The need for a new norm: Internet impact assessments – Bigstage 05 2022

From EuroDIG Wiki
Jump to navigation Jump to search

21 June 2022 | Start 13:35 CEST | SISSA Main Auditorium | Video recording | Transcript
BigStage 2022 overview

Session teaser

In 2020 the Internet Society released the Internet Impact Assessment Toolkit to support policymakers, technologists, and other Internet users and advocates to assess the implications of change – whether those are policy interventions or new technologies. Since then the Toolkit has been further developed with a new analytical lens to help assess how new technologies or policies may strengthen or weaken what the Internet needs to thrive. This presentation provides an overview of the toolkit and a presentation of one of the toolkit’s latest additions: Internet Impact Briefs.

Session description

Like any live ecosystem, the Internet is constantly evolving. This perpetual evolution without a centralized plan or control – but with thousands of people and organizations working collaboratively on standards, protocols, and their application in the real world – is what has made the Internet a success. But the fact that the Internet resembles an ecosystem, with complex relationships, also makes it hard to protect. How can we identify the policies or technologies that strengthen rather than harm the Internet?

This is why the Internet Society created the Internet Impact Assessment Toolkit (IIAT).

The Internet Impact Assessment Toolkit uses two technical papers as a framework for analysing impact on the Internet. The first describes the critical properties the Internet needs to exist, and the second describes the enablers that help it thrive as open, globally connected, secure and trustworthy resource.

The goal is to inspire the Internet community to champion a new norm, similar to how policy decisions are informed in the environmental field: to conduct impact assessments with an eye towards what the Internet needs to exist and thrive.


The format of this session is a traditional presentation.

Further reading



  • Carl Gahnberg

Carl Gahnberg is the Director of Policy Development and Research at the Internet Society (ISOC), where he is focused on issues related to Internet governance. In this role, Carl contributes the organization’s global policy development and its partnerships with international and regional organizations, engaging with global policy makers and non-governmental stakeholders on key Internet issues.

Video record


Provided by: Caption First, Inc., P.O. Box 3066, Monument, CO 80132, Phone: +001-719-482-9835,

This text, document, or file is based on live transcription. Communication Access Realtime Translation (CART), captioning, and/or live transcription are provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings. This text, document, or file is not to be distributed or used in any way that may violate copyright law.

>> Hello, everyone. I will be taking over for Nadia.

The need for a new norm, Internet impact assessments, which will be held by the Director of policy development and research of the Internet Society, Carl Gahnberg, he will be joining us online.

When you’re ready, the floor is yours.

>> CARL GAHNBERG: Thank you very much. Happy to be here. (Echoing). A big thank you to the organizer force inviting me it be online and to do this presentation online which is really super.

>> (Echoing).

(Audio quality too poor for captioning).

>> Can you speak a little bit more so we can hear?


>> CARL GAHNBERG: We’re back into the presentation.

As I mentioned, our idea of a solution to this, was that inspiration from the environmental field and the idea of doing impact assessments. So the idea today is to give you an overview of the toolkit that we created where we produced a framework for how you could do Internet impact assessments and then to walk you through one of the analysis that we have done, what we call Internet impact brief where is we use this framework to assess how a new policy could affect the Internet.

Starting with the Internet imPCT assessment toolkit, this is something that you could find on the website, it basically has three many components to the toolkit. The first one, it is what you would think about as the baseline, the framework that you use in your analysis, and I’ll speak more about those documents in greater detail, but those two documents that we produced are effectively trying to describe what the Internet needs to exist or important properties of the global network of networks and then the second document describes what the Internet really needs to thrive, how do we achieve those goals around an open, secure, globally trustworthy Internet.

It includes the methodology that’s the formal guide that helps the user to do these impact assessments themselves as well as a collection of assessments from both ourselves and from our community and partners that we’re collecting in a repository and we call these short analysis Internet impact briefs. Starting with going through this framework we have created, I wanted to start with the first document that you see on the screen to the upper left that we call the Internet way of networking. This was a document that we produced in 2020 and the goal there was really to try to figure out how can we describe the Internet in a useful way that allows you to assess change.

Many I’m sure are familiar with this description of the Internet as network of networks and we 100% agree, we figure it requires more detail if this it is to be useful in Internet impact assessments.

In consultation with our community and other experts, we set out to add a bit more detail to that description of the Internet as network of networks and identified what we called five critical properties of the Internet. In order for the networking model to be coherent, what we think about the Internet, it would have these five critical properties. For instance, when you’re doing networking, the Internet way of networking, striving to have a common protocol, striving to have a common global identifier system, for instance, DNS, IP addresses, you have a layered architecture with interoperable building block, open standards, all of those things put together and the additional details to the description of the Internet’s network of networks.

In this document, we describe these things in the ideal form and I’ll explain a bit more later on why we describe it in the ideal and the purpose is to really find the good reference point for where are we going what, should we strive for and for instance, many of you will say that the Internet actually has two protocols today and we have IpV4 and 6 and the current Internet doesn’t necessarily correspond to the ideal in all of its parts.

The important thing, when you do the networking the Internet way or striving to have a common protocol, our not – you don’t find it sufficient to have two protocol, you want to have one protocol, a column protocol within that document, what it gives you, what the Internet needs to exist.

2021 we talked about what the Internet needed to thrive. This is where we again went to our community and other experts and produced a document, enablers of a globally connected, secure and trustworthy Internet.

The approach is similar, but basically we take a high-level goal like an open Internet and try to break it down into its different parts, what are the different things that go into an open Internet. For instance, in an open Internet, you have the unrestricted access, you have the collaborative development and management in governance, one of the reasons for breaking the high-level concepts down like this, it allows you to get more nuance and analysis so that you could have a policy that might affect the easy, unrestricted access and may not have any impact on collaborative governance.

Those are the things we want to be able to tease out and describe how something may impact the Internet’s openness, its security, its trustworthiness, for instance.

Looking at the enabler, see what it looks like, we assume this enabler, easy and unrestricted access, now the description on the screen that is trying to present what that enabler looks like when it is in its ideal form, namely that it is easy to become part of the Internet, Internet connections are affordable, that Internet services are accessible and that there aren’t any unnecessary regulatory or commercial barriers. In this document, we describe those enablers and how they contribute to an open Internet and we’ll provide samples of things that may help us get closer to that ideal description. Things that move us a I way.

For instance, in the examples that we provide that point to the web accessibility guidelines, that’s something that’s a positive for the enabler, for instance those guidelines helps support screen readers for people with disabilities so that they can have greater access to online services. That’s an example of something that’s strengthening this enabler.

Likewise, we have examples of things that may move us away from the ideal, such as monopolies in the creation of Internet Act that could reduce affordability.

That’s a quick overview to how we describe these things and the rational for presenting them in this ideal form.

When you put this to work, it may sound cumbersome to do impact assessment, but what it actually turns out to be in practice, it is almost like a checklist.

This ideal description of thinking of how this work, you take a given policy, a technology, and they would – you would ask yourself a question like the one you see on the screen here, does this policy move us closer or further away from the ideal state of the critical property or enabler.

You would ask that question vis-a-vis the critical properties and each of the enablers and then once you see which direction that it is heading, you can provide an analysis as to why you see the direction going in the positive or negative direction and then you summarize this more collaborative report, what we call the Internet impact briefs.

It may sound cumbersome but it is really a checklist going through, thinking through the effects of the critical properties and enablers.

Finally, I wanted to provide you also with an overview of what this looks like in practice and I will straight this with one of the impact brief was published recently around interconnection rules in South Korea. This is an analysis that we published in mid-May, written by myself and the colleagues in the regions, other experts in the regions that contributed input it analyzing how the rules could affect the Internet.

To give you the short overview of what this is about, effectively there is a series of enacted, proposed revisions to the Telecommunication Communications Business Act in South Korea that spell out new interconnection rules for service providers and in this analysis we find that these new rules creates unnecessary costs and bottlenecks in South Korea’s ecosystem as well as increasing the risk of market concentration and dominance., that’s a summary overview of what we find in this brief. I wanted to take you through what it looks like and hopefully inspire you to read more of the full brief online.

In all of these briefs, we try to start with providing the relevant context for the analysis.

In this case, it is a matter of new interconnection rules, therefore we start the brief, we just describe how interconnection happens in the Internet, with the norms surrounding the Internet engagement. We present that as effectively there, as there are three forms of business arrangements on the Internet. You have first of all transient agreement, networks pay another network to pay others to carry traffic to the other parts of the Internet and we have pairing arrangement where they exchange network between customers coming in a form of peering, where the networks agree to have a compensation for when a traffic is being sent, but most often it is in the form of settlement free, 99% of all of the interconnection agreements are sediment free where the networks exchange traffic without any settlements between them.

The brief is provided in an important context to describe the environment in which these rules are then introduced.

I won’t go through the technical details of the policies that invest seen in South Korea, but rather summarize what they translate to, into, in terms of effect. The first one, is that the rules affectively mandate the practice of pay to peer and that needs to be put in contrast to the norms of Internet, 99% are sediment free but here there is a requirement to have paid pairing between the networks. The rules impose quality of service requirements from content providers and the rules are a bit ambiguous and could apply to service providers that don’t even have a direct relationship with the Korean ISP. Fine will he, these may not have a direct relationship with the Korean ISP, but those are effectively what the policies turn out to do in the environment and then we look at what impact this has on the Internet when we look at the impact vis-a-vis critical properties and enablers. This is a summary of what we found in the analyses, and as you see, there is a lot of red, the white is to indicate that we didn’t see any impact.

I won’t go through exactly the rational for each of these impacts described but rather I figure I would give you a snapshot of one of the critical properties impact and one of the enablers.

Looking at one of the critical properties we saw affected was this critical property around the general purpose and technology neutral network. In order to understand how we saw that impact, it is used to reiterate this ideal state of the critical property we described. In that ideal, the Internet is a general purpose network but it is also agnostic with regard to the connect neutrality of each data package. It is not optimized for any specific traffic characters. The real benefit with this, it allows innovators to innovate, pursue innovations without any permissions from other parts of the network and to do so knowing that the Internet has the benefits and draw backs.

The impact we see from the rules, it is that first of all, that quality and service quality gets conditioned on contracting with ISP, secondly that it has moved away from the agnostic networking approach and instead it is conditioned by regulation and to have contracts across the global network and in this regard we have also described that this poses a risk of fragmentation where users cannot access online services that don’t already have a contract with localized people.

To put it in clear example, let’s say you have an online service provider in Europe, they have not contracted with a Korean ISP, that Korean ISP could drop the packets and end user in Korea would not be able to access that service until European service contracts with the Korean ISP.

Then, we then go and look at one of the enablers. This is an enabler that was described previously, and one of these enablers that we call ease and unrestricted access, it talks about Internet connections should be affordable, services accessible, that there shouldn’t be any regulatory or unnecessary regulatory commercial barriers. The impact that we see from the rules is first of all that it effectively increases the South Korean reliance on transit links and desensitizes the arrangements in the ISP and a consequence of this great increased reliance on transit links is first of all that traffic within South Korea even goes on international transit links, accompanied by higher costs and higher latency for the user and has detrimental effect for the entrants in South Korea where a new application or service provider, as they grow, they necessarily node to increase the traffic costs as well and making them the victim of their own success.

This is an illustration of how we think about this impact assessment, how we use these ideal de descriptions of the world and then look at the different policies, how they may move news different directions either moving us closer to the ideal or away from the ideal.

I would invite everyone to have a closer look at both this report that we wrote for South Korea and also have a look at some of the other impact briefs that we published together with partners and community members. I think we’re up to around 16 at the moment on the website. We’re also inviting these are interested and may have a policy, a technology that we would like to put through our assessment framework to do so and also invite you to contribute that to the repository that we’re building with the impact assessments and that we’re hoping to have a global nature and really help share practices across the globe and insights across the globe around the impact of different technologies and policies.

With that, thank you all very much for listening. I put my email at the bottom of the presentation there if you want to get in contact with me. I’m happy as well to stay on and to answer any questions. Thank you.

>> CHAIR: Thank you. Any questions for Carl in the room at this moment? No? Thank you very much.