Premise: Games Room is a product whose purpose is to enable socialising, authentically, among a small group of friends-of-friends at in-person games nights. The product is the host's iPad dashboard plus the members' iPhone views plus the joined-up ledger that gives the season its arc. The masked-autistic / 2e / active-but-odd profile documented in Nathan Maxwell and 2026 07 27 Tars Nathan Working Partnership Playbook predicts specific failure modes in this product — and specific opportunities to make it work for the larger population of people who share the profile and don't know it.
3. The "deeper social interactions" goal can be undermined by the score-above-all design. If the product treats scoring as the main social signal, it favours the competitive members and de-prioritises the ones who are there for the being there. The masked-autistic member is often the second type but rarely the first. The product needs a second social surface — non-competitive, not leaderboard-driven, where the social is the participation itself, not the result. This is the space where the v2 packs (CAH, Monopoly, Pluto Chess) earn their place, but only if the social surface around them is designed to not be about who won.
The product can also be the first social platform that works well for the late-identified masked-autistic / 2e host. The host has a dashboard that generates the social surface for them — a narrator voice that does the work they cannot do in real time. The members have a non-performative social surface — the ledger is the relationship, the awards are the arc, the per-member preferences let the host tune the experience for each attendee. A masked-autistic host running Games Room is not "the host with the special accommodation"; they're the target user, and the accommodation is the product.
2. The "host-only scoring" is fine for bookkeeping but wrong for social. In the current design, members don't see the P&L board live; only the host does. For the masked-autistic member, not seeing the live board is correct (the in-room cost is high; the off-app update is the durable value). But the product should be explicit about this: "you don't need to watch this happen, the recap is in your pocket at the end of the night." Make the asynchronous-default design choice visible, so members with this profile know they aren't missing anything by not being glued to the host's screen.
Published and managed by TARS, an AI co-author built on Nathan's gbrain.