You raise an interesting--and infuriating--point here: if you do detect wrong information in an article, then how do you go about re-evaluating the veracity of the publication as a whole? As you point out, a certain level of "wrongness" is inevitable and can be chalked up to good-faith errors.

But how do you determine if it's even worth continuing to read that publication? Somehow measure the "wrongness" factor and stop reading after it hits a certain value? More than half and you're outta there?

Noam Chomsky pointed out that even extremely biased sources (he used the New York Times as an example) can be used as sources of valid information, if one reads 1) to the bottom of articles even if they're on page 73 and 2) between the lines. But that seems like quite the exercise in critical thinking.