Re: [tao-discuss] Review Request for Possible Revision of the Tao of the IETF

["Rob Wilton (rwilton)" <[e-mail]>]

I broadly agree with what Jay has written here.



When I joined the IETF, I'm sure that someone suggested that I read the Tao.  I'm not sure how far I got through the document but not very far, or perhaps I jumped to specific sections.  I just wanted to come to a meeting and present a draft, why would I need to read such a long document!  Mostly, I turned up at the meetings, got some things right, some things wrong, learnt from my mistakes, and asked things that I didn't know (e.g., what is a bis document).  However, I did also have a large pool of colleagues to draw on, which means that I got a different experience than many newcomers.



I think that this document effectively contains a useful condensed version of the key information that a participant who gets past the "new attendee" first 5 IETFs may well want to read/understand, but of course, that means that it has the duplication of diverging content issue that Jay mentions.  However, anything that removes the requirement for participants to say read RFC 2026 and all its updates would be a win for me.



I would really like all the IETF process documents to not be RFCs/BCPs at all (RFCs/BCPs should only be the output of the IETF), but instead migrate to a set of web pages/sites, probably with a lower review barrier than full IETF consensus for most of the content.  Specifically, I think that there should be certain key steps in managing our process that must be defined with full IETF consensus, but that the implementation of that process should have a lot more freedom to evolve with how the organization, people, and people's interaction with the organization evolves over time.  Small fixes, e.g., BCP45bis, really shouldn’t need to take so much time/energy to make some pretty trivial updates.  Of course, this is way beyond the scope of this document and would probably take 2 years of arguments discussion before we even got agreement to start it, if there was any consensus at all.



Now getting off the soapbox …



In terms of helping new people coming into the IETF, I agree that having task/goal related webpages that contain the key information would be more helpful and approachable for newcomers to the IETF.



Regards,

Rob





> -----Original Message-----

> From: iesg <[e-mail]> On Behalf Of Jay Daley

> Sent: 14 April 2022 15:42

> To: [e-mail]; IESG <[e-mail]>

> Cc: Salz, Rich <[e-mail]>; Lars Eggert <[e-mail]>; Greg Wood

> <[e-mail]>; Brian E Carpenter <[e-mail]>;

> Niels ten Oever <[e-mail]>

> Subject: Re: [tao-discuss] Review Request for Possible Revision of the Tao of the

> IETF

>

> This whole conversation seems very messy, reflecting an underlying mess

> around content.  Thanlfully, we have some strands coming together to produce

> a new strategy for our content, which will then provide the clarity to resolve

> discussions like this.  What follows is my analysis of those strands and then how

> those relate to the Tao, starting with the less contentious issues first.

>

> #  Review

>

> Review can mean a number of things.  It can mean that a document must be

> checked by a specific third party before it is published, which is how I think

> review of the Tao is being looked at, but it can also mean that content is open

> for anyone to review at any time, comment on and propose changes to.  For

> content that is going to be long-lived or small amounts of short-lived critical

> content then the former makes sense, but for large volumes of content that

> changes regularly the former becomes impossible and we have to use the latter.

> The website content is clearly of that nature.  Enabling this latter kind of review

> requires some work:

>

> - content published in a form/mechanism that allows for that "drive by" review

> and change requests, which we generally take to be GitHub.

> - clear responsibilities on who reviews the change requests

> - quick response to any issues and change request received

>

> We’ve partly achieved that for the two micro-sites we’ve recently launched -

> authors.ietf.org and chairs.ietf.org as those are now GitHub backed but the

> main website is not there yet.

>

> #  Structure of our content

>

> We have a huge volume of content, written by many dedicated people over

> many years and there are a lot of positives in that:

>

> - lots of people voluntarily put huge effort into writing documentation

> - people keep an impressive array of knowledge in working memory that they

> can use both to author and review

> - people really care and are trying hard to provide people with useful info

> - great level of detail

>

> However, much of our content is written for "IETF participants" in the broadest

> sense, rather than more targeted at specific roles or tasks and that creates a

> range of issues.  In other words, our content often seems to assume that all

> participants are going to speak in WGs, author drafts, be a scribe, ask questions

> at the plenary, etc, etc. The issues are:

>

> - our content often tries to cover everything that someone might possibly need

> to know, ever,

> - we say the same/similar thing in multiple places, which in turn makes it hard for

> people to find everything they need in one place

> - reading and understanding our content is overwhelming, there’s rarely an easy

> way in

> - leads to duplication as each of the locations tries to cover enough to be useful

> - when things change the incorrect text has to be tracked down in multiple

> places, which it rarely is and so stale content abounds

>

> What we need, and what Greg is working on with emodir, is a participant

> journey - a set of roles that participants take that we can flesh out and target

> our content at.  To be clear, I don’t mean the difference between beginner and

> expert, those can still be interpreted as "need to know everything" just at

> different levels of detail.  Instead I mean "roles" such as: mailing list lurker, in-

> person participant, author, wg chair, etc.  The new authors.ietf.org and

> chairs.ietf.org follow that principle - they aggregate all the information for one

> specific role into one place.  Those are not finished by any means, we don’t have

> a formal review team in place, there’s lots of old content around, we don’t have

> enough stratified content for beginners to experts, but they are getting there.

>

> # Our style of writing

>

> Most of our content, whoever the author, is written in a friendly open style.

> Authors have made a great effort to be clear and to be inclusive.

>

> However, despite that, there are two bad habits that fly underneath the radar.

> The first is to be defensive  - must do this, don’t do this, avoid that, watch out

> for this, and so on.  It’s as if successful participation in the IETF is predicated on

> learning all the rules not to break and the specific processes that must be

> followed to the letter.  The second is to believe that IETF memes or in-jokes

> have the same (or any) meaning outside the IETF as they do within.  The "many

> fine lunches" example from this conversation captures that perfectly - a

> completely meaningless phrase unless you’ve been part of the IETF for at least

> 10? years.

>

> # Relating this all to the Tao

>

> For me, the Tao incorporates both the best snd the worst of the above.  The

> effort and dedication that has gone into it are deeply admirable, the breadth of

> content is excellent and the tone is very friendly.

>

> The problems I have with it though are not minor. The name itself, ’Tao’, is a

> reference too far.  Sure some of use have read "The Tao of Pooh" and get the

> allusion but for most people, it goes straight over their heads

>

> More importantly though, I sincerely doubt that it is of any practical use to its

> intended audience - it’s far too sprawling, detailed and full of rules to be of use

> to a newcomer. To be quite candid, every time someone recommends to a

> newcomer that they read the Tao, I cringe.  What it does do is provide a useful

> reference for long-term participants (where else are the dot colours

> documented) but that’s not the intended audience.  It’s become more a series of

> authoritative statements that can be referenced/found individually than a

> flowing document.

>

> While I admire the effort that has gone into the Tao, I think we need to move on

> from this style of documentation.  Replacing it with several smaller

> documents/sites/pages would be much more useful: "Guide to the role and

> structure of the IETF", "Guide to participating in IETF working groups", "Guide to

> making the most of IETF meetings", and so on.

>

> Jay

>

> --

> Jay Daley

> IETF Executive Director

> [e-mail]<mailto:[e-mail]>