Member Associations to Liaison Relationships

[mnot]

I've heard people suggest that we need to transition Member Associations to a Liaison model, but didn't see an issue for it.

See also https://github.com/w3c/board/issues/180

cc @dwsinger

[frivoal]

Note that the Process does cover the existence of liaisons, but in my opinion says very little that about them that is of much value. I see very little in the relevant section that (a) is enabled by this text which we couldn't otherwise do, nor (b) is required by this text, which we would otherwise be free to skip. So much so that I've previously argued that we could delete all the Process says about Liaisons and not notice the difference.

If we want to move away from using Member Associations towards a Liaison model, we might need to define Liaisons a bit more.

Maybe we could start with an inventory of which Member Associations we currently have or previously had, figure out what these were for, and try to see if there's something we can generalize.

@plehegar Is this something we could get data about?

[chaals]

There's not a lot here about why. Membership organisations are sometimes not like other members:

• We want to be able to coordinate easily in particular to minimise genuinely redundant work that asks members to either choose only one set of collaborators or invest twice as much to keep up
• We don't want to have organisations that are potentially in competition with us (for members, as hosts of work, etc) to be as influential over our governance as we think it is reasonable for members to be
• There's often little appetite to swap the money required for mutual memberships
• The question of who is authorised to participate at what level is more fraught (albeit less so now we are much more open than the late 90s)

But then sometimes they are like other members:

• They're only really interested in some subset of what we do at any given time
• Their representatives are mostly focused on something other than W3C as their primary job