One side believes segwit will work, the other side believes it will make very little difference.
You're kidding me... you really think segwit won't "work" the way it was intended? It's been deployed for over a year on testnet (unlike segwit2x) and has been deployed on litecoin. Sure litecoin has precious little use for it but it's still in use and does exactly what it's supposed to. Or are you saying that exchanges and users simply won't use segwit transactions? Exchanges will jump ASAP for the benefits it provides and they do the bulk of the transactions (except during mempool spamming and then it's 'someone').
Work meaning to have no backlog thereby reduce fees and lower confirmation times, that's all I want.
I don't doubt the technical aspect of it working, but nobody knows
how much it will help with the above until it's implemented.
Segwit is coming and the question will be answered however.