The Internet Engineering Task Force (the IETF) proclaims to embrace an open (standards) process. Specifically, RFC2026 – one of the “founding texts” of IETF procedure and functioning states that (section 1.2, emphasis mine):
o These procedures are intended to provide a fair, open, and objective basis for developing, evaluating, and adopting Internet Standards. They provide ample opportunity for participation and comment by all interested parties. At each stage of the standardization process, a specification is repeatedly discussed and its merits debated in open meetings and/or public electronic mailing lists, and it is made available for review via world-wide on-line directories.
The IETF runs (open) mailing lists, and organises 3 annual grand meetings per year – trying to distribute these geographically to “spread the pain of travels equally among the participants”.
While participation in the IETF process is (in theory) possible through participating only in mailing-list discussions, in reality … attending these grand meetings brings the benefit of the bandwidth of a week-long face-to-face discussion, of getting to know people personally (which greatly helps in email discussions), of being able to scribble diagrams on the back of napkins in a bar at 3AM to get a point across, etc.
And according to the quoted text from RFC2026, some debates may happen (and, in my experience, does happen) only in open meetings:
“debated in open meetings and/or public electronic mailing lists”
My personal experience (having authored >20 RFCs over the years) is, that it’s a nice fantasy, but a fantasy all the same. It’s not possible to contribute meaningfully to the IETF process without being physically present, at least semi-regularly. And, I might add, having attended all IETF meetings from 2000 until 2015 (I think it was), each IETF meeting has been an amazing (and exhausting) intellectual experience, where I’ve also met several smart engineers who’ve gone on to become good friends.
It’s therefore incredibly important that the IETF organises its meetings so as to exclude nobody from attending. Obviously, that’s not an easy task.
In the recent 12 months or so, there have been several cases of this becoming an issue for the IETF:
- For the IETF 100, scheduled for Singapore, several IETF participants indicated that they would not be able to attend, due to national rules and regulations in that country. This gave rise to several long email threads, trying figure out how to proceed. The fact that there was this discussion was a testament to IETF trying to adhere to the RFC2026 text about “open meetings”. That IETF 100 is ultimately being held in Singapore is, then, testament to perhaps other considerations weighing heavier than “openness”.
- For IETF 98, and IETF 102, scheduled for locations in the continental USA, several IETF participants including Trustees of the ISOC (the ISOC is one of legal bodies behind the IETF) have indicated that they would not be able to attend, due to national rules and regulations in that country. As expected, this also has given rise to long email threads, the conclusion of course still pending.
On that very topic, Twitter of course also had something to say – since the IETF is not the only organisation that has the problem of selecting venues so as to be inclusive. Scientific conferences face the same set of issues:
And scientific conferences, like an IETF meeting, are incredibly valuable precisely because of the high-bandwidth of spending face-to-face time with a colleague, huddled over a scribble on a napkin, in a bar and at 3AM.
The chairman of the IETF, Jari Arkko, tries to take the bull by its horns, and writes in his IETF-official blog, in an entry entitled “Barriers to Entry”:
The situation is fluid. Legal and political processes around the imposition of barriers to travel will likely continue. We plan to track the situation closely in the US and elsewhere. We believe that Internet protocols develop best when people of many backgrounds can offer their contributions, and we are negatively impacted by policies that prevent such collaboration.
IETF meeting venues are always reviewed for potential impact on attendance by participants from different countries. Our next meeting is planned for Chicago, and we believe it is too late to change that venue.
In some sense, I think that Jari’s words can be summarised into “It’s really hard, but we’re trying“. To that, let me say “Thank you”.
Jari recalls that “the IETF does not make comments on political matters”. I agree, the IETF is an engineering organisation after all. But, even something as benign as selecting a venue to hold the meeting is an eminently politic statement in itself, as it turns out.
International travel has always been difficult. Getting a visa, or not…getting a student visa, or not to go to school, or a short-term residence permit – and, does attending a scientific conference, or an IETF meeting, qualify as “business” or “leisure” when you stay a day or two after. (Co-)organising a conference or a meeting in a foreign country (as the IETF does regularly) is incredibly challenging and frustrating.
All this is not new. It may be getting more complex in some parts of the world – and that’s of course unfortunate. But it’s not new.
The Internet made a fantastic promise come true, of a computer system being able to communicate to any other computer system with the same ease, whether they were hemispheres distant, or adjacent – whether they are on different continents, or sitting next to each other the sane desk: “to an IP Datagram, the entire world is but a send() away”.
The IETF, thus, did it once already – made distant and adjacent access the same.
This would seem to be the occasion for the IETF to once again apply engineering to make distant and adjacent access the same, by doing away with the need for physical meetings for international collaboration. Applying clever technology for making it as for colleagues to collaborate whether they are hemispheres distant, or adjacent – whether they are on different continents, or sitting next to each other around a conference table.