Actually I'm surprised that reddit has something like this ...anyway they seem to gather articles from several scientific websites and newspaper science sections and that's it.
For me there are two problems: The way articles are selected: it seems mostly stuff which is interesting enough (or even audience-grabbing) for non-scientists, so it is a very biased selection and surely not giving a good and exhausting overview about what's going on in science, but more a type of tabloid press view on science.
And this leads to the second problem (IMO): the hundreds or thousands of comments for each article which are mostly not worth reading (at least the samples I read were just annoying/boring/stupid/redundant).
Therefore if I want this I'd prefer the original source (Science, Nature, Sciencedaily etc) or directly from my newspaper where usually the same science news are printed (as in all newspapers).
This is something I never understood: often websites give a more understandable article about a paper ... but they almost never give the link (or title) to the paper that they used to make the article. I do not understand this. Even good newspapers like the guardian do this (or even more scientific newspapers/websites often do not give the link/title of the article used.)
Not sure if they really read them (and some journalists perhaps even don't know if such a paper exists ) or if they use the text from e.g. university press office or other sources that give an easy to read abstract.