News

Preview of IETF98: .homenet, new transport developments and data protection politics

2017-03-24 News

By Monika Ermert - Despite some alarm bells raised over a potential ban for travellers from mainly Muslim countries and limitations on bringing laptops, iPads or cameras in carry-on luggage when traveling from eight countries to the US, the IETF is back in the United States after some absence. Meeting in Chicago on 26-31 March, it reacted in its own way by announcing that EMC/Dell will be offering those in trouble with their laptops some “loaner” devices for the week.

The Chicago meeting will see the Alissa Cooper, Cisco, taking over as IETF Chair from Jari Arkko, a change noteworthy in several regards: Cooper is the first women to head the still very much male engineer crowd. Cooper's employer, Cisco, a long-term supporter of the IETF – which Cooper underlined in her self-portrait blogpost – has just considerably cut down the ranks of its IETF expert group.

Who can allocate .homenet?

A turf war of sorts might finally be declared between the IETF and ICANN over how much leeway the IETF has to allocate “specialised” TLDs. ICANN Chair Steve Crocker chimed in on an already heated discussion on the DNSOP mailing list of the IETF over “should .homenet be added to the root?”. On both sides of the aisle, people are starting to question the special allocation process being used as a route around the feisty ICANN processes.

ICANN stakeholders – the registries – brought this up for the first time at the recent meeting in Copenhagen. IAB member Russ Housely called the .homenet approach something like “an end run around” the ICANN process. And one of the reviewers plainly told .homenet draft author Ted Lemon he thought “chances of getting a 'yes' out of the ICANN community is very nearly zero”, given that there was already talk of lawsuits over .home and .corp. DNSOP might take some time from its busy agenda to look into the issue. There is more .homenet stuff coming, like a draft on naming in that zone in the homenet WG.

Another discussion slot for the .homenet naming is provided in the DNSSD WG, which has some interesting topics on its agenda this time including the relation of DNSSD and IoT, as well as a new draft by Christian Huitema (Microsoft) on DNS-SD Privacy.

A ton of DNS work and RegExt discussing RDAP

A thick pack of 187 pages of documents are part of the meeting package of the DNSOP group for Chicago. Some are more maintenance-like things, like a draft on DNS Terminology by ICANN's Paul Hoffman or even an attempt to standardise how to capture DNS packets for monitoring. But there are also several proposals for authenticated denial of existence to reduce the load on the root servers.

For the DNS crowd, continued deliberations of the WHOIS-successor protocol RDAP is also of considerable interest. The RegExt WG will be experimenting with a new meeting format, allowing for more practical, running code workshops. Besides the RDAP session, which will address the “ongoing policy development” and “non-browser clients”, there is also an EPP session covering questions like “should a premium domain name be returned as unavailable in the check if the fee extension is not passed, since the create would most likely fail later in the purchase flow?”.

Transport modernisation – full steam ahead

The renaissance of transport protocol development also continues with Quic being on top of many agendas. For those interested in getting a Quic introduction, there is a tutorial about the light-weight protocol originating from Google's labs on Sunday. After an intersessional in Tokyo, the Quic WG is pushing ahead with discussions on transport (presentation), recovery, TLS (presentation) and HTTP issues, and allows itself a full hour for open issues.

The rebirth of transport is also illustrated in a total of three Transport Area meetings, which are taking the entire Monday afternoon and are covering the somewhat contradictory issues of the Transport-independent path layer state management ideas (former Spud ideas) on the one hand, and the potential privacy problems in transport protocols on the other.

Privacy receiving more attention in protocol-making

On Monday at the Transport Area Open Meeting, an activist from European Digital Rights, Diego Naranjo, will be presenting on the subtleties of the EU's ePrivacy Regulation. A discussion on Tor's perspective on privacy and traffic analysis resistance for encrypted protocols will follow.

A data protection official – or Edward Snowden, for that matter – would be rather happy to see how much attention is being increasingly given to privacy issues in the IETF. On Tuesday night, Snowden said to a large crowd at the CeBIT fair in Hannover that it was not the law that would help regain control on the net, but rather good engineering. Voilà, check out the agenda of the IRTF's Human Rights Considerations in protocols research group meeting in Chicago, which talks about distributed Architectures and rights, the IEEE Global Initiative for ethical considerations in AI & AS on the consideration to have “no domain left behind” in the “Let's Encrypt!” campaign of the Electronic Frontier Foundation and on 'The Internet's Standardisation Bodies as the Guardians of Privacy Online in the Face of Mass Surveillance'.

What is certainly doing the job are the practical steps taken, for example, for securing email the Using TLS in applications (UTA WG meeting), on pushing DANE for DNSSEC Authentication in TLS in the TLS WG, on privacy enhanced RTP conferencing (PERC) or work on IoT security, presented in the Security Area Advisory Group.

Will the “Protocol for Dynamic Trusted Execution Environment Enablement” become an addition to security and privacy attentive specifications? The TEEP BoF, one of only four in Chicago, could become an addition to the list of human rights issues that receive more attention.

TEEP, as well as new work on coordinated IP address space management (currently done in multiple unstandardized ways) in the CASM BoF, the GitHub BOF WUGH BoF and the IASA 2.0 BOF, reviewing the structure of the IETF administrative side, were all recommendations for the Chicago meeting by the outgoing IETF Chair, Jari Arkko. The IETF has changed over Arkko's tenure in quite some ways.