Yes, it was a good discussion. I've always been a proponent of smaller set-sizes; as I recall, I was one of the first people to be urging that the initial 72-piece V-Sets be cut down to 54, and then several years later to 36. Personally, I'd be fine if we only made 36 cards per year, but at this point I think 54 per year is a very good spot to be.
I'm also a major fan of the playtester-designed pieces; anything to incentivise and thank our playtesters is a good move, in my book.
I want to offer some clarification about how the SW Imperial Assault v-set process works (they call it the IACP..."IA Community Project"), since I was the one who initially described it to Randy.
--There is a group of designers (I think there are 4 of them) who head up the IACP; they come up with the new designs and all of the official changes go through the design team. People on this team, along with some others, also handle communications with the community, manage the website, write articles, and manage the graphic design of new cards, etc.
--The year is broken up into a rotation of seasons: Competitive Seasons and Playtesting Seasons. Whenever a new set of cards is released for playtesting (they're currently in Season 7), that begins a Playtesting Season, during which the entire community has access to all the cards.
--During a Competitive Season, it operates largely as we do, with competitive tournaments (with prizes), Swiss rounds, etc. There are also Competitive Leagues, which basically function as a single-day tournament that is spread out over several weeks.
--During a Playtesting Season, all of the tools are updated (I created and maintain the IA Vassal mod, someone else does the same for the TTS mod, someone else maintains a squad-builder app) with the new stuff.
----A Playtesting League begins, during which players have scheduled games in which they try out the new cards and try their best to come up with broken combos and interactions.
----Statistics are tracked for each new card (frequency of usage, win %, etc) and players are encouraged to submit their thoughts and have discussion (via the community's Slack channel) on anything being playtested. This is frequently where the solution is found: for example, the new Din Djarin was winning too often, and it turns out that he needed his Damage toned down by 1pt and one of his abilities adjusted, and now he's properly balanced.
----During a Playtesting Season there are 3 phases, at which point mid-season tweaks are announced (and the tools updated again), so that players can playtest the new version to determine if the tweak was effective.
----At the end of the Playtesting Season a Playtest Survey is sent out to the whole community, where players vote on each new card (which has often been tweaked via playtesting in some way), either to approve it exactly as is, or to not approve it (if they don't approve it then they're asked to indicate why...usually because it's either too strong or too weak). I think the margin for approval is 70% for each card; any card that has a high enough approval rating becomes official. If a card is not approved then, depending on the reasons why that is the case, it is either postponed for more changes in an upcoming Playtest Seaon or scrapped and started from scratch.
----Once a card is approved then it is approved forever and is usable in all Seasons (Competitive or Playtesting) going forward. But if it later becomes apparent that a card needs adjustments--either because it has too much synergy with a new design or because it just turns out to be underwhelming--then the community is asked if they'd like to tweak it. This is relatively easy to do, since all the cards are electronically supplied, and people print their own cards.
----All of the changes ultimately come through the design team (because too many cooks spoil the broth), but the team is extremely receptive to community input and frequently draws its solutions directly from playtester suggestions.
I no longer play IA, but I've been keeping up with it from the outside, since I still update the Vassal mod for it. Regardless, if anyone has questions about how this process works then I'd be happy to try to provide answers if I can.