SR2 Communications

irl@sr2.uk

mallory.knodel@nyu.edu

internet Research Lab

gurshabad@irl.works

safety measurement monitoring research Internet measurement is important to researchers from industry, academia and civil society. While measurement of the internet can give insight into the functioning and usage of the Internet, it can present risks to users, such as breaches of their privacy, legal consequences, or disruption of services they rely on. This document describes briefly those risks. It also outlines considerations for researchers to reference when designing internet measurements to ensure that those measurements can be carried out with user safety as a priority. This document is a draft. It is not an IETF product. It does not propose a standard. Comments are solicited and should be addressed to the research group's mailing list at pearg@irtf.org and/or the author(s). The sources for this draft are at: https://github.com/IRTF-PEARG/draft-safe-internet-measurement

Measurement of the internet provides important insights and is a growing area of research. Similarly the internet plays a role in enhancing research methods of different kinds. Performing research using the Internet, as opposed to an isolated testbed or simulation platform, means that experiments co-exist in a space with other services and end users. Researchers using the Internet as part of a scientific experiment include academic, industry and civil society researchers. This document outlines considerations for such researchers so they may mitigate risks to the safety of users and other affected parties. It is common today for certain internet measurements to require an ethics review before gaining approval and to report details of that review in the resulting work. Previous work in the area of ethical concerns includes the Menlo Report and its companion document , ethical concerns for censorship measurement , and as an example of putting ethical guidelines into practice in ethics review. Many organizations publish specific guidelines around safe measurements, e.g. . The measurement landscape and community is wide and varied. The specific context for Internet measurements should be taken into account by researchers/reviewers and applied appropriately. This document does not attempt to supersede any recommendations such as those cited above, but rather to act as a point of reference for high level considerations with particular emphasis on user consent and data collection. In the growing area of research that includes internet measurement we see this work as part of a larger effort to better equip review boards to evaluate internet measurement methods. Future work may specialize these considerations for specific types or contexts of internet measurement.

The document contains considerations for how to measure the internet while prioritizing user safety. The considerations are particularly germane to projects that involve the generation, collection and/or analysis of traffic from humans. When performing research on a platform shared with live traffic from users, that research can be considered safe only if the risk of danger or injury to users and other affected parties arising due to the research, now or in the future, is no more than minimal or, where it is greater, has been accepted through informed consent by those who bear it (). A measurement poses minimal risk when the probability and magnitude of harm anticipated from it are not greater than those ordinarily encountered in everyday use of the internet. Avoiding harm is necessary but not sufficient: safe research also requires that any remaining risks are justified by the benefits of the research (), that users have given consent appropriate to those risks (), and that researchers are transparent and accountable for how the research is conducted ().

• The considerations presented here are not an exhaustive list. Depending on the measurement there may be other factors that need to be taken into account when evaluating the safety of the measurement.
• Following the considerations contained within this document are not a substitute for institutional ethics review processes, although these considerations could help to inform that process.
• Similarly, these considerations are not legal advice. Local laws must be considered before starting any experiment that could have adverse impacts on user safety.
• The scope of this document is restricted to considerations that mitigate exposure to risks to user safety when measuring properties of the internet: the network, its constituent hosts and links, or user traffic.

Threat model: A threat is a potential for a security violation, which exists when there is a circumstance, capability, action, or event that could breach security and cause harm . User: For the purpose of this document, an internet user is an individual or organisation who uses the internet to communicate, including those whose data is carried in communications over the internet. In certain contexts, users also include individuals who are not the direct subject of the measurement, for example a contact whose data appears in a measured communication. Affected party: Users, and any other individual or organisation that may be harmed by a measurement, such as the operators of infrastructure that carries, receives or responds to measurement traffic. Active measurement: Active measurements generate or modify traffic. Passive measurement: Passive measurements involve the observation of existing traffic without active intervention. On/off-path: A measurement that is on-path happens on the network. Off-path indicates activity in a side-channel, end-point or at other points where the user, their connection, or their data can be accessed. One-/two-ended: A one-ended measurement is like a probe or a trace. A measurement with two-ended control requires the cooperation of both endpoints, which might include the network itself if that is the measurement target.

Any conceivable internet measurement study might have an impact on an internet user's safety, including users who are not subjects of the study, and on other affected parties such as infrastructure operators. The measurement of generated traffic may also lead to insights into other users' traffic indirectly as well. It is always necessary to consider the best approach to mitigate the impact of measurements, and to balance the risks of measurements against the benefits to impacted users. Some possible ways in which users and other affected parties can be harmed as a result of an internet measurement study: Breach of privacy: User privacy can be affected or violated in many ways. The first consideration is at the stage of data collection itself. First-order data, such as name, can distinguish a person. Second-order data, such as IP address, can be used to track behaviour. The second privacy risk is an internet user's data being shared beyond that for which a user had given consent, through accidental, malicious or compelled disclosure. A third privacy consideration relates to the risk of re-identification or de-anonymization of obfuscated or masked user data. These risks include data about users other than the study's subjects, such as parties to a measured communication. Inadequate data protection: A scenario where data, either in transit or at rest, lacks sufficient protection from disclosure. Failure to meet user expectations for data protection is a concern, even if it does not result in unauthorized access to the data. This includes cases of improper access control (i.e. people having access to user data who do not need it). Traffic generation: A scenario where undue traffic is generated to traverse the internet. Traffic modification: A scenario where users' on-path internet traffic is non-consensually modified. Impersonation: A scenario where a user is impersonated during a measurement. Legal: Users and service providers are bound by a wide range of policies from Terms of Service to laws, each according to context and jurisdiction. A measurement study may violate these policies, and the consequences of such a violation may be severe. At the same time, legal requests may compel disclosure of sensitive data to government entities or to courts in lawsuits. Unavailability: Users or other entities may rely on the information or systems that are involved in the research and they may be harmed by unexpected or planned unavailability of that information or systems . System or data corruption: A scenario where generated or modified traffic causes the corruption of a system. This covers cases where a user's data may be lost or corrupted, and cases where a user's access to a system may be affected as a result. Emotional trauma: A scenario where a measurement of or exposure to content or behaviour in an internet measurement study causes a user emotional or psychological harm.

This section sets out considerations for researchers designing and carrying out internet measurements, to mitigate the risks described in . They cover general considerations, obtaining consent from users, data collection and sharing, and the impact of measurements on the network. Not every consideration applies to every measurement, and researchers should apply them in the light of the specific risks their study poses. Where a consideration in this section allows an exception, or where it is unclear whether a measurement poses more than minimal risk, the decision should not rest solely with the researchers who would benefit from it. Refer it to an independent party, such as an institutional ethics review board. Where none is available, as is often the case for civil society and independent researchers, seek review from an independent body, or from peers outside the project. Record the decision and its reasoning, and report it with the results ().

The benefits of internet measurement should outweigh the risks . Internet measurement studies should generally advance knowledge and create meaningful results. Benefits of internet measurement studies can include but are not limited to analysis of internet topology; understanding the effect or prevalence of new internet technologies, protocols or applications; characterisation or prevention of malware or other security and privacy threats, and much more. The risks of internet measurement studies, particularly to user safety, are summarised in . Consider those risks, but also those stemming from use of auxiliary data (e.g. third-party data sets). Note that while a privacy risk may not be immediately apparent or realisable, in the future increased computing power may then make something possible. Weighing benefits against risks is not only a matter of comparing total benefit against total risk: the Menlo Report's principle of justice also asks how risks and benefits are distributed . In internet measurement, the risks often fall on specific users, such as those whose traffic, devices or networks are measured, while the benefits go to researchers, operators or the internet community as a whole. Identify who bears the risks of a measurement and who receives its benefits. Where these differ, and especially where those at risk are vulnerable or cannot practically consent, reduce the risk as far as possible () and explicitly justify what remains. A study is not justified simply because its total benefits exceed its total risks. No aggregate benefit justifies exposing people to more than minimal risk without their informed consent (). Ensuring that a project's benefits outweigh its risks requires applying the considerations in this document, as well as regular community feedback (particularly from those who will ostensibly benefit or be at risk through the project), proper auditing of the study, and impact assessments.

Have “a general policy of openness about developments, practices and policies with respect to personal data.” Despite best intentions, things fall apart. In the case of any adverse events, researchers should “responsibly inform affected stakeholders.” Contact affected users directly where possible. In all cases, information about the discovery of security vulnerabilities, data compromise, and instances of unauthorized access to information should also be published, once doing so would not itself put users at further risk (for example, once a vulnerability has been fixed). At a minimum, report the following with the published results of the research: the outcome of any ethics review; the risk analysis (); how consent was obtained (); any exceptions taken (); and any adverse events. For long-running experiments, also institute a regular accountability and reporting mechanism, such as an annual review by an advisory board. Such regular accountability mechanisms can include an assessment of the experiment's impact on user safety and privacy, analysis of security events, and an evaluation of whether the experiment is meeting its expected goals.

Proactively identify your measurement to others on the network. "This allows any party or organization to understand what an unsolicited probe packet is, what its purpose is, and, most importantly, who to contact." Example: For a layer 3 IP packet probe you could mark measurements with a probe description URI as defined in RFC9511. This guideline may be ignored if attribution runs contrary to the purpose of the measurement study or would otherwise render it ineffectual. For example, for an experiment collecting information about local internet censorship, the network may provide different results if the connections can be identified as serving that purpose. A decision not to identify measurement traffic should be reviewed as described in , and should be disclosed when results from the measurements. Identifying measurement traffic also helps to reduce the impact of a measurement on others' infrastructure, as discussed in .

Accountability and transparency are fundamentally related to consent. As per the Menlo Report, "Accountability demands that research methodology, ethical evaluations, data collected, and results generated should be documented and made available responsibly in accordance with balancing risks and benefits." A user is best placed to balance the risks and benefits for themselves therefore consent must be obtained. The form of consent required is set by the risk a measurement poses to users (), not by how practical consent is to obtain. Researchers should obtain informed consent for any measurement that poses more than minimal risk to users (), however difficult that is; if it cannot be obtained, the measurement should be redesigned to reduce its risk or not performed. Proxy consent or implied consent may be used instead only for measurements that pose minimal risk to users, where informed consent is not practical. Consent obtained from the users who take part in a measurement does not extend to other users it affects, such as contacts whose data appears in measured communications, or to other affected parties such as the operators of measured infrastructure, many of whom cannot practically be asked. Measurements should be designed so that those who cannot be asked for consent are exposed to no more than minimal risk, using risk analysis () and data minimization ().

Informed consent should be collected from all users that may be placed at more than minimal risk by an experiment. For consent to be informed, a reasonable coverage of possible risks must be presented to the users. The considerations in this document can be used to provide a starting point, although other risks may be present depending on the nature of the measurements to be performed. In addition, it should be clear from the language of consent request who the asker is, and what the terms of data observation and/or collection are. Example: A researcher would like to use volunteer-owned mobile devices to collect information about local internet censorship. Connections will be attempted by the volunteer's device with services and content known or suspected to be subject to censorship orders. This experiment can carry substantial risk for the user depending on their specific circumstances. Trying to access censored material can be seen as (network) policy infringement or breaking laws. Consequences can range from disciplinary action from their employer to imprisonment by government authorities. If the experimenter wants to expose volunteers to this kind of risk, users must be fully informed, and voluntarily give consent to run the measurement. Even then, experimenters should seriously consider designing their experiment in another way. Researchers using shared measurement platforms whose vantage points are hosted by volunteers should also consider the same obligations.. Note that informed consent is notoriously tricky to obtain. Conveying all possible risks of a measurement is often simply impractical, depending upon how technical the user audience is, the context of the consent prompt, what the tool is normally used by users for, etc. In addition, consent can have network effects. For example, asking a user to consent to sharing information about their communication with others can have impacts on users who have not personally consented to the study (see ).

Where a measurement poses minimal risk to users and it is not practical to collect informed consent from all users of a shared network, it may be possible to obtain proxy consent. Proxy consent may be given by a network operator or employer that would be more familiar with the expectations of users of a network than the researcher. In some cases, a network operator or employer may have terms of service that specifically allow for giving consent to third parties to perform certain experiments. Example: Some researchers would like to perform a packet capture to determine the TCP options and their values used by all client devices on a corporate wireless network. The employer may already have terms of service laid out that allow them to provide proxy consent for this experiment on behalf of the employees, in this case the users of the network. The purpose of the experiment may affect whether or not they are able to provide this consent. Say, performing engineering work on the network may be allowed, whereas academic research may not be already covered.

Where a measurement poses minimal risk to users and even proxy consent is not practical, as in many larger scale measurements, implied consent may be presumed. Consider that users of a network will have certain expectations of privacy and those expectations may not align with the privacy guarantees offered by the technologies they are using. As a thought experiment, consider how users might respond if asked for their informed consent for the measurements you'd like to perform. Implied consent should not be considered sufficient for any experiment that may collect sensitive or personally identifying information. Nor should it be considered sufficient for active measurements that may cause harm to individuals, such as those directed at personal devices rather than infrastructure. If practical, attempt to obtain informed consent or proxy consent from a sample of users to better understand the expectations of other users. Example: A researcher would like to run a measurement campaign to determine the maximum supported TLS version on popular web servers. The operator of a web server that is exposed to the internet hosting a popular website would have the expectation that it may be included in surveys that look at supported protocols or extensions but would not expect that attempts be made to degrade the service with large numbers of simultaneous connections. Example: A researcher would like to perform A/B testing for protocol feature and how it affects web performance. They have created two versions of their software and have instrumented both to report telemetry back. These updates will be pushed to users at random by the software's auto-update framework. The telemetry consists only of performance metrics and does not contain any personally identifying or sensitive information. As users expect to receive automatic updates, the effect of changing the behaviour of the software is already expected by the user. If users have already been informed that data will be reported back to the developers of the software, then again the addition of new metrics would be expected. Note that the reduced impact of A/B testing should not be used as an excuse to push updates that might compromise user expectations around security and privacy. In the event that something does go wrong with the update, it should be easy for users to discover that they have been part of an experiment and roll back the change, allowing for explicit refusal of consent to override the presumed implied consent. Example: A research project looks at networked "things", yet users' only interface with the network is through a device that does not provide interaction to the degree that would be sufficient to obtain informed consent at time of use. In this case the user can be informed of the use of data for internet measurement research in the device's terms of use and privacy notice, which can be included in a printed, physical manual for the device or accessed at any time via a webpage. Such a notice informs users rather than asking for their consent, so continued use of the device can at most be taken as implied consent. It is therefore only appropriate for measurements that pose minimal risk to users.

When collecting, using, disclosing, and storing data from a measurement, use only the minimal data necessary to perform a task. Reducing the amount of data reduces the amount of data that can be misused or leaked. When deciding on the data to collect, assume that any data collected might be disclosed. There are many ways that this could happen, through operational security mistakes or compulsion by a state authority. When directly instrumenting a protocol to provide metrics to a passive observer, see section 6.1 of RFC6973 for the data minimization considerations enumerated below that are specific to the use case.

Collect only that data that is required to perform the study and discard data that is not required. When performing active measurements, be sure to only capture traffic that you have generated. Traffic may be identified by IP ranges or by some token that is unlikely to be used by other users. Again, this can help to improve the accuracy and repeatability of your experiment. For performance benchmarking, requires that any frames received that were not part of the test traffic are discarded and not counted in the results.

Data should be stored in a secure location, with appropriate access controls and full-disk encryption if possible. Access to the measurement data should be minimized to only those personnel who require access to perform research.

For any data collected evaluate which minimization techniques can and should be applied to that data in order to minimize the risk to users whilst still safely providing sufficient data for the measurement to be effective. A range of pseudonymization or anonymization techniques are available . IP addresses are particularly identifying, as they can often be linked to a specific subscriber, household or device, and to an approximate location. Where the measurement does not require full IP addresses, they should be discarded, truncated or pseudonymized using the techniques below.

Mask data that is not required to perform the task. This technique is particularly useful for content of traffic to indicate that either a particular class of content existed or did not exist, or the length of the content, but not recording the content itself. The content can be replaced with tokens or encrypted. It is important to note that masking data does not necessarily anonymize it .

When collecting data, consider if the granularity can be limited by using bins or adding noise. Differential privacy techniques can help with this. Example: presents a case-study on the in-memory statistics in the software used by the Tor network.

There are various techniques that can be used to reduce the accuracy of the collected data and make it less identifying while still meeting the needs of the measurement. The use of binning to group numbers of more-or-less continuous values, coarse categorization in modeling, reduction in concentrations of IP address by geography (geoip) or other first- or second-order identifiers, the introduction of noise and all privacy-preserving measurement techniques that allow researchers to safely conduct internet measurement experiments without risking harm to real users .

Further to use of measurement data, data is often shared with other researchers. Measurement data sharing comes with its own set of expectations and responsibilities of the provider. Likewise there are responsibilities that come with the use of others’ measurement data. One obvious expectation is around end-user consent (see "Implied consent" above). Where data obtained from others was collected in ways that raise ethical concerns, disclose this when using it or publishing results based on it. Allman and Paxson provide "a set of guidelines that aim to aid the process of sharing measurement data... [in] a framework under which providers and users can better attain a mutual understanding about how to treat particular datasets." Their guidance for data providers is to:

• explicitly indicate the terms of a dataset’s acceptable use
• convey what interactions they desire or will accommodate.

Their guidance for researchers is to:

• be thoughtful in the reporting of potentially sensitive information gleaned from providers’ data.
• comply with the indications and interactions of the data providers.

Example: Researchers have obtained network measurement data from more than one provider for purposes of conducting analysis of protocol use on both. Where privacy partitioning techniques are used, the researchers' findings may inadvertently collude to uncover private information about users. Once realised, researchers should mitigate this privacy risk to end users as well as disclosing this result to the data providers themselves.

If your experiment is designed to trigger a response from infrastructure that is not your own, consider what the negative consequences of that may be. At the very least your experiment will consume bandwidth that may have to be paid for. In more extreme circumstances, you could cause traffic to be generated that causes legal trouble for the owner of that infrastructure. The internet is a global network that crosses many legal jurisdictions and so what may be legal for one is not necessarily legal for another. If you are sending a lot of traffic quickly, or otherwise generally deviating from typical client behaviour, a network may identify this as an attack which means that you will not be collecting results that are representative of what a typical client would see. One possible way to mitigate this risk is transparency, i.e. mark measurement-related data or activity as such, as also recommended in . For example, the popular internet measurement tool ZMap hardcodes its packets to have IP ID 54321 in order to allow identification .

When performing active measurements on a shared network, maintain a list of hosts that you will never scan regardless of whether they appear in your target lists. When developing tools for performing active measurement, or traffic generation for use in a larger measurement system, ensure that the tool will support the use of a "Do Not Scan" list. If complaints are made that request you do not generate traffic towards a host or network, you should add that host or network to your "Do Not Scan" list, even if the request is automated. A request may be declined only where honouring it would hide the very thing being measured. For example, in censorship measurement, a request could be an attempt to hide the fact that a certain target is unreachable in a particular geography. A decision to decline a request should be reviewed as described in , and the requester informed where possible. You may ask the requester for their reasoning if it would be useful to your experiment. This can also be an opportunity to explain your research and offer to share any results that may be of interest. If you plan to share the reasoning when publishing your measurement results, e.g. in an academic paper, you must seek consent for this from the requester. Be aware that in publishing your measurement results, it may be possible to infer your "Do Not Scan" list from those results. For example, if you measured a well-known list of popular websites then it would be possible to correlate the results with that list to determine which are missing. This inference might leak the fact that those websites specifically requested to not be scanned. On the other hand there may be benefits to publish your "Do Not Scan" list including the opportunity to reconsider the list over time, invite feedback, ensure research reproducibility, or bootstrapping other measurement projects or tools.

This document as a whole addresses user safety considerations for internet measurement studies, and thus discusses security considerations extensively throughout regarding collection and storage of user data.

This document has no actions for IANA.

Many of these considerations are based on those from the adapted and generalized to be applied to internet research. Other considerations are taken from the Menlo Report and its companion document , and research and discussions of the internet measurement and ethics communities. Comments and contributions from Marwan Fayed, Jeroen van der Ham, Arturo Filastò, Christian Huitema, Tobias Fiebig, Greg Skinner, Oliver Gasser, Craig Partridge, Eric Rescorla and Shivan Kaul Sahib greatly improved this document.

Open Source Development Lab This document is a republication of RFC 1944 correcting the values for the IP addresses which were assigned to be used as the default addresses for networking test equipment. This memo provides information for the internet community. Tor Project Tor Project OECD University of Michigan CDT Nokia Siemens Networks Skype NeuStar, Inc. ULD Janet EthiCS 2023 workshop, NDSS Symposium RIPE Labs CAIDA ACM SIGSAC Conference on Computer and Communications Security (CCS '18)