Reply to thread

Message

<blockquote data-quote="Tsyr" data-source="post: 433667" data-attributes="member: 354">I'll explain to you very simply why that doesn't work: It doesn't actualy measure the persons (Or, in this case, the products) ability in and of itself.As an example. Say you had a class that gave math tests, each one with 100 problems, 10 times a semester. Assume you have 50 students. 10 always get all the questions right. 10 always miss 1. 10 always miss 2. 10 always miss 3. And ten always miss four. (We are assuming this is an advanced class, I guess). Now, individualy, if someone said &quot;I got 96% on all my tests this year!&quot;, you would think they did pretty well, right? I sure would.But by your model, they would all fail and have to repeat the class. Despite having done good work. Simply because there was someone slightly better than them.Now, granted, that's a fabricated example. But I do think the point it examines is valid. I mean, if you were to assume that only a &quot;D&quot; grade would let you pass a class, by nessesity, in your system, 40 percent of every class MUST fail, regardless of how well they did. Or, to shift back to reviews, if you reviewed ten really really good products... lets say... Wheel of Time, FRCS, Rokugan, Manual of the Planes, Monsternomicon, Kalamar players guide, Oriental Adventures, Traps and Treachery, the EQ RPG, and Spycraft... Of those, 2 MUST, no matter how good they are (assuming you are still going on a hard 1-5 scale) earn a 1. Likewise, if you reviewed 10 horrible products, two MUST earn a 5. This system is deceptive and doesn't really tell you anything. For example, if I see that Psion gave the Manual of the Planes a 1, I might think he hated it, when in fact he simply loved it less than other stuff. That's not really useful... it forces me to read the entire review to understand how he felt about it. And while I do make it a point of reading the reviews, sometimes I just want a quick baseline... which your system would not give me.</blockquote>

[QUOTE="Tsyr, post: 433667, member: 354"] I'll explain to you very simply why that doesn't work: It doesn't actualy measure the persons (Or, in this case, the products) ability in and of itself. As an example. Say you had a class that gave math tests, each one with 100 problems, 10 times a semester. Assume you have 50 students. 10 always get all the questions right. 10 always miss 1. 10 always miss 2. 10 always miss 3. And ten always miss four. (We are assuming this is an advanced class, I guess). Now, individualy, if someone said "I got 96% on all my tests this year!", you would think they did pretty well, right? I sure would. But by your model, they would all fail and have to repeat the class. Despite having done good work. Simply because there was someone slightly better than them. Now, granted, that's a fabricated example. But I do think the point it examines is valid. I mean, if you were to assume that only a "D" grade would let you pass a class, by nessesity, in your system, 40 percent of every class MUST fail, regardless of how well they did. Or, to shift back to reviews, if you reviewed ten really really good products... lets say... Wheel of Time, FRCS, Rokugan, Manual of the Planes, Monsternomicon, Kalamar players guide, Oriental Adventures, Traps and Treachery, the EQ RPG, and Spycraft... Of those, 2 MUST, no matter how good they are (assuming you are still going on a hard 1-5 scale) earn a 1. Likewise, if you reviewed 10 horrible products, two MUST earn a 5. This system is deceptive and doesn't really tell you anything. For example, if I see that Psion gave the Manual of the Planes a 1, I might think he hated it, when in fact he simply loved it less than other stuff. That's not really useful... it forces me to read the entire review to understand how he felt about it. And while I do make it a point of reading the reviews, sometimes I just want a quick baseline... which your system would not give me. [/QUOTE]

Verification