Great questions!
V2 is technically out, the proposed Simple DVT module itself is based on the architectural foundation that Lido V2 enabled by moving to a modular framework. I don’t think the Simple DVT approach can bring 5000 new node operators, but I do think that it’s a useful way to battletest the modular design and relevant tech infra so that modules developed in the next 1-1.5 years can (including with permissionless aspects)!
Will leave the respective teams to answer this, but from my perspective in order for the module to go live on mainnet I would like to see audits/reviews of whatever client and on-chain code versions would be utilized as at the time of mainnet launch plus all findings (apart from INFO/LOW) addressed or sufficiently explained if they’re of the “won’t fix” variety.
Important point! I think it would be important for the LNOSG to put forth the criteria as the testnet develops (since there are a lot of dependencies here), but I would suggest that they be primarily of a performance nature and exclusionary criteria might only be something like indications of sybiling, malfeasance, etc.)
It’s what exists in the cover vault (it has since grown since the funds were set there as it’s denominated in stETH) since the self-cover option was adopted (see Snapshot and Aragon vote 134). It was basically most of the rewards from the first 1.5 years of operation of the protocol. Discussions about if that amount should increase and how are also ongoing.
Once there’s a few offers provided then I think this analysis can happen. IMO the risk here is more about technical risks associated with DVT infra (especially at a scale not yet run) and being a “first mover”, so the question is whether the DAO would consider it prudent to source additional cover specifically given the circumstances of the “early adoption” of the tech, rather than stuff like slashing events.