I just skimmed through some of the comments. Time to correct the common views of The Digital Picture:
But first, of course everybody needs to use their best judgement when reading a site. I agree with neuroanatomist that numbers are helpful, although they need to be taken in context. Some cameras that rank highly on DXOMark are not appropriate for some uses compared with others, and of course money is a part of that equation.
Specific criticisms of Bryan Carnathan's site I spotted here:
"He only reviews Canon." This isn't correct. He's uploaded a lot of images (which I gather he took himself) of Nikon F-mount lenses, and notes he plans to review a lot of them. He started out with Canon, but he has reviewed Sigma lenses (his comments on the 120-300mm f/2.8
"He never comes right out and says something is junk." Maybe he's a bit more diplomatic than some, but he definitely gives a bottom line for each piece of equipment. Sometimes he says a lens is one of his favorite, most-used, or most recommended (from memory, the 70-200mm series covers the first two, and the 17-55mm and the 100-400mm L are examples of the last two), but sometimes he also plainly says when a lens is cheap and doesn't surpass its price segment in terms of quality and even that it should be avoided for an alternative. For the Canon 50mm f/1.2, for example, he says that it isn't sharper than the f/1.4 at some apertures, and the price is hard to justify unless you need its qualities. Seems like a fair assessment to me. I also note that he went to pains to do a side-by-side comparison of the defocus area quality (boke) of all three lenses in that focal length because they are commonly compared. Can't ask for much more.
The resolution charts are amazingly complete and comprehensive, I'm not aware of anything like it elsewhere on the web. (Although I wonder how many lenses look better or worse than their usual uses due to apparently being shot at different distances for the sake of a consistent magnification; some lenses seem to perform much better in close focus - that's been my observation, anyway).
Ken Rockwell (avoiding the temptation of a direct comparison) may occasionally put in "Photoshop lens correction factors," so there are useful resources out there if you know where to look. You might have to go through some text to get what you need with no guarantees, on either site, but all the information is generally helpful. Ken Rockwell is big on talking about compatibility with film lenses (in this day and age!) and Bryan Carnathan's reviews tend to be a complete tour of a lens (from his perspective but he covers things that should be useful for most everybody). Both guys could do more, but I don't pay them a subscription to put out content that I personally need, so I see no need to complain. When I want more information about a lens (i.e. is it parfocal and/or does it suffer focus breathing) I know to look other places as well (though sometimes you can infer that from the text).
My own personal (small) gripes are that Bryan sometimes does not really demonstrate the possibilities of a lens (mainly with TS-E lenses where he really could grab the bully pulpit and show off some of their great effects which people generally need more help grasping) and sometimes the image samples are hard to judge and not representative of all uses of a lens. But, like I said, this is not a big problem because other resources on the Internet cover those angles.
Photozone.de is another useful site, but sometimes you have to just look at the data and do your best to compare across formats (you can make a guess if there is a lens tested on both that you are using as a central comparison point) where there are holes in the data, or when the stated conclusion seems to be flatly contradicted by the actual test (the recent review of the Sigma 120-300mm f/2.8 APO EX DG OS is an example of this - given between three to almost four stars in optical quality on full format just because the edges of the frame get marginally softer than the center; all the samples shown pretty much completely contradict this finding, and some are outstanding and couldn't be expected from any other lens on this format, so yes, Photozone falls over too sometimes).
The only way to make a stab at overcoming bias is to read, read, read and view EVERYTHING you can find on the internet (and if possible investigate in person and talk to users in person) about a given piece of equipment.