Kotaku didn't attend the review event; we have yet to run a review; and our audience cares more about our reviews being thorough than early. This isn't some sort of unsolvable problem. Getting rid of review scores would certainly help.
That's good. Is that a standard policy in all instances, or is it a decision that's made on a case by case basis?
I broadly think Kotaku does a pretty good job of reviews. Can't think of any instances where you've given a game a positive review and then the game releases and it turns out the review was unrepresentative of the actual game. The worst offender for something like this of the big sites is probably Polygon, and some of their worst reviews have been conducted at review events, and they've had to edit these reviews after the game has come out and they've already recommended that their readers buy it,
Certainly there is some reviews I disagree with, but that's just opinions, not a result of attending review events or bad review policy.
I guess my point is that you shouldn't really hit publish on a review unless it is a review you are going to stand by, and I can't really see how you can stand by a review with any conviction if you've not played it in real world settings, on your hardware, and with the same online infrastructure that customers will be using.
(Disclosure for anyone else reading this, I did write a little bit for Kotaku UK, but wasn't really privy to their editorial policy outside the scope of my work, and Kotaku UK is largely independent from Kotaku.com).