Anyone heard or xsight labs? They are making networking gear for AI and spacex satellites. I'd be curious if anyone can shine a light on:

- Does full switch programmability actually matter in production, or is it one of those things that sounds great in a slide deck but nobody uses?

- Is "on-path cores" in a DPU genuinely better than Pensando's P4 approach or NVIDIA's off-path model, or does it converge at scale?

- Can you actually swap merchant silicon under SONiC in practice, or is Broadcom lock-in deeper than the software layer?

- Is there a real market for a power-optimised 12.8T switch when the spine is moving to 51.2T+, or does the spine vendor always drag the leaf along?

- How big a deal is open-sourcing the ISA? Do people buying this really care to pay up for this type of thing?

Thanks!!