Web Standards and the Fall of the House of Iamus (Alex Russell)

Discussion
, , ,
1

Every feature begins life as a non-standard proposal, often shipped. This is not a bug.

This is Alex Russell’s post on web standards, and has a lot of discussion about W3C working groups.

I found it a really really useful read that informs how I’m thinking about convening people here.

To state again, we are collectively defining, forming, and educating people about what “ATProto Working Groups” are. They are not a standards process, they are a collaboration tool.

We’re kind of in between a classic open source project (shared code and a rough governance process) and protocol evolution (multiple implementations, interop, seeing who will help put design, engineering and other time into different implementations).

I’m going to quote a few things for the article, starting with this big chunk from the beginning (highlight mine):

Working Groups don’t gate what browsers ship, nor do they define what’s useful or worthy. There are no seers; no Delphic oracles that fleetingly glimpse a true logos . Working Groups do not invent the future, nor do they hand down revealed truths like prophets of the House of Iamus.

In practice, they are diligent, thoughtful historians of recent design expeditions, critiquing and tweaking what they return, then spreading the good news of proposals that already work through Web Standards. These are usually ratified years after features first ship, serving to licence designs liberally to increase their spread.

In the end, this is the role of Web Standards and the Working Groups that develop them: to license patents that read on existing designs, reducing risks to implementers for adopting them.

Anyone who tries to convince you otherwise, or invites you to try your hand at invention within a chartered Working Group, does not understand what those groups are designed to do. Sadly, this includes some folks who spend a lot of time in them.

ATProto is heading to the IETF, and the role should it be accepted there, will be to “ratify things years after features first ship”.

Invention – as well as R&D – can and will come from working groups, but it also won’t be “design by committee”. IETF style running code and rough consensus is what we can practice today. An implementation and 100 people using it is more persuasive than anything else (which is why we also need to be in community with bsky – their market power determines a set of top level adoptions and use).

I realize now that it’s in part why I’ve been uncomfortable with @snarfed.org talking about Lexicon Community being “a baby standards org”. That’s pre-mature! LC is really a big Lexicon working group, not a pretend standards org! It’s also a locus for coordinating on Lexicon adoption, education, and learning together

(I almost said best practices above, but we don’t actually have enough experience to know what a best practice is, that only comes from shipping and also slowing down the pace of innovation, and we’re still moving very fast in all of ATProto land!)

Read the whole thing! Highlight other things that hit for you, ask about how/why ATProto Working Groups - let’s discuss.

6 Likes
2

Thanks for sharing this.

This strongly resonates with my motivation for sharing concerns about co-locating the privacy discussion at the IETF.

Like Alex, my experience of having been through a few of these cycles suggests that meetings like the one in Montreal in November can be good – but it’ll bear repeating and reminding a thousand times the orientation that Alex is pointing to, because there will be some who forget and get understandably excited about inventing the future, instead of sharing insights from the [recent] past.

These latter well-intentioned folks represent the group that Alex points to who populate working groups, but don’t understand the work that they’re doing. Sadly, this orientation can, in my experience, be deeply counter-productive.

So, hurrah for collaboration, and thanks to the stewards who are creating the space for it! <3

4 Likes
3

I gasped when I read this and was excited for you to see it!

2 Likes
4

I wonder if there’s a way to encode the full article into a single sentence:

“In this working group, we only accept statements in the form ‘On project {a}, this approach, {x} is how we are approaching problem {i}‘ or questions in the form ‘I have a project {a} with problem {i}, are there any approaches {x,y,z} that might help?’ or ‘I have a project {a} with problem {i}, can someone help think through the implications of approach {x}?“

(Possibly heavy handed, but maybe a worthwhile starting point!?)

2 Likes
5

Maybe that isn’t a good “rule” per-se, but as a starting point for thought it does get across a lot of the general idea.

6

I’m usually not one for rules! :sweat_smile:

I think one of the challenges that we encounter in this space in particular, that Alex doesn’t really dwell on (but maybe should?) is that the work is important, but overwhelmingly attracts people who like to engage in bike-shedding or theoretical design as a way of [trying to] encode the world in the way they’d like to see.

The problematic comms styles crowd out people who are actually trying to ship things to real people, and the thing we end up with if we don’t set guidelines like my rough suggestion is that we’ll end up with a space that draws well-meaning people in, wastes their time, and doesn’t meaningfully engage people who are actually shipping things.

(Those problematic comms styles are also often accompanied by folks who can be … persistent, and oblivious to their disruptive & unproductive behavior. :sweat_smile:)

2 Likes
7

For context, the founding executive director always called the decentralized identity foundation, where i’ve been active for like 5 years now and counting, a “standards org” or even an “SDO” (standards development org). in W3C terminology, we’re more like a “CG” (community group), or what the IETF would call a “pre-WG mailing list” or a “RG” (research group); since it’s a linux foundation org, exec directors since have gone with the much humbler “pre-standards incubation org”, for reasons entirely explicable by alex russell’s essay. the chain agnostic standards alliance is called that because it’s a trade group of implementers agreeing to common/baseline APIs, nothing more.

i was at an unconference monday where, like every other unconference i’ve ever been to, people say they want “standards” or “standardization” when what they really want is harmonization, co-design, cross-org research, and/or market alignment. all of those are valid, and good, and a crucial part of the ecosystem and processes that can result in standards some day, and are no less valid and useful and worthwhile if they don’t.

1 Like
8

what i try to do at DIF is make really explicit the timeline, the outcome, and the commitment. “can you commit X hours/week for Y weeks, so that at the end of it we’ll have Z?” Important not to call Z a “standard”, but also important to talk brass talks about how much work it is and what it DOES get you to bring Z into the world. i’m often accused of being a killjoy or a party-pooper, but it wastes time if people think a co-authored map can change the territory

2 Likes