I know it's always a dangerous move to make generalizations like "this entire field of journalism should be like this, not that," but I'd like some feedback on this thought. When I'm not programming, I like to write and maybe work on some video, and I recently had the idea to start trying to critique games in some atypical ways. We're all familiar with the Gamespot/IGN style, right? Their reviews are generally pretty shallow. Even though they don't break the game into categories like Gameplay - Audio - Graphics - Story anymore, you can still see them trending toward that in their writing.

There are people doing things a little differently. Rock Paper Shotgun for example has a pretty unique style; but I think they focus too much on plot and setting, and sometimes they go a little overboard with the writing. Sequilitis is a pretty amazing (if short) video series that breaks down games focusing almost exclusively on design, but, like The Angry Video Game Nerd, it only critiques old games. That sort of thing has its value, too, but it's easy to criticize old games. We've had so long to look back on the paradigms of those days.

So I'm interested in trying to find a new way to bring all of those styles together to critique modern games. I'd like to make a video series that developers and players can learn from--the sort of thing that's interesting and educational.

So do you have any input on this? What direction would you take something like this if given the task?

So a blend of Angry Joe and Bennett The Sage ?

Game reviewers are all over the place. Here are some more you missed. HERE and HERE

You have to be very entertaining and unique in your approach to your videos, or you'll get lost in the very large crowd.

The trouble with using reviews as educational critiques is that you don't necessarily know the personal tastes of the writer. As a gamer the best way to use the information is always to find a reviewer with similar tastes to your own, because what they say will likely hold weight for you personally.

Imagine I say:
"I absolutely LOVE racing games but I've never really understood or been into first person shooters. By the way, don't get the new Halo game because I hated it".

Obviously the second sentence is invalidated by the first one, in terms of useful objective information, and I would hope that example is more extreme than anything in typical videogame journalism, but I do think unseen biases (that quite likely don't match your own biases) are skewing the words you read a certain way.

We all have biases. And I think in fact game reviews would probably be boring if they weren't able to reflect the writers themselves as individuals. So I don't think it's a bad thing by any means, but if you want to be usefully informed you just have to know to some degree the writer of whatever you're reading.

In terms of what I'd like to see done, I have had a slightly related idea for a while that I think would be cool.

Basically, a site that aggregates videogame reviews and lets you rate the reviews you read. These ratings of reviews wouldn't have to be public - just on your personal account, as a tool to allow you to see reviewers that consistently match your individual tastes rise to the top of the mix (this would be different for every user). So each user for this site would see lists ordered from top to bottom based on their own input over time.

The evolution from there would be to have last.fm style "neighborhoods", whereby you're linked with people that share similar tastes to your own, opening up game discoverability by association and recommendation.

I guess in terms of tying this back to learning for game developers - maybe that would be a matter of letting users opt-in to data sharing for the larger gaming community to analyze associations and "clouds" of gaming tastes, and individual developers would also be able to see which reviews of their games had the greatest number of "agreeing" customers in order to know that those critiques are the most significant.