back to article IETF rates itself 'minimally acceptable' on key measures of community, efficacy

The Internet Engineering Task Force (IETF), the open standards organisation that defines and develops internet standards, has surveyed its membership, and found that many of its key efforts are merely "minimally acceptable" – rather than efficient or inclusive. Those findings emerged in the IETF's first ever community survey, …

  1. Yes Me Silver badge
    Paris Hilton

    M/F

    Another item rated as a key finding is that females make up between 9.2 and 10.7 per cent of its community.

    To be clear, the demographics of the IETF aren't controlled by the IETF. Mainly they are controlled by the employers that choose to fund people to participate in the IETF. Admittedly, there's also the question of who wants to participate and why they might decide not to.

  2. Warm Braw

    Asia home to just ten per cent of respondents

    It's a while now since I've been to the IETF and my attendance was always as a "non-combatant". It inevitably reflects the priorities and the culture of the companies (principally US companies) that find attendance pays dividends exceeding the cost of fees, travel and accommodation. However, the IETF's role is not entirely passive - it has made efforts to diversify the location of meetings and I was at the first non-US meeting in Amsterdam as far back as 1993 - but it has nevertheless clung to its US-centricity. It was always seen as the strength of the IETF that it had no particular mandate to represent anyone other than those who turn up at the meetings and therefore that it could "get things done" unlike those cumbersome national standards bodies that demanded a stake in the outcome even in the absence of an IT industry of their own. The price of getting things done, though, was that international considerations were often a grudging afterthought.

    The climate in which the IETF thrived has changed. Computer use is no longer a majority US (or indeed Western) phenomenon. The Internet is no longer under the de facto funding of Western educational institutions. And, of course, we've just been through a difficult period that has proved that the technology we have assiduously been developing is a productive substitute for meetings in person. I don't see that the IETF in its present form has much more to contribute: it has no official status and the interested parties have no dependence on it for their continued collaboration. If you're a group of predominantly Chinese firms wanting to come up with a uniform approach to a technical problem what value does the IETF provide?

    Organising physical meetings for the world to discuss technical problems in English seems peculiarly anachronistic in a age of remote working and machine translation.

    1. Doctor Syntax Silver badge

      Re: Asia home to just ten per cent of respondents

      "If you're a group of predominantly Chinese firms wanting to come up with a uniform approach to a technical problem what value does the IETF provide?"

      That your approach is not only uniform between yourselves but uniform with the way the rest of the internet works?

      The whole system depends on standards. What body works those out? AFAICS if it didn't exist it would have to be invented, if only to prevent ISO or the ITU taking over.

  3. Anonymous Coward
    Anonymous Coward

    "If you're a group of predominantly Chinese firms wanting to come up with a uniform approach to a technical problem what value does the IETF provide?"

    How about asking those Chinese firms who repeatedly show up at IETF meetings? Huawei typically sends > 50 people.

  4. R Soul Silver badge

    I don't see that the IETF in its present form has much more to contribute: it has no official status and the interested parties have no dependence on it for their continued collaboration

    That must surely come as a surprise to the likes of Cisco, Google, Juniper, Huawei, Akamai, Microsoft, Cloudflare, Nokia, Apple, Ericsson, etc who are heavily invested in the IETF and its outcomes.

    If the IETF hasn't got "much more to contribute", where do you propose the work on core Internet protocols (routing, DNS, HTTPS, TLS, etc) should get done? Do you want this stuff happening behind closed doors at the ITU or 3GPP or in some banana republic's Ministry of Stupid?

    If you're a group of predominantly Chinese firms wanting to come up with a uniform approach to a technical problem what value does the IETF provide?

    Technical/engineering scrutiny by subject matter experts who have experience and insights into what does and doesn't work.

    1. Yes Me Silver badge
      Happy

      No Official Status

      it has no official status

      This has long been considered a major advantage of the IETF by participants, including the companies that R Soul lists. Those of us who tried to interest ISO and ITU-T (formerly CCITT) in the Internet in the 1980s are pretty glad that we failed.

      The cold dead hand of governments is refreshingly absent from the IETF (and the W3C). The Internet is much better as a result.

  5. Warm Braw

    where do you propose the work on core Internet protocols (routing, DNS, HTTPS, TLS, etc) should get done

    Where it mostly gets done these days is inside Google, Microsoft, Apple or Amazon. Along with most of the contributions to W3C output.

    If the IETF and/or W3C is to have any useful purpose in future it would be in ensuring a wider participation because even if these organisations lived up to their claims of "technical/engineering scrutiny" - which is perhaps not as robust as some of the participants would like to believe - there are a whole range of societal and operational impacts that largely don't get the attention they deserve.

    And it's the sneering dismissal of "banana republics" and other bodies (and I note that in the mobile phone space they've managed to go from 2G in 1991 to 5G while the IETF has been struggling with the acceptance of IPv6 over almost that whole period) that pretty much sums up why there won't be wider participation in the IETF: because the present participants like it the way it is.

    1. J. Cook Silver badge

      (and I note that in the mobile phone space they've managed to go from 2G in 1991 to 5G while the IETF has been struggling with the acceptance of IPv6 over almost that whole period)

      To be fair, going from IPV4 to IPv6 is not a trivial thing for most companies, who are quite satisfied with putting their DMZ hosts behind a NAT; Most ISPs are loath to hand out static IPv6 addresses either.

      To wit: what they got works- why fix it when it's not broken?

    2. Yes Me Silver badge
      Headmaster

      Where it mostly gets done

      Where it mostly gets done these days is inside Google, Microsoft, Apple or Amazon.

      You are confused. Let me fix it for you:

      Where it mostly gets started these days is inside Google, Microsoft, Apple or Amazon or Cisco, Huawei,.... Where it gets polished, made interoperable, and therefore useful, is in the IETF (or W3C, or WHATWG, or ...).

      If you don't understand why it's this way, then you don't understand how intellectual property law and anti-trust law interact with the open standards process.

  6. Anonymous Coward
    Boffin

    Mired in history

    At its start, the IETF was tied to the US government. Its first meeting consisted of 21 US government sponsored researchers and, since it was in 1986, I would guess that they were all white males. Over the years, the government aspect toned down but the US centrism remained as did the numbers of white male participants.

    Because of this history, it is not surprising that potential non-US and/or BAME participants question whether their voice will be heard and their decision not to participate maintains the status quo.

    1. Yes Me Silver badge
      Megaphone

      Re: Mired in history

      One of the great plus points of modern standards development (and of open source development) is that you can't see the other people most of the time (unless they choose to show you their picture or video). So the idea that non-US voices (such as mine) or BAME voices (there are many) will not be heard is wrong. Plain text email has no gender or skin colour. It does, inevitably, convey an accent.

      The IETF (like the W3C) uses English. Well, so does almost every international collaboration in science and technology. [Feel free to cite exceptions, but they are exceptions.] If you have an open source AI that can change this so that everybody can work in their preferred language, please let us know and I'm sure we'll adopt it. For now, we use English, much of it written by people who have learned it as a second language.

      For an extraneous reason, the IETF hasn't held an in-person meeting since November 2019. It's worked remarkably well. Travel to meetings, and the heavy associated costs, have gone away for a while. Now's the time!

      P.S.: since it was in 1986, I would guess that they were all white males

      Wrong, as it happens. Lixia Zhang was there, and she was at IETF 111 last month.

    2. Anonymous Coward
      Anonymous Coward

      Re: Mired in history

      "it is not surprising that potential non-US and/or BAME participants question whether their voice will be heard and their decision not to participate maintains the status quo."

      That's an issue for the IT sector as a whole, not just the IETF. Your own IT department (or company boardroom or parliament or...) is unlikely to have a number of non-US and/or BAME (or gay or disabled or....) people in proportion to their share of the overall population.

      Non-US and/or BAME participants are and have been in prominent leadership positions in the IETF. Their voices haven't just been heard, they've had the support and peer recognition to become co-chairs of working groups, area directors and so on. That should at least show non-US and/or BAME participants that there are role models who got those positions on merit. BTW, the current chair of the IAB (which oversees the IETF) is a German woman. The previous IETF chair was an American woman.

      That's not to say everything's fine at the IETF because there's clearly more work to be done on this topic. A bunch of predominantly old fat guys and greybeards does have an image problem.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon