Oh sure, for something of particular interest, which is indeed what I do, and publications like NS will always provide links. But not everything can be of particular interest, so you do need to be able to have a degree of trust in the reporter's qualifications, honesty and rigour. Something like NS has a reputation to uphold, which helps - for any claim of substance, they will always contact at least one other expert in the field to find out what they think about it. Where I'm often left shaking my head is with the misrepresentations, misunderstandings and lack of checking (whether wilful or accidental) you find in the generic 'quality' press like the Guardian, etc (and all too often the BBC). There, you really need to follow the links every time, so they're of limited value.
I agree on your characterization of the relative accuracy, I'm just not seeing much of a need for NS either in a discussion. Either the links correctly support the claim being made, in which case one could just give those links directly, or they don't, in which case the article does not support the claims and hence should not be used at all.