Achievable pathways to low-carbon energy systems
  6. VOX (2017) – Why humans are so bad at thinking about climate change
  7. DEMAND (n.d.) – Dynamics of Energy, Mobility and Demand
  8. Jason Koebler (Motherboard) (2017) – Apple Forces Recyclers to Shred All iPhones and MacBooks
  9. Arielle Duhaime-Ross (VICE) (2017) – No mining required: Apple promises to stop mining minerals to make iPhones – it just isn’t sure how yet
  10. Apple (2017) – 2016 Environmental Responsibility Report
  11. Michael Oghia (CircleID) (2017) – Shedding Light on How Much Energy the Internet and ICTs Consume
  12. Telefónica Internet of Things (IoT) Team (2017) – Infographic: 5 saving and efficiency Key Factors in Smart Energy Solutions
  13. World Health Organization (WHO) (2017) – Inheriting a Sustainable World: Atlas on Children’s Health and the Environment
  14. E-mail exchange between Michael Oghia and Vint Cerf (2017) – What fraction of the power consumption does the Internet (and its access devices) take?
  15. Internet Governance Forum (IGF) Dynamic Coalition on Internet and Climate Change (DCICC) (2016) – Linking ICTs and Climate Change: Towards COP21 and SDGs
  16. Michael Oghia (CircleID) (2016) – The Internet’s Climate Quandary and the Inconvenience of Practicing What We Preach
  17. David Franquesa, Leandro Navarro, & Xavier Bustamante (2016) – A Circular Commons for Digital Devices: Tools and Services in eReuse.org
  18. Emily Cox, Sarah Royston, and Jan Selby (UK Energy Research Centre) (2016) – The impacts of non-energy policies on the energy system: A scoping paper
  19. Danny Bradbury (The Register) (2016) Super cool: Arctic data centres aren't just for Facebook
  20. Karmenu Vella (European Commission blog) (2016) – To achieve the Sustainable Development Goals (SDGs) we must change our linear economic model
  21. Hazas et al. (2016) – Are there limits to growth in data traffic?: On time use, data generation and speed
  22. CircleID (2016) – Data growth, IoT will lead to unlimited energy consumption if not controlled, scientists warn
  23. GSMA (2016) – Mobile Industry Impact Report: Sustainable Development Goals
  24. Global e-Sustainability Initiative (GeSI) (2015) – SMARTer2030 ICT Solutions for 21st Century Challenges
  25. Jason Koebler (Motherboard) (2015) – How to Fix Everything
  26. Jessica Benko (New York Times) (2015) – Making and Unmaking the Digital World
  27. CCCB Lab (2014) – How polluting is the Internet?
  28. The Atlantic (2014) – Inside a Massive Electronics Graveyard
  29. Leyla Acaroglu (New York Times) (2013) – Where Do Old Cellphones Go to Die?
  30. Greenpeace (2012) – How Clean is Your Cloud?
  31. International Telecommunications Union (ITU) (n.d.) – ITU-T Goal 3 Sustainability
  32. UN Climate Technology Centre & Network (n.d.) – CTCN
  33. Bitcoin Wiki: Myths (n.d.) – Bitcoin mining is a waste of energy and harmful for ecology

Additional resources on sustainability in Europe

  1. Rescoop.eu (n.d.) – European Federation of Renewable Energy Cooperatives
  2. European Union (2016) – Green Digital Charter
  3. European Union product policy (n.d.) – Cool Products.EU
  4. European Union Resource Efficiency Coordination Action (n.d.) – EURECA Project
  5. European Telecommunications Standards Institute (2015) – Environmental Engineering (EE): Methodology for environmental Life Cycle Assessment (LCA) of Information and Communication Technology (ICT) goods, networks, and services
  6. European Commission (2014) – Study on the practical application of the new framework methodology for measuring the environmental impact of ICT – cost/benefit analysis
  7. European Commission (2013) – Towards an overall measurement methodology of the carbon and energy footprints of the ICT sector
  8. European Commission (2013) – ICT footprint: Pilot testing on methodologies for energy consumption and carbon footprint of the ICT-sector
  9. ICT Footprint.eu (n.d.) – European Framework Initiative for Energy & Environmental Efficiency in the ICT Sector

Additional resources on data center efficiency

  1. Google data centers
  2. Efficiency: How others can do it (Google)
  3. Google’s green data centers: Network POP case study
  4. Machine learning applications for data center optimization
  5. The data center as a computer: An introduction to the design of warehouse-scale machines
  6. Data center best practices guide: Energy efficiency solutions for high-performance data centers

Additional resources on e-waste

  1. UN University Institute for the Advanced Study of Sustainability (2014) – The global e-waste monitor: Quantities, flows and resources
  2. e-Reuse
  3. International Telecommunications Union (ITU) activities on e-waste
  4. iFixIt e-waste website
  5. WHO e-waste
  6. Solving the e-Waste Problem (STEP)
  7. Flashdrives for Freedom
  8. Pacific Hazardous Waste Management (PacWaste)

People

Focal Point & WS 11 Wiki Editor

Subject Matter Expert:

  • Chris Buckridge (RIPE NCC | Netherlands)

Key Participants (for workshop)

  • Mike Hazas (Lancaster University | UK)

Dr. Mike Hazas is a senior lecturer at Lancaster University, who works at the confluence of human-computer interaction and social science. His research combines qualitative and quantitative methods to understand everyday practices and technologies, how they can be related to carbon emissions and energy demand, and more sustainable trajectories.

Moderator

  • Michael J. Oghia

Michael is a Belgrade-based, independent consultant working within the Internet governance ecosystem. His work includes a focus on sustainable access, development, and digital rights. Michael is also a YOUthDIG 2017 focal point & co-organizer. For more information, see his LinkedIn profile.

Remote Moderator

  • Dan Waugh (Union of Students in Ireland | Republic of Ireland)

Daniel Waugh is the vice president for campaigns for the Union of Students in Ireland, and represents 354,000 students across Ireland. He is an Open Con alumni, an award-winning journalist and campaigner, and EuroDIG alumnus. Dan is also a YOUthDIG 2017 focal point & co-organizer. For more information, see his Linkedin profile.

Organising Team (Org Team)

  1. Suso Baleato (CSISAC | Germany)
  2. Suzanne Taylor (RIPE NCC | Netherlands)
  3. Alexandra Groza (Independent | Romania)
  4. Mike Hazas (Lancaster University | UK)
  5. Robert Clark (ITU | Switzerland)
  6. Amali De Silva-Mitchell (Futurist/Consultant | UK)

Reporter

Elisabeth recently obtained her MA from the University of Graz. Her academic focus lies on human rights law and international relations theory, in conjunction with environmental sustainability management. Elisabeth is also a YOUthDIG 2017 focal point & co-organizer.

Current discussion, conference calls, schedules and minutes

The organizing team – hereby referred to as org team – took some time to shape, as the topic at hand has been traditionally absent from the wider agenda of Internet governance. The first pre-org team meeting was held between Michael Oghia (focal point) and RIPE NCC staff members Chris Buckridge (subject matter expert) and Suzanne Taylor (org team member) on 16 March 2017 while in Copenhagen, Denmark, for ICANN58 (the minutes of that meeting are available here).

Two weeks later, Chris and Michael met once again at RightsCon 2017 in Brussels, where Michael was able to connect with many new individuals who may be interested in getting involved. These include an energy-conscious data host (Greenhost), a sustainably sourced smartphone maker (Fairphone), and a member of the Access Now team (which hosted RightsCon) that cares about sustainability. In addition to Suso Baleato of CSISAC (OECD), who formally joined the org team as well at RightsCon, Cristobal Irazoqui (independent) met with Michael on 29 March 2017, and he agreed to join the org team as well but eventually had to step down. Chris also put Michael in touch with a contact at Telefonica, which has conducted environmentally friendly research, and Michael reached out to Google. As of 4 April 2017, Michael is still waiting to hear back from the ITU secretariat and Study Group 5, but he am also in touch with a contact who works at the UN's Climate and Technology Center & Network (CTCN) to see if he would like to join the org team (see these notes for more information).

On Monday, April 24, Michael sent the first message to the organizing team (available here), followed by discussion and follow-up emails (available here). We agreed to host our first online meeting (via Zoom) on May 8, 9, or 10 depending on the Doodle poll.

The org team held its first online meeting via the Zoom platform on Wednesday, May 10 from 14:00-14:30 (GMT+2) with Chris, Michael, and Robert present. They discussed the following agenda:

  1. Overview of our progress so far
  2. Review the draft agenda
  3. Discuss key participants
  4. Brainstorm intended outcomes for the workshop given our draft agenda and key participants
  5. Moderator, reporter, and remote moderator roles

Michael gave an overview of the progress made so far, and updated about collaborating with an e-waste recycler in Tallinn to provide e-waste recycling bins during the session. We agreed to simply the agenda to allow for more open and organic discussion, and discussed key participant recommendations (including inviting an operator to join the discussion (Chris will follow-up) and/or inviting Dominique Lazanski from GSMA). Michael will also reach out to Konstantinos Komaitis (Internet Society) who confirmed his interest and availability. Michael will also reach out to individuals in his network he knows are attending EuroDIG to find a remote moderator and reporter. Lastly, we discussed our intended outcomes, which include: (1) linking the work of the ICT for sustainability community with the Internet governance community, (2) Promoting sustainable access and concepts such as green IT and the circular economy, and (3) Address why it is relevant to Europe and the European context.

Contact

Get in contact with the Org Team by sending an email to ws11@eurodig.org or Michael: mike.oghia[at]gmail.com.

Video record

https://www.youtube.com/watch?v=ZEsKI41s6O0

Messages

  1. As data traffic rises in tandem with the increase in the number of Internet users, Internet of things (IoT) technologies, and machine-to-machine traffic, so does energy consumption; thus, the Internet will account for a significant share of global energy consumption. There has to be an effort made to monitor and address this foreseeable rise in a sustainable way.
  2. Planned obsolescence and decreasing incentives to repair devices leads to high levels of electronic waste (e-waste), which is often discarded in dumps in the Global South. Eco labeling could bring an incentive for consumers to invest in more sustainable devices and might be used for software in a similar way. On a regulatory level, a switch to a circular economy model should be made a policy goal.
  3. Sustainability also manifests on a societal level, and the Internet will gain even more significance as more people are connected to it. Projects and policies that address ecological issues and find a solution that involve a sustainable social benefit should be advanced.
  4. Internet governance processes should collaborate with new stakeholders to address sustainability, and acknowledge a responsibility to incorporate sustainability in their discussions as well as the core of our work.

Transcript

Provided by: Caption First, Inc., P.O. Box 3066, Monument, CO 80132, Phone: 800 825 5234, www.captionfirst.com


This text is being provided in a realtime format. Communication Access Realtime Translation (CART) or captioning are provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings.


>> Hello.

>> We are going to wait a couple more minutes, to give people time to come in from lunch, and also the EuroDIG General Assembly just ended.

Some people told me that they would be coming, but it's okay, they will probably come in late. It is not that big of a deal. Let's go ahead and get the session started. My name is Michael Oghia. I'm the moderator for the session. I'm also the focal point for this workshop. I'm an independent consultant within the Internet Governance ecosystem, and I've been working now on let's say the connection between sustainability and the Internet since last year when I started writing about this topic.

This is the first time as far as I know that sustainability in this capacity is being addressed at a Internet Governance, sorry, the EuroDIG event. But I don't know if that's completely true. Either way, I want to thank some people that helped make this workshop possible, including Chris Buckridge and his staff at the RIPE NCC, our remote moderator Dan and our rapporteur Elizabeth, and then there is also, also there is a local eWaste recycling organisation called ESS Ringlus, and they have given us a bin to provide eWaste, in case anybody brought some, but they can't get rid of. So I really appreciate that.

Also, so I want to start with the concept of sustainable access. I've been asked a few times since I've been at EuroDIG what is sustainable access, and actually, it made me think what does sustain ability refer to. I looked it up, it is the ability for something to be maintained at a certain rate or level.

When I was thinking about how do we define sustainable access, I wrote it as it's, it refers to the ability for any user to continue to connect to the Internet over time.

What does this involve at least as it relates to us? It involves reliable infrastructure and the kind of energy that is supplying critical Internet infrastructure or whatever is including servers, how much power ICTs are consuming, how such power is being generated and the energy cost of data generation, storage and transit.

It involves sourcing, manufacturing and recycle ability of Internet connected devices, human resources and skills are also paramount to making sure individuals can connect to the Internet and stay connected. But then it's also concepts such as digital pollution and the availability of resources such as IP addresses and AS numbers, and of course the ecological impact.

Why is this a problem? It's a problem because of things like this. This is an eWaste dump, I'm not sure exactly where this one is located. I found it on, it was a unspecified place. But with the passage of the Paris agreement in 2016, but then the recent notification of withdrawal by the United States, it's absolutely critical that we all in our own capacities begin to address this very urgent topic.

But with that said, even within the context of Internet Governance, we want to connect the next billion people to the Internet, well, are we also considering the long term viability of our current processes and practices in making sure that those people that are connected will be able to maintain that connection? How will they connect with what devices will they connect with, what happens after those devices break, will they go to a dump like this, where what will happen? Considering especially with one billion more people connected to the Internet that is one billion more YouTube videos, e Government services used, pictures shared, Skype conversations or video conversations with loved ones, etcetera.

I was telling someone about this, that whenever I was telling people about this session, starting out, some people don't even know what eWaste is and how it's, for instance, exported to different countries to be disposed of. When I say disposed of, it's dumped. If you look at the EU, we know now that it sends it to India, to Nigeria, Ghana has a huge eWaste dump for instance, China has eWaste dumps as well.

The idea that the Internet, we always talk about the cloud, but the cloud is not the cloud. The cloud is infrastructure, it's on the ground, it's under the oceans. It's in our world. It's on the earth.

So what I really want to focus on is the fact that more importantly, this concept is really absent from our discussions, sustainable access is not part of our current discourse. We should not look back and address sustainability in retrospect. It should be integrated into the core of our work, public policy and considered a basic requirement, especially since we have a role to play in, for instance, protecting the environment.

Now, I wanted to quickly address kind of what are the aims and objectives of this expected outcomes that we hope to get. First of all we want to link the work of the ICT for sustain ability community. There are researchers doing work on sustainability, specifically how it relates to the Internet and ICTs, with the Internet Governance community and highlight that interconnection. There is a lot of room for work, especially as it comes to initiatives to connect people such as community networking and the like.

The second one is promote sustainable access and really catalyze a paradigm shift in how we talk about connecting. Then also address why sustainability and sustainable access are relevant to Europe and the European context which as many of you probably know, the G20 meeting is happening next month. There is a civil 20 group that was focusing on climate. I was part of that just a mailing list and I kept talking about sustainable access there as well, this horizon 2020, various EC initiatives that are going on with renewable energy, etcetera. There is a sustainable development goals of course and sustainable access, ICTs period are a huge part of realizing those.

Even then, we talk about the future, whether it be at this year's Internet Governance forum, whether here at EuroDIG even, we talk about the future, sustainability must be a part of that. Otherwise who knows what that future will look like. It might be a lot more grim than we hope.

Ultimately, Europe is a global leader in promoting sustainability and renewable energy and considering that the environmental challenges that we face, the EuroDIG community in particular I think is in a prime position to make sustainability a core component of the Internet Governance discourse. With that said, I want to outline the session.

We are going to, it is a bit unorthodox, I know it's not exactly, usually implied, but given that this session is meant to be bridge between communities so we are going to have a 15 minute keynote presentation, and then after that we will go into the, directly into the round table. I have some questions, but I really hope that each of you can drive the discussion as well.

With that said, without further ado, I want to introduce one of the people that really was one of the reasons why I started doing this work, because I came across his research through CircleID, and that inspired me to go, wait a minute, why aren't we talking about this? That is Dr. Mike Hazas, senior lecturer at Lancaster University who works at the confluence of human computer interaction and social science, his research combines qualitative and quantitative methods to understand every day practices and technologies, how they can be related to carbon emissions and energy demand and more sustainable trajectories. I will give the floor over to Mike. Can we switch the presentations as well?

>> MIKE HAZAS: Can you hear me out of the speakers? All I hear is myself. There we go. That is better.

Thanks, Michael. Seems I cannot stand in front of my screen. That's fine.

Very quick preface, I've done this in collaboration with colleagues, which we will talk about in a second. I gave a version of this talk first at the conference for computing within limits, that was about a year ago, pretty much exactly a year ago. And then again at a seminar series on the ethics of big data in Cambridge back in February, so there have been previous airings of these ideas. But I was excited to be able to come here and to be contacted by Michael to talk to this community, because a lot of times I'm presenting to computer science audiences, sometimes sociologists, and we particularly computer scientists focus on what should designers do, how should we design apps differently. But there are a lot of important higher level I think policy implications and governance implications and that's why I'm happy to be here. Thank you for the invitation.

This was done in context of the demand center which is a 5 million pound research center based at Lancaster University over five years. My immediate colleague is Janine Morley within that center, and the demand center as a whole concerns itself on understanding what energy is for. That might be whether it's on line shopping or home heating or whatever, and understanding how those resources have grown over time and how those change, with what people do so every day practice.

We work specifically on domestic IT within that larger center.

I codirect the social digital sustainability group, based in the school of computing and communications at Lancaster. As you are probably picking up from some of the disciplines on this we have sociologists and computer scientists. I have a checkered past. I come from an electrical engineering background and did a PhD in mobile computing, and then decided that technology and its impact can't really be understood without understanding people. I did a degree in sociology in order to think more strategically about socio technical systems and interested in the broader understandings of technology and how they relate to important problems like climate change.

This whole bit of work, the argument of this talk lays out is a response to writing by Kris Decker for Low Tech Magazine, which asserted there are no limits to growth when it comes to the Internet except for the energy supply itself.

That is in contrast to roads and hospitals, which have physical and material constraints. The Internet because of the nature of fiber and how easy it is to relay and also to reuse with faster endpoints, basically the backbones, same thing for cellular networks, we have to maintain them, but it's relatively easy to do. It is not like building a national network of hospitals when we roll out for example 5G. Uniquely among infrastructures, the energy is the Internet can only stop growing when energy resources run out unless we impose self chosen limits. This is his assertion. This is interesting. We couldn't think of anything immediately to refute that. We thought we would take that forward. That is what this talk does.

The talk goes in four parts. For the first, I'll assert, which I hope everyone here would agree with, that traffic volumes are growing, and how those are composed, and then I'll link those to a rising energy demand due to the Internet and its services. Then I'll talk about how human time and attention are also linked to the growth, potentially, and obviously I'll make the assertion that human time itself probably has some sort of limit.

Then I'll think about machine to machine communication, and going toward the future. A lot of the, and this is very future oriented itself, but as I go through the first three points I'll be giving lots of examples of how traffic is composed, how energy demand is composed now which I think will be hopefully of interest to this audience. A couple caveats. Michael showed the picture of the eWaste site. In any energy figures I give or growth figures, I'm not talking about the embodied emissions or energy in manufacturing and transport. Those are significant, they are important. But I generally won't be talking about them.

I also won't talk too much, a little bit, about the direct energy that digital stuff consumes in homes and offices and things like that, so this is the home electricity consumption in the UK and around 35 percent between computing and consumer electronics most of which is digital, is around 35 percent. That is currently of demand, and that is increasing as a share, as a trend partially because of other categories are becoming more efficient like lighting.

Returning to the assertion that data traffic volumes are growing, this is the basic per month average, per month per household broadband usage in gigabytes, going from around 20 in 2011, up to around 87 in 2015, and it keeps going up. So there is a clear rise there.

The Amsterdam Internet exchange published their traffic throughput, and we are seeing between 20 and 40 percent depending on the year annual growth in the last five years. This should all be fairly familiar stuff. When we talk about the Internet stuff generally does that, so that is what all these graphs are going to do.

Has anyone run into Sandvine? They do data consulting for Internet service providers. I'm seeing at least one nod, for Internet service providers, they publish aggregate figures from a selection of providers which might be representative of the current picture. If you aggregate, there used to be twice yearly, now they are yearly but for each region reports, you see similar sorts of trends. Basically volumes tend to rise steadily in North America, whereas in Europe they doubled over the past few years, and mobile access which is on the bottom, of course those numbers are smaller so these are megabytes, not gigabytes like they are for fixed lines like broadband at the top. But it still grows around 50 percent per year.

CISCO and Ericsson tell us the same things, for mobile traffic, 50 to 60 percent per year. That is the CISCO graph. That is the Ericsson forecast, sorry, current state of play. They forecast between 50 and 60 percent ongoing out through 2021, which is the point there.

It's fairly, mobile grows in subscribers, grows in traffic as more people get better phones, more people get signed up, as the mobile gets deployed in more places but it applies to fixed line too, just a slightly slower growth.

That hopefully makes the case for growth. I'll move on to the energy demand use of the Internet and its services. This is a survey of literature. This is from the lifecycle assessment community, which spends a lot of time trying to first estimate where the energy and emissions are going for various systems, food, transport and also ICT and Internet systems, and they revised their estimates every year. They argue about it. There is a occasional survey article. It is a lot of reading but it's, there is no other way of even vaguely knowing. So at least having a estimate has been helpful.

Depending on how you count it, if you don't count all the end user stuff like laptops and phone and televisions, it is around 5 percent, maybe 7 percent, depending on what you are looking at. Once you include all the stuff that draws upon the Internet, you are looking at 10 percent of global energy goes into digital stuff, right, and Internet traffic.

If anybody wants any of these references, by the way, they are in the limits paper, I'll happily share a link to that, some of the slides are, I tried to reference most of the slides, so for what I was saying.

Again, this shows up in the media quite a lot, whether it's cat videos or assertions like Britain will shut down, the power grids will shut down based on the load of the Internet, which is probably not outside the realm of possibility but maybe won't happen tomorrow, then as well the emissions due to data centers.

There are some alternative things like more and more providers are using renewable energy. There is a strong financial incentive for them to build renewable energy alongside data centers, places like Google, but still there is quite a lot of risk including Michael's article which I came across independently, it was before I met him. So this is not a shameless plug. This was in the presentation before he knew I was coming.

If you return to the lifecycle analysis literature, they do forecasts and modeling to say how things might change. The fixed access wired use and data centers are going to grow as a share as we move to 2030. This is the expected scenario. But the curve is like this. If you think of we are at 10 percent global energy now and other things are fixed, like transport, air travel, those things tend to grow a little but not as quickly, then as a share the Internet is going to keep growing as a share, so we could see 20 percent by 2030. Yep.

(someone speaking off microphone).

Yes, I agree. I wonder if that is because of something in their model, maybe they anticipate higher traffic. Things do get more efficient, of course. But they don't, the efficiency doesn't keep up with the growth in traffic. But it would be interesting to know what in their model. There is other best and worst case cases. We might flat line. It could happen, but it could also be a lot worse. But it depends on the assumptions in the model. I share your skepticism there.

Moving on, how much things keep growing does depend on human time and attention. I want to spend a little time on that. Turning to Sandvine data, this is the profile of peak traffic fixed on your left, and mobile on your right. So you can see that there is sorry the most significant parts being audio and video streaming, with social networking and of course web browsing which supports a number of practices and things that people do. Again, quite significant areas of practice.

Really, as you probably experience yourselves network services have moved into many areas of daily practice including things that didn't involve the network or mobile devices before, like exercise or watching things like that. And these have kind of been part of, you might tweet about your jog or you might whatever.

There is that kind of, it seems to have a hand in the growth. That people are changing how they use technology. There is also a lot more, we were able to use concurrent devices so we can stream something on television while we are also using our phone, both are accessing the network perhaps in a intensive way. So it's, that didn't used to be, you sat at a laptop or desktop and used the Internet as it were, but now it's easier to use two, maybe three devices concurrently.

What do we know about time use on line? That is a hard thing to quantify. It is not that people haven't tried. Back in 2005, people self reported that it was around 9.9 hours a week, that was based on surveys done by OFCOM, their mobile communications report, that's gone up to 20.5 hours a week, recently, 2014. In the U.S. similarly, they surveyed time spent paying attention to digital activities, which similarly has grown not quite double, but has increased.

It depends on the wording of the survey, depends how people think of their own, what is digital and what is not, things like that.

I'd like to point out that based on the time use survey data in the UK, there are certainly shifts. If we compare 1974 to 2005 in the category TV, media and games, this is not the newest data, the newest data allows people participating in the time use survey to indicate whether there was a digital component to what they were doing. They might have gone for a jog, but they can now indicate, and I use my phone on the jog or the phone was involved or something like that. There is the change in how they account for it. But you can already see that between '74 and 2005 there are shifts in how people spend their time with TV, media and games, perhaps maybe more in the morning, maybe less in the early evening but it gets pushed later as well.

Obviously there is going to be a limit. First of all, the global population is finite and less than half the world is deemed to be on line or have regular access to the Internet. In some parts of the world, mobile phones subscriptions are approaching saturation, particularly in Australia and Scandinavia where you have one or more than one phone, one subscription per person. But there is many more there. That could be a limit.

The hours in the day available to us is finite. We can only spend so much time not sleeping. We probably push that a lot, right? We get less sleep perhaps because of mobile devices. But surely there's got to be a limit there as well.

Eventually, you would think that it could, doing this it will eventually stop once the Internet reaches more of the population, once it takes up a certain amount of time that we can't bear.

I'd like to point out that despite those, the fact that we might do similar things like say watch a video or listen to music or check social networking, even though those practices what we actually do stays the same, and we might do those for the same amount of time day to day, they can actually slowly ramp up the amount of data required. So Net Flix started early with fairly close to standard definition, then it went to HD and then super HD and now you get 4K ultra HD. Nothing to say there won't be services for virtual reality which would require more bandwidth and similarly social networking have had increasingly video based content. That might be adverts, videos created by people you know, easier to share those. Facebook's auto play causes it to load video on your behalf without you having clicked on it. There is stuff that happens behind the scenes that ramps up the data. When Facebook switched on auto play there were blips like large surges, I should say, in global networks as it starts to auto play those. There was a lot of bad press, because it killed people's mobile quotas.

Yeah, there must be some kind of limit, but arguably, we should somehow have some say in the discussion of how quickly we approach that limit and when. But as well, apart from human time, there is this big data traffic potential for nonhumans, so for machines.

I'll return to my Sandvine breakdown of peak traffic. You can see marketplaces which can be app store updates and things which are pseudo automated, 6 percent of traffic. We did empirical work recently with 400 participants with two weeks of data from their phone. OS and background processes which includes notifications and stuff like that, as well as updates and backups, comprised 15 percent of traffic. This is nothing that necessarily would be fired off when they were doing stuff or just running in the background. This is in addition to all the listening, watching, social networking and other stuff that they were doing. This is happening on their behalf in a fairly automated fashion.

Currently there is the estimates around six and a half billion connected devices which is starting to approach the world population. Could be 21 billion by 2020 by some estimates. We have references in the paper, where the machine to machine communication could be approaching half of traffic by 2021.

It's arguable whether these will be particularly data intensive. I suppose in some ways they will. You can think of things like self driving cars, which have fairly wireless cameras, off site monitoring for homes or business, medical, wearable medical devices which have sensitive intensive data associated with them. There is lots of potential for things to keep increasing, even when we run out of time, even when everybody in the world is on line using three devices at once.

To wrap up, how am I on time, Michael?

>> MICHAEL OGHIA: You can take a couple minutes.

>> MIKE HAZAS: Good. I hope I made the case that there is arguably that the Internet is perhaps becoming less closely tied to human attention although still quite involved how we use the Internet is very involved and it's data intensity. But certain services have been increasing, in particular social networking and high fidelity content delivery video in particular, and where this background connectivity and more connected things tends to be more common.

This automation, this ongoing trend and as well as the Internet of Things, more broadly, makes that long term limit unclear. I suppose where we got to was we probably agree with Decker's original assertion that perhaps the Internet will only stop when it runs out of energy unless we take a more active role in its shaping.

That shaping of that dynamics is still up for debate. I think we sort of would do well to be part of that. So, thank you.

(applause).

>> MICHAEL OGHIA: Thank you so much, Mike. I really, the whole reason why I asked Mike to be the keynote is to give context to our discussion, because there is obviously a lot of data involved, there is a lot of research involved. And I wanted to make sure that this can inform our discussions. With that said, I want to also quickly mention that I remember when I saw his work and whenever I saw it published on CircleID, I remember seeing that in a interview that he gave with someone, he was saying that eventually, all of our energy could eventually be going to our data and to the Internet. Is that correct? Was that a potential future?

>> MIKE HAZAS: I suppose that's one, surely we will keep some in reserve for transport and heating and cooling. But yeah. It's hard to see how it won't keep rising as a share and it's not clear where that would stop at 90 percent of global energy, 95? It's not clear. So yeah.

>> MICHAEL OGHIA: Thank you so much, Mike. With that said, if anybody wants to speak, please speak into the microphone, because we have a few on line participants. Before I open the floor, I want to give the floor to Marco, who is with RIPE NCC, because he wants to address specifically the fixed component that somebody was asking about.

>> Thank you. As you watched the fixed components big, and that is something that is slowly bubbling up in policy discussions in terms of 5G, is frequency spectrum is a very scarce resource, and especially at the higher frequencies that are still open, the coverage is very low. That means that basically you need a lot of antennas, and those antennas all need a cable. I heard somebody refer to the middle mile now in relation to that.

That could be cause for a rise in sort of what is depicted as fixed wired access, because all of the fibers and all the connections going into that antenna will pull out of energy. While I have the floor, if you allow me to ask a question of Mike, and that is you shared a lot of aggregates or averages on monthly basis. So my question regarding that is, as energy sector people probably know, is what kills your grid is the peak usage.

In that sense, I wonder if you have taken that into account. The second question I want to leave with the panel is consider the actual energy requirements of storing data, the fact that we just park it on a desk and never look at it again, where it used to be on tapes and tapes don't need energy, but these days apparently we store on electronic devices. I wonder if you can, have you included that? Or do you have a opinion on that?

>> MIKE HAZAS: The first question was about the timing of the demand, not just that is a great question. I'm glad you asked it. I didn't have time to cover it.

But in the paper we just presented last month we looked at, it was called demand around the clock. It looked at the timing of demand and timing of demand for different services. The bad news is, as you pointed out, it's all about the peak, and particularly at peak times, the carbon intensity of the grid is very high. Certainly in the UK that is true. The bad news is that the band coincides nicely with peak demand and so that is partially because that is when people are watching video, that is close to prime time in the evening, cold winter nights, firing up video alongside your heating. That is kind of bad. Social networking peaks in the morning, along with transport peaks, so as people go to work, as they wake up, start moving around.

These are linked to human time and how we spend it. Right now, there are things like backups which might run at night, automatic downloads sometimes, although sometimes the downloads run at peaks times which is certainly something addressable. So yeah, not great news there, I'm afraid.

The second question was about storage and mobile devices, is that right? For everyone?

>> Yeah, you can start, but yeah, the fact that we tend to store data on line in data centers and basically never look at it again while it still eats energy.

>> MIKE HAZAS: Yeah. That is a huge problem. I don't know, because a lot of the data centers, where your data is being stored and how that energy is provided for and how it's resourced are generally fairly secret. It's hard to account for it which is why we have to turn to the lifecycle assessment community who do the modeling and analysis for that. But yeah, that is surely a problem because it is going to be sitting on a hard drive somewhere. I don't know how they make decisions about how to sustainably store that, or delete it. That is not transparent at all actually, which would be a huge problem for governing it, I would say.

>> Hello. Luca, I work for the center Internet society. I had two comments, two questions.

The first one I found very interesting, the graph you were showing about increase of data traffic, and early consumption of fixed network.

But I wonder if that is really a problem, because if you have a simultaneous transition to renewable energy, actually even if there is a increase, you would not produce too much waste, because the increase in energy consumption would be say absorbed, or at least let's say the potential negatives would be limited by the fact that you use renewable energies to supply power you need.

So actually that could be mitigated.

So I was wondering if you had considered this, and the second comment is with regard to the community networks, that I don't know if you are aware about the existence of community networks or crowdsource network. We are working a lot on community network within the IGF coalition committee, even produce a report. If you want I will be happy to give you a copy. And what is very interesting, one of the many fascinating elements of community networks is that many of them being built in rural areas, they have also to integrate sustainable energy production tools or mechanisms in order to power the various antennas and towers and routers that are used to provide connectivity to the rural community.

They automatically integrate sustainable energy production into sustainable connectivity production. What I was wondering, if you had already considered this in your research, or if you have already elaborated some recommendations on how to make connectivity and fixed access more sustainable.

>> MIKE HAZAS: Yeah, I suppose to answer your second question, or just now, yeah, it's hard to, it is hard to work out what the way forward is on the sustainable networks. I think, I've done field work in Shetland, there are communities that set up their own networks with fairly advanced kit actually, it is all microwave stuff that sits on top of hills and runs off of wind turbines and solar which is fantastic, great to see that. That was eight households were grouped together, deploy the equipment, run a fiber cable to the nearest access point, so that is great. That seems to work. They have to replace the ball bearings on the wind turbine every six months. But apart from that it's fairly sustainable.

So there are definitely ways forward. I'd be interested to look at the report. Yeah.

Your first question was about renewable energy and the role of that, yeah. I had a similar question at saint Andrews last week when I was visiting there.

I think, I suppose, yeah, if we can guarantee that all the energy that goes into the Internet going forward and its growth is renewable, then I have trouble thinking of a reason to be worried then, I guess, if it truly is renewable. That would be great.

The problem is that, and this does happen with things like data centers, where they will build a geothermal plant next to the datacenter, and then they are good to go or so they claim. But the problem with the Internet is it's so distributed, and once it works its way out beyond that datacenter and goes into other networks and things like content distribution networks and they rely on multiple sites, as well the middle and last mile, those networks, all the energy is difficult to trace, pin down, you are dealing with local peaks in different regions, things like that.

It will be, it's difficult to make the case that we will be able to go all renewable going forward because we don't know where all that data is going to end up and how it is going to be used all the way down. I think we have to be careful there, because the Internet is so global, is so multi regional.

>> My comment was more about given the fact that it is possible to identify what is the percentage of energy consumption produced by connected devices, at least from the graph that was my assumption, one could argue that the same percentage should be supplied with renewable energy, I mean policy wise speaking.

>> MIKE HAZAS: Yep, that's an interesting idea. Can we look at the different paths of energy in the system, and base the policy on that. That is a very interesting idea.

>> Anybody else, do raise your hand. I don't want to eat up all the panel's time. But permission to ruin somebody's day, because that does worry me, and that is in the context of IoT and community networks coming on line, that basically harvest their own energy and that is very sustainable. In terms of IoT we now see what I call self harvesting devices, energy consumption, will continue to run forever. And you started your presentation with this huge big garbage pile of old and used devices.

The problem is that we are spreading that out, because once you glue that sensor in the street and it is harvesting its own energy from people walking over it, its limiting factor is the prediction quality. Sadly we got good at building stuff that last really long, that is going to be there for 50 years. Even when it breaks, who is going to be the one that is going to unglue it from the street.

In that sense, it does worry me that on the energy at least when you get your power from the grid, you pay a price for it. Then you know where it is. But once it sort of triggers its own life, it is really out of control.

(someone speaking off microphone).

>> It might not even because it might be broadcasting data that nobody is interested in listening to, but it still uses the spectrum. That is the other worry there.

>> Thank you, fantastic presentation. I'm interested, because I think it was in 2010 when I attended a rather left leaning summer school on green IT, and the focus there was really on the question, what can consumers do, and how can we save time that we spend on the Internet. But from the perspective of CO2 emissions, so how can we Google one thing less, and I think it goes very much counter the modern lifestyle of myself and most people to think about that.

I see myself just Google something out of, just because I went on my browser and that is the first thing I did when I thought about something. The question is now, do you see any implications for, of this kind of research or of the findings for personal usage, and do you see anyone mobilizing around this? Because when I think of things like the car or the airplane, these were technologies that rolled out and everyone was very happy to get everyone on board. But at some point it switched. People are starting to think you are morally doing something wrong if you drive too much with your car. Will that same thing apply to using the Internet, to using Net Flix at some point?

>> MIKE HAZAS: Yeah, I think I can, yeah, I think working at the user, at the individual level, is quite a challenge. Everyday life gets in the way. That's the thing.

I think our main challenge there is just to raise awareness initially, like you said. These things happen slowly over time. Perhaps people will start searching less, just like at some point they decided to, that it was bad to leave lights on. At some point there might be some sort of voluntary curbing. But I don't think it's entirely realistic and certainly not any time soon, but that we can definitely do a lot to raise awareness.

There are other things that designers and technologists will often, some of the audiences that I talk to, look at, can we do things like not only make the system more efficient for delivering video, using some sort of peer to peer distribution or whatever, but can we also clamp down on fairly intensive connections like Facebook, is actually quite intensive, Facebook and other social networking sites between 20 and 30 percent of Internet traffic globally.

Part of that is because of all the video content. What happens if we clamp down on those connections a little so they consume less. What content gets through? Is it stuff that is meaningful to people? Or does the whole thing just shut down? There is lots of different things we might try, which may be automated ways of slowly reducing demand through software effectively.

>> MICHAEL OGHIA: Unfortunately, I did reach out to Google and Apple, and I couldn't get in touch with Facebook but to see if they could join us, and they were not able to.

But this is potentially a good place to go from here, maybe next year or in the future discussion, to say how can we kind of link these two communities, the Internet Governance with what ICT for sustainability and the private sector to see how could we leverage each other's work to see how to make that better. That is one idea. I want to know if anybody else has any other ideas for that.

>> With regard to the Internet of Things it was mentioned by Marco, where we are connecting almost any device that will have sensors not only to understand to monitor data around the device, but could have, could explore the same sensors to monitor the energy consumption of a device, so if the device is so smart, it should be able to understand how much energy is used and to communicate it, so that could be a very reliable sort of information to research. Yeah. I think that is also maybe some, a piece of discussion that should be integrated into the more general IoT discussion, knowing that things are being connected without any criteria or any analysis of the potential externalities that those things can produce in terms of human right assessment or environmental assessment.

>> MICHAEL OGHIA: It would be great to have product manufacturers here, for instance, Ericsson since they are based in Sweden, or even Chinese company representatives.

Actually, since we just, you mentioned IoTs. We have been talking about this, Mike, something that you actually even started your presentation with is you are only really addressing the energy component. We are not talking about necessarily products themselves.

Did you want to add something? We are not talking about products themselves. Manufacturing a product, where do those components come from? I reached out to Fairphone, I was hoping they could make it. They weren't able to. But Fairphone is a kind of, I guess the best way to describe it is a phone that is modular, so that you can swap out the components, and we know if you want a new camera instead of throwing away the whole phone or getting rid of the whole phone, you just buy the new piece of the component for the camera. There are initiatives out there. But otherwise if we are creating a new device, where is the device coming from?

What materials are they coming from? Also something slightly related, is it coming from a conflict zone? A lot of the materials that go into phones let's say or into other devices that come from conflict zones which are already exacerbating the situation on the ground there. But one thing that somebody had been talking about during the discussion, I'm sorry, I wanted to give context of this specific point I want to bring up, is planned obsolescence.

If you are not familiar with this concept, it's the idea that some, many devices are only supported for two years let's say, and there is, from what I gather and I don't know if you know about this, Mike, but sometimes that two year mark is rather arbitrary. It is not to say that that is the only way. But then that is both at the software and hardware level, even though a phone that is produced can often be carried for many many years longer than that.

So how, I guess one of the discussion points I have a question is, how can we cut down on planned obsolescence? And is labeling devices with their life cycles, would that be an effective way to do it, so that before I buy my nexus 5 here I say that its components are only, one of its components is going to last for five years, can we make the entire phone last for five years then? That sort of thing. Basically, using consumer choice. Does anybody have any opinion about this? It's okay if you don't. Anybody have any thoughts about this? Sure.

>> Hi, I'm Rosy from Romania, I'm here as a environmental educator.

I think that the end user level is very important, as you guys mentioned, and also the personal examples, the planned obsolescence thing could be fought on a personal level but it's very difficult. In order to exchange components in my 7 year old Mac, we have to import screwdrivers set from three countries.

So you are not encouraged to fix your things. But there are very good movements like the one restart project in the UK, I know from the green movement, I'm also a environmental activist, and if you have the opportunity to push the agenda, the research agenda, to push forward the research on this area, to get funding for research, to get reliable data, not only large scale aggregated, maybe not so relevant data for local communities, that is a must. And the fact that this workshop is here now is a good sign that it is moving, the issue is moving from a personal and small scale to a political and policy level.

On the other hand, there is the systemic impact. You cannot have a systemic impact if you don't work on an institutional level, with the big players. But until then, should we wait for Santa Claus or do what we can? My idea is don't wait for Santa Claus. We do what we can. So we educate our students. We educate our families, our peers, to fix their stuff, to reuse some of the ICTs in a way that is not only creative but also functional.

I don't think that we should keep our ICTs, as we keep at home to reuse it sometime in the future, we should push forward local actors and regional actors, because that's the level where you can make a difference, I think as an individual I mean, as an active individual.

And also at the institutional level, who says you cannot do selective I don't know the word in English when you collect waste in a selective way. Collect selectively the waste in an institution, or don't print as much as you could, or to put a policy. I'm a department director, and it's a policy that the equipment have to be switched off, for instance. It's a very small local decision, or that you print as little as possible. So what's such a big deal, that's the personal small level, and this is the bigger level.

>> MICHAEL OGHIA: Thank you very much. We are going to actually come to some of the points that you were making toward, later in the session.

I want to keep it for sure, but does anybody have anything to add about specifically with either a response to her point, or with something about planned obsolescence and the importance of, I guess of ensuring our, how we are connecting to the Internet are recyclable or manufactured in a way that are sustainable.

>> Yeah, I got something to say about that. I'm from Netherlands. We are throwing too much the fastest way which are still functioning quite well, but the thing is hardware and software.

Now the software manufacturers are spending 95 percent of the time on supporting devices that are older than the newest.

So for innovation that's really, yeah, not good.

But as consumer, we want to have the new devices, we want the new software. But the new software does require a better hardware. So maybe there should be something in between, that we won't get an update that requires all the hard stuff, but only some of the little updates. You know what I mean?

>> MICHAEL OGHIA: I do know what you mean. Marco, did you want to speak or Chris, did you want to say something as well?

>> Maybe I'm less worried about the consumer electronics, it's no worse than regulation but we fix the bottle system by putting a deposit on it in many countries. I personally don't see why I can't leave a deposit when I buy a phone, that this is an incentive for me to bring the phone back when I buy the next one. In that sense, I think that chain, we have got prior experience, we know how to control it. We know how to incentivize it.

What I'm more worried about is the collective and institutionalized systems, the stuff that nobody really owns. That's really where a lot of the IoT devices come in, the cables we put into the ground, 50, 60 years, fiber, the theoretical life span of a fiber is 40 years. Most of the fibers went in there sort of late '90s, early 2000s. In five, ten years time we are going to decommission the first few trunks probably.

And I don't know. I haven't heard anybody on how to dig them up or any smart way of unplugging a fiber. The stupid thing is if they have incentive to dig it up, it's worth money. You can still make a buck out of it.

Fiberglass, I haven't found a market for it yet. There is no incentive there, and also with the burst of the bubble a lot of the original companies who put it in went bust. Something to think about, I don't have a solution for that one. The phone is the easy part.

>> MIKE HAZAS: On the theme of better citizen or user awareness and engagement, you were talking about planned obsolescence, we could envision an energy or kind of a eco labeling basically for devices, why not, right? There might be some things which are C and some things which are A and A plus. As time goes on, as it has with refrigerators, the goalposts will move, where then you have to have an A plus plus and A plus plus plus. But right now we have nothing. You go and you buy a phone and buy it without considering maybe that it was, materials had to be extracted from the ground and coal had to be burned in the far east in order to build it, right?

That is one thing.

You can also envision eco labeling for software. You could look at films that you might stream that might have a eco label. 4K film which takes 7 plus megabits per second might have a eco label of a D or perhaps a C, because you might watch the same film in super HD, HD or even center definition and the rates would go up to be A plus. Again, at least, of course people will still pick the 4K if they have a big display that is capable of displaying they will pick that of course, but at least there is awareness there.

>> The last point, the market provides incentives there because when I go to Apple iTunes the SD version is cheaper than the HD version. There is a incentive there. Actually on streaming servers Net Flix, Net Flix will automatically scale down if they don't think they have the bandwidth to deliver. I can easily see that, I don't have the energy to deliver to let's scale down.

In my phone, if I go to general and there is a regulatory tab, somewhere in the middle is at least a icon that says I should not throw it into the general waste. There is labeling. It is hidden, but there is a little sign here saying I shouldn't just toss it away.

>> MIKE HAZAS: Similarly we can label apps. You might have Facebook might have a certain rating or sort of maybe a instant messaging client base, add jabber might have a certain rating as well. You can envision ratings for apps, not just media.

>> From electronic frontier, Finland. I have a case for you. I bought a TV, some six or seven years ago. After the first year, pretty much one week after the guarantee had finished, the TV broke. And the reason was a blown capacitor in the tele. I did a Google on that problem. There is 1.9 million hits on that with Samsung.

I was wondering, the first time, luckily in Finland we got pretty good consumer protection there. They still agreed to fix the TV for no extra cost, although I did have to pay for the transportation there and back which was like 50 Euros. We are talking about a part which cost about one Euro.

I did that. They fixed it. The same thing, the same symptoms, again like one year after. I was like, this can't be happening. It's got the same fault that was happening, and I did a search and I found out that, yeah, a lot of people have exactly the same problem with exactly the same tele.

Then I had a friend who is a bit more of a electrical genius than I am, and so he was just like sort of watching over what I did, it was really easy. I changed the part. Now it's worked six years at this point. I just gave, the capacitor could take a little more power in there, that is all I had to do and it worked six years.

My question is, can you think of any kind of regulatory or some other kind of solution that would actually prevent companies from doing this planned obsolescence? I think the example of this tele is exactly that. They put in a component that will blow up in about one year's time.

>> MICHAEL OGHIA: If I can quickly answer that, one of the organizing team members I've been put in touch with said how amazing would it be if there could be a regulation out there, that says something to the extent of a device must be supported until its lowest, to its until its least component is no longer able to be maintained, something like that. Basically instead of having a component that lasts for one year and then the other components last for three years, imagine if everything else could just be for three years.

That could be at the regulatory level. I don't know how realistic that is. But at the same time, again it would be, it would have been incredible, for instance, if somebody, I've been in touch with the EC and there is somebody at the EC that is actually quite interested in these issues. Unfortunately, again, he couldn't make it to Tallinn.

There is definitely room for discussion about this, I would say. There is room to discuss it with the regulators, specifically in collaboration with the private sector, etcetera.

But that is just one little bit, I wanted to go to Mike real quick. Did you have something you wanted to say? I thought you were raising your hand.

>> No.

>> MICHAEL OGHIA: Sorry. Marco.

>> Would you ever buy television again?

>> It lasted fine since I repair it myself.

(off microphone).

>> It has come up in our workshop yesterday, things are replaceable, as long as consumers have a choice within the market, how to correct these kinds of things.

>> MICHAEL OGHIA: There is also the concept that the circular economy, this is becoming more, this is growing in general, it's the idea that whenever something is being designed, then ultimately produced, it should not right now our current time line is that something is designed, it's produced and then thrown away, and then what happens to it? Who knows. The circular economy, it's designed, it's produced and then like a circle it goes back once it's no longer used, it gets recycled, its components go back whatever is recycled is recycled, used again, or it's melted down.

The idea is that it creates, so even in our manufacturing processes, there is a significant issue in the sense of how are we approaching it. Are we approaching it with the sustainability in mind? Or are we approaching it purely from a, let's say the bottom line perspective, and how many units can we ship, how many, how much can we make, for instance.

Quickly, I wanted to go to you and then Luca.

>> When I think of companies forced to support technology a bit longer, my first thought is it might block the innovation part. My second thought, if I look on the iPhone which changes every year with some minor improvements, if they are forced to hold them two years, we have, well, not the iPhone 8 but we have the iPhone 7, then iPhone 8 two years later for the innovation part, we called it from the costing part more efficient to create update with the resources on all components, than to have those short upgrades just to get the people to buy new stuff, because it's marketing. We in Luxembourg have one institute called Digital Inclusion, and you can send your old devices, computers, smart phones, laptops, TVs. They will be repaired by refugees. They teach the IT and administration staff. Then they were sent to refugees to get access to the Internet, and then we have cooperation with the local training who provide on line training, so they can in the time they are not able to get a job or not allowed to get a job, they can learn the language, to learn some staff, some teams. So it's something that could be more global and European, provide good initiative.

>> Win, win, win, win, there is two other wins in there. Across the board, you are helping everywhere. Luca, did you have something to add?

>> Two questions actually, I want to ask Mike, because I was thinking, hearing discussion, is that many problems could be solved with some corporate social responsibility for entrepreneurs that could be, would have a obligation to recycle at least the components of the product that they sell.

I want to ask if you have ever tried to elaborate some recommendations to this extent, and if you ever had the possibility to speak with policymakers and what was their reaction in case this happened.

The second comment on the edge of my mind when I thought the second time about the graph on Internet traffic, is that I was wondering if you have ever considered the consumption, both the energy consumption, traffic consumption of blockchain, because and if not, that could be a very good path for research, because we all speak about blockchain as the solution for every problem we have in the world. But we always forget that to create every block of the chain, you need a lot of nodes, consumes a crazy amount of energy and of bandwidth.

I was wondering if you have already considered this in your research.

>> For the record, Luca, I did bring that up yesterday as a point in the blockchain session.

>> MIKE HAZAS: Yeah, thanks for the questions.

In short, the answer to both of them is no. The longer answer is, I haven't really been able to interact with basically policies stakeholders. That is one reason I've come here.

So not really, although, it varies by country, sort of the, in many sort of I guess European and in Canada and the United States, you can go to the manufacturer and insist they recycle it and I think they have to. If you take it to them, there Is no incentive for you to do so, but if you take it to them there is a obligation there. It varies. That is the trick, it varies globally. There is no unified strategy, no unified way of getting them to recycle it even though in many places they do have to, if you take it to the right place if you know where the place is. That is something to think more about.

The blockchain thing is interesting. I haven't looked at it yet. I'm involved in a project starting in August with the university of Edinburgh, I keep teasing the principal investigator there, because it's using blockchain to dictate transactions for solar generation in Sub Saharan Africa. I keep telling him that the blockchain is going to eat up all the energy that they generate. But yeah.

I'll be interested to see how that works. The longer blockchain goes on, the more complex it becomes, the more energy it consumes. Hopefully for this particular solar project it will be small enough that it won't actually eat all the energy. But there is a broader point that I wanted to touch on, briefly.

We have been talking about awareness and eco labeling and end user awareness in particular, citizen awareness. But I want to hear the reactions of the people here, whether kind of, on the note of corporate responsibility, social responsibility. In a way, there is no, well, one of the phrases we throw around in the demand center at Lancaster is that there is no policy which is not also energy policy because in the end policy is about the allocation of resources in society and so it's hard to make policy which doesn't move energy around or justify expenditure in some way or not. Having a lack of policy on Internet and its growth is kind of basically we are hands off in terms of the energy, and that is why, that is one reason anyway you see it doing this. The other reason is that people are spending more time on line and more devices. Again that is not necessarily a given. It doesn't have to be that way.

What I wanted to throw out was, does it make sense, is it realistic to engage with companies? You might say with the explosion of video, whether it's advertising or user content on social networking, that is the reason we have seen it rise to 20 to 30 percent of global traffic, for example, but if it had stayed text and picture based, we would probably have a different state of affairs right now, for example.

Another thing is, there are certain checks in place, as you are pointing out, with video, depends on screen size, depends on bandwidth, and in particular 4K subscription to Net Flix costs more. That surely will be some sort of cap. It doesn't cost much more but at some point it might be free. Is there any mileage in engaging companies to justify decisions that they are making on these huge global scales regarding millions of devices, and whether, does it make sense to ascent to justify the extra energy, for example a software update which increases the video fidelity or video contents, does that even make sense? I don't know. That would be a responsibility thing, because when they roll out that update, it raises, it does, it has raised energy demand during peak times all over the world.

>> MICHAEL OGHIA: Chris and then ...

>> Thanks, Michael. I've sort of had some thoughts here which I've been actually holding back for the session, partly because I was thinking what is the connection, how do we substantiate the connection with all this to Internet Governance, this process we have. Part of why I was holding back is actually that there has been a lot of discussion here today of quite practical measures that can be taken, which I think was one fear that maybe wouldn't, and what I'm going to do is going much more out to the sort of paradigm sort of level, and I think that your question there cuts to that a little bit.

I think one thing that is clear with this discussion relating to Internet Governance is that what Internet Governance often means is not sort of Internet policy, it's going to relate to all sorts of policies. It's going to relate to manufacturing policy, energy policy, going to relate to corporate policy, all these things that the Internet affects everything it touches on but are not necessarily what you think of as Internet Governance, but Internet Governance has an opportunity for us to have input there, which is valuable and useful.

But I think then we need to think of it in terms of an Internet Governance paradigm or Internet paradigm even, and one of the things I think the cliches talking about the Internet in whatever context is marvel at how quickly it's grown. Presentations start off with, it was only 30 years ago that the Internet was this size and now it's this size and everything is up and to the right and the scale is incredible. That is the paradigm of the Internet, that it's, even now we talk about Internet for development is that development, how do we get to the next billion, the final billion. It's all that growing this technology.

There has to be a limit to that, I think. We have to at some point turn around and say, maybe it's not about Internet Governance for development but Internet Governance for sustainability, how we treat this technology in a way that we are going to be able to continue to have it without destroying the planet, without destroying anything.

To go in a slightly pessimistic way, even from some of the discussions in different sessions here, we have seen signs of what you could argue are that we have already moved into that unsustainable phase. For example, we talk about the IoT and talk about the fact that there is this sort of problematic cycle, where customers want to pay less because they are tiny devices and you don't want to pay for them. Manufacturers see the possibility to spend less on manufacturing, by cutting corners on security and they don't see a down side to that. There is this unvirtuous cycle, which continues.

But that is an example of not a sustainable model, still very much a growth model. Another example for IP addressing, we are looking at, technology is looking at managing the growth of the number of users, the number of applications using things by putting more users on a single IP address. What are the down sides to that? We had law enforcement here today telling us exactly what the down sides of that are for them.

A model that allows for growth but maybe doesn't allow for sustain ability. Perhaps what we need to do at a high level Internet Governance paradigm is to change that and start to say, as we look at what the outputs of Internet Governance would be, engaging with policymakers, engaging with corporate users, maybe part of what we need to focus on is how do we actually build in sustainability as a idea to what people are doing and how it rules the Internet.

>> MICHAEL OGHIA: I can say, Chris, that is exactly what my long term goal is with something like this, is how can we make this the thing is though, first of all, two things. The first is that, for instance, I think the IGF theme last year was on sustainable and inclusive growth or something. Nobody talked about energy at all. I was just like pulling my hair out, what little is left, and thinking to myself, why aren't we talking about energy?

I felt like I was becoming almost fervent at one point, you are talking about sustainable growth but there is no energy. But at the same time if we look at other aspects especially when it comes to development, I'm looking at this side and not this side, sorry, but especially when it comes to development, I know Luca for instance community networking, they know exactly why energy is so important. A lot of the community networks that exist around the world they are in underserved areas, in remote regions, they are in places that are very far from population centers. A lot of these places, for instance, that Luca and the colleagues that are involved with these, with this initiative, they are going into centers that don't necessarily have grid electricity. How are we supposed to connect people to the Internet, if we don't even have the energy to power the routers or satellite connectivity that we have or the radio frequencies or the spectrum or the TV white space, for instance. All these are technologies that are used to connect remote regions.

How, they know the importance of that, what I hope is that especially us that are not so removed, I'm sorry, that are very removed from these rural and remote regions, how can we incorporate that into our work, to catalyze this paradigm shift so that something like energy, we are not struggling to incorporate into the discourse but is something that is a given, something that is a given. Then Alka, your point as well.

>> A quick response. I think one thing I would want to make the point is that I think the development paradigm is still very important. There is certainly reason to look at how do we go to this equitably, make sure everybody gets the benefits from that. That is one thing.

To give it a more optimistic spin, there are areas where there is still, and the example of spectrum is one, the ITU and the work that is going on in spectrum there, that is a sort of environment, community, whatever, a field, where they are very much aware of the limits of their resource, and they have history and experience in how do we use this, how do we re claim it, how do we make sure it's, the whole thing about white space now going on is an example of that.

But again, the IoT thing is perhaps also throwing a spinner into that process where spectrum use is being used by more things, other existing processes, something like the ITU has is going to be able to scale and scale quickly enough to be able to continue to use the spectrum there in a sustainable way.

>> Alka, Netherlands. I want to throw up another problem, because I just read an article about that every two years we create as much data as we did forever. I also read the same story in 2010, and then it was 7 years.

So it's grown as you have said but I guess, it's just a assumption, that also has to do with backups, because we are making them back up and after that making backup after backup. How can we as, how can private and public sector solve that problem? That is a question for the panel.

>> There is no panel.

>> Okay, for this round table.

>> MICHAEL OGHIA: If nobody has a direct point to make about that, that is a good point. What it amounts to is that that could continue to add, continue to be a footnote on the discussions. There are these outstanding questions that we don't necessarily have answers to right now.

I'm going to ask Elizabeth if you can start creating the messages that we can create for the session. As you are doing that, maybe the last question that we can get to today, I'm really excited about the fact that there is so much to talk about, but the last question that I had is, and Chris broached the issue, but it's how can the Internet Governance community specifically do more? What do we need to do in order to start to make that paradigm shift happen, because I like getting up and talking about sustainabilities and maybe it's quaint or endearing for a little bit. But after a while, somebody might start throwing something at me. How can we make this something that other people are buying into?

>> Primarily continue the discussion here, I think this is especially with things like IGF or EuroDIG, it's good. We are certainly running out of time so I'll leave that for next session, I guess, is also, a bit of Chris's point and people committing, it's also to recycle responsible. As people brought up like spectrum, standards 20 years ago was written with a 20 year life span. Modern devices are far more energy efficient. We don't switch off the item. There is still a handful of people using a old fashioned phone. That is the dilemma that also then on the technology part you are faced, because if we say we can be much more energy efficient and switch off, everything is going to be 4G, then you have, you are wasting valuable resources and making the device unusable. We saw the same with television switch off and switch overs. Especially it is not as much the Internet but in terms of multistakeholder dialogue, you have to look at all sides of this.

It is going to take time because everybody has their own little thing and of course, nobody wants to waste money or waste energy in things that they don't see a direct return on. I think that is the biggest nut to crack. A lot of this is future work. There is no direct return.

>> MICHAEL OGHIA: Getting more stakeholders around the table that can address this and also inform us about what they are doing to address these problems and these challenges, I think is also quite important. But yeah, quickly, Luca.

>> Need perspective I would argue to exploit what already exists. There was convention in Europe that defines not only the right of the general public to be informed and to participate to discussion, but also foresees the creation of institution, I mean multistakeholder institution that, Internet Governance environment, are to a great extent modeled on environmental governance. This multistakeholder is not invention of Internet Governance. It already exists in the environment.

>> MICHAEL OGHIA: For sure. If we are going to address this, it is going to have to be multistakeholder.

>> There are already national agencies and processes that are built on this format. I think very good idea would be maybe to include in the debate those agencies starting for the next debate you organise, Mike.

>> MICHAEL OGHIA: Thank you very much, Luca. Elizabeth, I'm going to hand it over to you to give us our messages.

>> I have come up with four broad points. First one is, as data traffic goes up with a increase in IT technologies and machine to machine traffic, so does energy consumption and the Internet account for relevant share of global energy consumption in the future. There has to be an effort made to monitor and address the rise in a sustainable way.

The second point, planned obsolescence, list a lot of eWaste, incentive for consumers to invest in more sustainable devices and this might be used for software in a similar way. Regulatory level, switch to circular economy model should be made a policy goal.

Next up, sustainability as a manifest on a societal level and Internet will gain even more significance as more people are connected to it. Projects and policies that address ecological issues and find a solution that involves sustainable social benefits are to be put forward.

My last point will be Internet Governance processes should acknowledge a responsibility to incorporate sustainability in discussions.

>> MICHAEL OGHIA: Does anybody disagree with those? Are those all okay based on consensus? Okay. Wonderful. Thank you very much, Elizabeth. I want to say first of all, I want to thank the tech team for all of your help with this. I want to of course, this is something that is not done very often, but I don't know who is captioning, but could you just write your name?

(hi there, I'm Mary Kay).

>> MICHAEL OGHIA: Hi, Mary Kay. Thank you very much for captioning this session.

(It's my pleasure!)

You are working remotely, but I really appreciate your work as well. Thank you.

Of course, thank you to Dan for remote moderating, to Elizabeth for being our rapporteur. Of course, so much to Mike for coming all the way here to Tallinn and being a part of this, and hopefully creating more bridges between the work that you do in your networks with what we do.

And especially, I know, that's okay, especially I really, absolutely, I can't thank all of you enough for coming to this. This means so much to me. I really want to continue to push for these, for this topic and advocate for this. Thank you for your support, and let's try and work even harder to see how we can continue to engage in this discussions, engage with new stakeholders, with the stakeholders that need to be at the table as well, and see how we can continue to make the Internet more sustainable.

Thank you all so much.

(applause).


This text is being provided in a realtime format. Communication Access Realtime Translation (CART) or captioning are provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings.

Session Twitter hashtags

Hashtags: #eurodig17 #DrowningInData