I fail to see how the cast is in a nonsense order. The main cast (first 10 people) are in the same order as they are in AFI's catalog entry for the film; and all the uncredited roles are listed last in accordance with the guidelines.

Thank you for your response and an opportunity to discuss my comment. The order is nonsense because it does not reflect the relative importance of the various characters in the finished film.

The treatment of the "uncredited roles" is particularly out of logic in that the character played by Nedrick Young is reduced to the very bottom of the long and substantial overall cast list by the unreasonable use of last-name alphabetical order to organize the uncredited players. Young's character is, in fact, a significant supporting role, occuring throughout the story and essential to the story's finale. Why Young was omitted from the screen cast I don't know, but he certainly has more to do than, say, Charles Buchinski (Bronson) as a mute bogeyman or the fellow who bats paddle-balls incessantly at the screen at one long point solely in order to demonstrate that the filming was in 3D. Meanwhile, other smaller roles and rank extras are all jumbled together and shoved in on top of Young, and anyone interested in finding who played his role are left to scroll aimlessly around the long list of uncrediteds to find him or any of the other performers who were omitted from the slender screen cast. This method of cast list databasing is not user friendly. Adding insult to injury, this method in TMDb's "The Guidelines" is nothing but an import from IMDb, not even original.

I have no idea what Afi used to determine the "main cast" but it simply replicates the screen cast (minus Young, plus paddle-ball guy) which places the film's hero after the secondary police inspector, probably based on advertising "name value", or political "contract negotiations" and having nothing to do with importance of the roles to the story. By your citations of AfiCat and "The Guidelines", I take it you haven't seen the film, or at least taken much notice of its characters and performers. Those of us who had can see the nonsense easily, and should at least be able to get in there and organize things better as defined our understanding of the film, not by arbitrary and ill-considered "guidelines" taken, unquestioningly, from other sites or sources. At best, these methods may be useful as starting points for bulk data entry, but hardly for a finished product.