This shouldn't be down voted. While photonic-bandgap fiber can approach speed of light in a vacuum (to ~99.7%) it is not commercially deployed in any volume AFA

This shouldn't be down voted. While photonic-bandgap fiber can approach speed of light in a vacuum (to ~99.7%) it is not commercially deployed in any volume AFAIK, and standard fiber propagates signal at only about 70% of vacuum. Starlink will of course involve a minimum fixed distance penalty of a few hundred kilometers on each leg and some amount of inherent constellation bounce penalty each sat. Even so over distances of a couple thousand kilometers or more having the signal go near maximum (even up/down, air is not vacuum but with an index of refraction of only 1.0003 is damn close) vs 70% actually looks to be quite significant. I see no reason while standard fiber couldn't be beaten even under optimal point-to-point conditions (like NY to London, around 5500 km) if they were able to get it up fast enough in the near term. And that's point-to-point from an HFT viewpoint, whereas of course in most situations fiber will not be even close to taking a direct route between locations. Even if photonic-bandgap fiber got deployed for transoceanic or continental backbone links Starlink could still of competitive practical latencies if the satellites are orbiting low enough and impose a low enough overhead.