Only needs to be marginally watertight, enough to survive just one hour in (only) one meter of water
Well, in a submarine perhaps but airplane terms that is not really 'marginal'!
DO160F 'Waterproofness' doesn't even have immersion listed as a test method.
If the ELT is specified by the manufacturer as "Permanently Hermetically Sealed and airtight' (I don't know if that is the case, but I doubt it) then DO160 exempts it from further waterproofness testing, if not then the relevent tests for airplane use will be in DO160 Section 10 and will be selected from four test methods, in increasing order of severity:
'Condensing Water Drip'
'Drip Proof'
'Spray Proof'
'Continuous Stream' (this one is fun - think big pressure washer jet aimed at the most vulnerable joints and connectors!)
The choice of test will be made based on equipment type and it's location in the airplane. I don't know which was selected for the ELT in question.
So to say "787 is to wet for electronics" is clearly inaccurate however there may need to be a review of matching test method to equipment type and location if, and only if, there are repeated and multiple examples of random bits of electronics failing due to water ingress. I don't think we have any facts to show that yet?
Don't confuse occasional poor manufacturing quality with design failings - and before I get jumped on, I accept repeated manufacturing errors can indicate poor design of course.
....such as several occurrences of the same wires getting pinched when a battery that is designed to be replaced in service is fitted for instance....... (This example picked entirely at random you understand!)