That, and after the ship full of military people all caught it a second time, it seems like maybe immunity isn't a thing for this virus.
That's been discussed before. There isn't enough evidence for re-transmission.
For this specific ship, over 1000 tested positive, then out of them, 5 people tested negative twice, then tested positive again. It's more likely that they didn't get over the first time, and the two negative tests were false negatives. After all this is only half a percent of the number of crew who got infected, 1 per 200 people. Let's say that the test picks up the virus 90% of the time. Then, out of 1000 known infected people, 100 will falsely pass the next test. Then, out of those 100, you do a follow up test, and 10 will also falsely pass that test. Account for some who did actually get over it on their own in the meantime, and 5 out of 1000 falsely passing two tests then failing the third would be
about what you'd expect just from chance.
https://www.sciencedaily.com/releases/2020/05/200526173832.htmresearchers at Johns Hopkins have found that the chance of a false negative result -- when a virus is not detected in a person who actually is, or recently has been, infected -- is greater than 1 in 5 and, at times, far higher.
So the thing is, the tests individually have a pretty high false negative rate, depending on the test, and if you have a large enough sample-size of people you're testing then of course some of them will have this pattern of two false negatives followed by a positive. You need better evidence than that for retransmission, since you can always find these anecdotes by testing enough people.
The only reason that you're not seeing even more false negatives being picked up is that some of the false-negative people
will actually get better on their own before the follow-up test, or the follow-up test itself will be a false-negative too, so they'll slip through the system.
"omg someone tested positive after we cleared them the first time, they must have got it again" sounds more like they're trying to cover their ass that their testing regime isn't robust enough, rather than being proof of retransmission.