Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
The halfway house for at-risk ideas

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.




Two-dimensional rating scale
  [vote for,

So many critics rate things on a one-dimensional scale of 0 to 10. Now, there are often different attributes; for instance, computer game critics might rate on four scales of "graphics", "sound", "gameplay", "price". You could say that's a four-dimensional scale, but it's really four one-dimensional scales.

What I'm proposing is a true two-dimensional scale for rating stuff. Here's how it would work. You'd have two non-negative values in the form a/b, where a is at least 0 but at most b. b represents how important this scale is for the product; a represents how good the item is on the scale. For example, say the scale was "graphics". Then:

10/10 : "This game has awesome graphics that really make the game enjoyable." 0/1 : "This game has crappy graphics, but when you play this game, you don't care about graphics." 1/8 : "This game has poor graphics, which really detracts from the game." 3/6 : "Not much to say about the graphics; they contribute a bit but they're pretty average."

Two advantages to this system: (1) Critics no longer have to feel uneasy when their ratings system doesn't do a critiqued-item justice. For instance, say a movie critic has a scale for "Plot" (because it's important for a good movie to have a good plot). But say the critic just watched a really good action flick with a crappy plot, but still enjoyed it. The critic could plump a "0/2" on the "Plot" scale, and now the other scales are shown to be more important. (2) You can use arithmetic to combine several scales to get an overall rating. For instance, if you added all the first numbers to all the final numbers, you'd get a true rating that weights the importance of each scale in the critic's view (instead of most current systems, where adding 'em up forces you to treat all scales equally). You can even give users a customized scale; for instance, say a user thinks "sound is really important for me in a computer game". He could get a customized review for him where all the sound scores are "scaled up double", i.e., 1/3 becomes 2/6, 5/10 becomes 10/20, 8/8 becomes 16/16.

Another tangent: some web sites like using color to represent their scales. Usually they vary the hue; for instance, a site might use red for crappy, and green for great. We could represent the "importance" number by saturation; so as the scale becomes less important, the color becomes greyer and greyer.

weihwa, May 07 2001


       Well, PeterSealy, it is the same and it isn't. There are extra properties that are useful. For instance, the Cartesian x,y plane is isomorphic to the complex plane, and both are two-dimensional, but they serve very different purposes. The complex plane introduces extra concepts and operations (such as taking square roots) that have a practical use that the Cartesian plane wasn't intended for.   

       Same thing here; most review sites that use a multi-dimensional rating scale do so because they want to convey extra information that they couldn't do with a single scale. Sure, this is just proposing another scale ("importance"), but it's intended to be useful.
weihwa, May 07 2001

       Cool; I'm not aware of any rating system where the judge gets to choose the relative importance of each attribute. (That is, I don't think it's Baked, despite what PeterSealy says.)   

       In any case, the difference between "a four-dimensional scale" and "four one-dimensional scales" isn't material to the idea at hand.
egnor, May 07 2001

       I like the idea, but if graphics aren't that import to you in a game, then you aren't going to pay that much attention to the graphics rating.   

       [I'd give this idea 2/2, where the idea isn't that important, shouldn't this be getting greyer?]
[ sctld ], May 07 2001

       Do you have an example, Peter?   

       [sctld]: The point is not whether or not graphics are important to you per se, but whether the graphics significantly contribute to the enjoyment or lack thereof of this particular game, in the reviewer's opinion.
egnor, May 08 2001

       Nice idea. I think it would remove some ambiguity and give more information to work with in evaluating ratings on something. I haven't really seen widespread use of anything like this (that I can remember).
PotatoStew, May 08 2001

       [Mephista]: You misunderstand.
egnor, May 08 2001


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle