In reply to john arran:
I've thought a lot on this, and my only half-baked ideas are:
1) All error corrections must receive AT LEAST as much exposure as the original error. Errors include blatant lies, misquotes, and most importantly - lying by omission. For example - a page 1 lie, needs a page 1 correction.
In practice, that would mean "£350m? We could spend it on the NHS." would never have flown. It would be an error by omission to exclude the additional funding we receive from the EU (£200m a week was it?), and that it's likely our trade market will shrink causing a much greater loss of income - thus we won't have £350m extra, we will be -£400m (or whatever) - and while it is still possible to spend an extra £350m on the NHS, we would now have a £750m sized hole in the budget that would require cuts in other areas"
What constitutes an error, is a problem, and hence why we might need...
2) Some kind of reference requirements. It's not workable to require all stories to have references, but perhaps stories without them could be required to have [citation needed] on all unsupported facts. Would be tremendously unpopular at this moment in time, but I don't see much other ways. There would need to be some kind of "source ranking" as well, as obviously a newspaper can't quote itself objectively, and a scientific study has more merit than an anonymous source.
I don't see any other ways that can be more-or-less objectively policed (it needs to be as objective as possible IMO). A possible extension to 2) would be to have an automatic ranking of articles by their references, so the "truthiness" of an article can be approximated by the average truthiness of its sources (or lack thereof). That way people could actually see "oh, the DM
is full of lies!".
Post edited at 14:21