Way to reduce # of 5 star ratings

Actually, the reason why I do this is that if an album gets 5 stars from me, it is exactly the same as any other 5 star album.  A masterpiece is a masterpiece, you can't have a *better* masterpiece, because that implies some imperfection, which is not the case in a true masterpiece.  The same goes for a one star rating.  If it's deserving of 1 star, it has no redeeming qualities, and is therefore just as bad as any other album without any merits.

Another major problem, at least in my eyes, is the backlash received here by fans if you rate something here poorly.

For my own case I've had several occassions where I was more or less yelled at and abused because I rated a particular album poorly.

You also will lose a lot of friends if you attempt to be fair here. That's another issue, as I think some members here have some growing up to do.

I'm going to keep doing things the way I'm doing them, leading the cause for balanced rating systems at PA.

Usually I don't like reviewing albums I don't like, because it means that fans of the band/album will be disappointed when reading the review. But sometimes it's necessary ... and after all these reviews tell more about a reviewer than the 1001st review of Close to the Edge.

I guess that the most effective way to write interesting reviews is to simply focus on albums which you see differently than the "public opinion" - works both ways (negative/positive).

somehow I think that your post lacks context.

How about redesigning the algorithm to include the number of reviews a reviewer has made? For instance, a reviewer who only reviews 10 albums will have a smaller effect on any given album's total rating score than a reviewer who has reviewed 100 albums. The more albums you review, the more of an impact you will have on the rating score of any given album. The effect of consistently doing more and more reviews would also retrospectively increase the impact on previous albums the reviewer has reviewed. In addition, some checking on score spreads, as noted previously, would lower the impact if a reviewer gives everything five stars or one star and has done it to say a couple hundred titles.

I believe something similar to this is used at rateyourmusic.com. I don't know if such an idea is statistically helpful as I'm not a statistician. But at the very least, I think it's worth discussing.

progaeopteryx wrote: How about redesigning the algorithm to include the number of reviews a reviewer has made? For instance, a reviewer who only reviews 10 albums will have a smaller effect on any given album's total rating score than a reviewer who has reviewed 100 albums. The more albums you review, the more of an impact you will have on the rating score of any given album. The effect of consistently doing more and more reviews would also retrospectively increase the impact on previous albums the reviewer has reviewed. In addition, some checking on score spreads, as noted previously, would lower the impact if a reviewer gives everything five stars or one star and has done it to say a couple hundred titles. I believe something similar to this is used at rateyourmusic.com. I don't know if such an idea is statistically helpful as I'm not a statistician. But at the very least, I think it's worth discussing.

Peter Rideout wrote: Who's going to buy those albums which I probably won't like for me? You? Max or Ron? (Certainly not me!)   Or should I illegally download them, then spend lots of my limited leisure time listening to music I don't care for? Why? Where is my motivation to do that?

You live in Canada ... one of the big advantages of this magnificent country is that Napster is available there. So for the price of a few dollars each month you can have access to all these albums that you wouldn't normally spend money on.

Having people review bands they don't like would be like what is happening now, just in reverse.  You would still not be getting an unbiased opinion.  Music reviewing can never really be conducted in a vacuum, even by professionals.  But do I have an answer to fix it, no. 

But one thing I do really appreciate in a review is the referencing of the band's other albums (or albums of like bands).  When a new album comes out, any references to previous albums and how it compares does more for me than just trying to describe sound with text.  Maybe that could be incorporated in the suggestions for reviews.

^ fair enough, who has time to listen to bad music ... but it's not fair to complain that Napster is not a free service ... after all, good music doesn't just make itself. 10 bucks for more than 2 million tracks still sounds quite fair to me.

BTW: I also couldn't care less about the percentage of 5 or 4 star ratings in my profile ...




Edited by MikeEnRegalia - November 08 2006 at 16:06

MikeEnRegalia wrote: ^ fair enough, who has time to listen to bad music ... but it's not fair to complain that Napster is not a free service ... after all, good music doesn't just make itself. 10 bucks for more than 2 million tracks still sounds quite fair to me. BTW: I also couldn't care less about the percentage of 5 or 4 star ratings in my profile ...