Game design and protocol design (was: Re: Centralization dangers of applying OpenID Connect to wallets protocols)

TL;DR: A board game analogy for those of you that might be having a hard time
following the discussion about centralization and protocols.

On 3/26/22 11:38 AM, Markus Sabadello wrote:
> I agree with what Daniel writes, "the nature of the technical choices 
> enshrines incentives".

I found myself happily nodding in agreement with the vast majority of what
Daniel said as well. I also agree with Dmitri, JoeA, DaveL, Markus, Anders,
and good chunks of what Adrian has been saying.

I hope it's clear to everyone that there is disagreement over the current
trajectory of OpenID as it pertains to VCs, and it's not just from 1-2
troublemakers -- it's a broader concern. That said, Tobias, Oliver, DavidW,
DavidC, TonyN, and MikeJ have also made some solid points and provide key
insights on better ways to structure the conversation. CHAPI, nor any of the
other protocols, are without their risks, limitations, and downsides. Most
everyone seems to continue to have a rational debate, which is great... we're
getting somewhere[1].

As a relevant aside for those of you that are not steeped in protocol design
or worrying about market centralization, I'm reminded by one of the most
important things that Christine Webber taught me. It was in the dead space
after a Rebooting the Web of Trust event, the one in Boston, that she said:

"Have you ever stopped talking to a friend because of the way they acted in a
game before?"

And I said, "Yeah! I mean, not permanently, but some of my friends in high
school... we used to get into the biggest fights over D&D campaigns and then
not talk for days!"

To which, Christine responded:

"More protocol designers need to study game design."

We were talking about protocol centralization, and how some protocols drive
certain implementer, and corporation, behaviours. Most everyone on here is
probably familiar with social media protocols, like: incite strong negative
emotion and polarization so that people will stay more engaged in social media
so we can feed more ads to them. More info at: "Angry by design: toxic
communication and technical architectures"[2].

We were theorizing that there were similar societal effects wrt. corporations
and how they implement technical protocols and how, sometimes, even with the
best of intentions, individuals designing protocols result in corporations
optimizing for things that the protocol allows for, but was never the intent
of the protocol.

It's rare to have game designers also design technical protocols, but we could
all learn a lot from that particular field of study.

When you design a game, you're looking for a particular outcome from the
player(s). If it's a group game, do you want people to collaborate? Do you
want them to be adversarial? If it's a single person game, do you want the
story to progress in an open world (RPG), or on rails (shooter)? How many
times a day do you want the player to throw their controller across the room?

The rules of a game are a protocol. The player does X, the game responds by
doing Y. Similarly, with computers, the protocol client sends X, the server
responds with Y.

... and this is where it gets really interesting.

There is a LOT of thought put into walking that fine line in a game between
frustration and mastery -- you want games to be fun, but if they're too easy,
they're boring:

https://www.theguardian.com/technology/gamesblog/2013/feb/14/frustration-in-game-design

... and that's just one game mechanic... there are MANY game mechanics to pick
from (single player, cooperative, bribery, catch the leader, hexagonal grid
movement, rondel skill use, etc.):

https://boardgamegeek.com/browse/boardgamemechanic

... and when you put those things together, you drive very different
behaviours in players... and it's possible to put the wrong combinations of
things together and end up with a game that's not fun for anyone but the winner.

... and that's what we have with OpenID Connect today. It was never intended
to have the centralizing social login outcome that it did, but here we are
with centralized social login. Why did that happen? That's what we're
exploring in this thread so we can try and avoid repeating that w/ DIDs and
VCs and digital wallets.

There's this beloved Reddit meme that quotes Masahiro Sakurai where he says
something to the effect of "No! This isn't how you're supposed to play the game."

https://knowyourmeme.com/memes/no-this-isnt-how-youre-supposed-to-play-the-game

... and it's often used to point out ways people duck the intended rules of a
situation to achieve an outcome that is the opposite of what the rules were
supposed to enforce.

Whether or not OpenID Connect was intended to have the outcome it did (social
login centralization), while interesting to debate, is a bit of a red herring.
That was the outcome -- centralization -- and saying "No! This isn't how
you're supposed to play the game." to those corporations that centralized
isn't going to get them to change what they're doing. Doubling down on the
previous strategy is also highly unlikely to get the decentralized outcome
some of us want to see.

Just like how game design has "game mechanics", protocol design has "protocol
mechanics" -- different rules that we can put together that drive different
outcomes. Not everything can be controlled technically, true, but there are
SOME things that can nudge behaviours one way or the other, and the more we
fine tune these mechanics, the more we can optimize for a particular outcome.
That outcome isn't guaranteed, but the OpenID game today is rigged... and some
of us don't want to play that game.

The good news is that the OpenID game can be fixed, if that community can 1)
identify the mechanics that are not working, and 2) find the motivation to fix
them.

-- manu

[1]https://docs.google.com/spreadsheets/d/1Gp5V5lTO3pyQ94hS-UR_WTWg0DkBMv0ubXYaKjIPqnU/edit
[2]https://www.nature.com/articles/s41599-020-00550-7

-- 
Manu Sporny - https://www.linkedin.com/in/manusporny/
Founder/CEO - Digital Bazaar, Inc.
News: Digital Bazaar Announces New Case Studies (2021)
https://www.digitalbazaar.com/

Received on Saturday, 26 March 2022 19:57:07 UTC