The FIVE STAR rating system

I disagree with ratings being more important.

A rating says nothing to me, I don't even know if the guy has heard the album or is rating just to manipulate charts or because he loves a determined band.

The point 5 system might be a good add-on as many already state this rating in the review.
But coming back to the point about the abundance of 1 or 5 star reviews - over and over again you have to ask yourself who could or should rate the ratings ? Maybe I would allow Ivan the right to correct low ratings on Styx releases, and the T can take care of Dream Theater albums
In essence, IF I am allowed to rate  & review, then I should be allowed to judge the albums as I see fit. Expertise, knowledge, heck, even good taste , are not mentioned anywhere as criteria for an acceptable review. 
The other sticking point that we keep coming back to, and this one is spelled out plainly is the "prog value".
And again, who judges who. It is true that many give it a 1 or a 5 based on the music without any consideration as to its' progness. But then should Foxtrot be judged less prog than say Henry Cow' Legend ? Is King Crimson's ITOTCK less prog than Yes' Close to the Edge ?
Can you see the endless debate that will settle nothing ? Because no objective measure will magically come into existence that all could easily apply in sharing their opinion of an album.
Instead, let's discuss why Ange is slighted, and Klaatu is prog related; while Arachmoid are praised and Barclay Harvest are accepted as crossover/symphonic whatever prog

Personally, I think that ratings can either be useful or worthless, depending on the situation.  In particular, I think it depends on what motivates you to be reading reviews of an album in the first place.  For example, if I'm browsing through ebay and find a cheap used CD and I want to know if it's something I'll like, I'll head over here and read some reviews.  In this case the rating is completely worthless because I'm basing any decision on the reviews. 

If on the other hand I decide that I want to try getting into a new artist that I have little knowledge of (especially one with a big discography, say Frank Zappa), I'll head over the the artist's page and I'll need to decide which albums I want to read the reviews for.  So naturally I'll gravitate towards the albums with the higher ratings (for better or for worse).  You can't expect a person to read through reviews of 50 albums in order to find the best album to start at.  So I'll look at five or so of the higher rated albums and then read the reviews.

In either case the reviews are the most important thing and are what I'm really using to determine whether or not I'm interested in a particular album (combined with sample tracks if available, etc.).  But It's entirely possible that the ratings help me out along the way.  So I think that they really do serve a purpose, even if they aren't as important as the reviews.

Since I think they're important, I think it's a good idea to be able to rate albums on a ten point scale as that makes the rating a more accurate measure of this community's thoughts on the album.  Of course for an album like Close to the Edge it's going to make little to no difference.  But it could make a big difference for an album with only a few ratings.  Although I agree that more emphasis should be placed on the review than on the rating, I think we all know that many people around here rely on the ratings more than they should.  And if that's the case, I think it's wise to make the rating as accurate as possible.  After all, allowing half points could very well change the rating by a couple tenths of a point if there aren't many ratings.  And if a user's primary means of determining what albums they should get is the ratings, then this difference could be very important.

Rubidium wrote: I think it's wise to make the rating as accurate as possible.

NotAProghead wrote: Rubidium wrote: I think it's wise to make the rating as accurate as possible.    To those who think half stars or 10 stars systems are better: how these things should be displayed? Now it's easy to see ratings on album pages.  Counting 6 - 8 stars is not convenient at all.   Those who really need information can find it even with existing rating system. Fans of statistics will be always searching for the ways of improving and one day someone will find that 100 stars system is much more accurate than 10 stars.

It'd be best to keep the maximum at 5 stars, but allow for rating at half star increments.  That way you don't have to worry about counting 8 stars, which I agree can be cumbersome, although it'd be easier in a case like that to count the number of stars that AREN'T shaded in (so that you'd only count to two instead of eight).

[Warning: The remainder of this post may only make sense if you understand the inner-workings of my mind, which is a very dangerous thing]

Theoretically we could partition the five stars into as many equal increments as we want, and still make it out of five stars maximum (at the moment there are 5 increments, and 10 is being proposed in this thread).  We could even partition it into a million increments if we wanted to without any technical problems, but of course that wouldn't be practical.  The key would be to choose the optimal number of increments to maximize the accuracy of the ratings while still being able to determine the difference between ratings in one's head (for example in my mind I know what an album rated at 3.5 star album is like [namely one that is better than what I would consider a typical 3 star album but not as good as a 4 star album], but I don't know what the difference between a 3.5 and a 3.6 album would be, aside that the 3.6 album is "slightly better").  I think 10 is a simple and natural number to use, although some would argue that 50 would be better, as I've seen many reviews were the reviewer states that they give the album 3.6 stars.  Heck, I've even seen a rating of 4.71, where I assume that the reviewer has distinct criteria in mind as to how a 4.71 star album is better than a 4.7 star album.  I find it difficult to believe that reviewers are capable of distinguishing between two such albums on a consistent basis, and are more or less pulling random numbers out of the air.  As the number of increments increases, the rating will converge to the true population rating.  So in this sense, more increments is better.  But as the number of increments increases, the marginal improvement will begin to diminish and we will reach the point where the infinitismal increase in the accuracy of ratings is outweighed by the general confusion of people giving Close to the Edge a rating of 4.78273 stars, Foxtrot a rating of 4.77325 stars, etc.

Ratings should be reserved for albums listed in prog categories only - one can hardly give a 5 star "masterpiece of progressive rock" to a prog- related or non-prog album, can they?

Edited by mystic fred - July 16 2008 at 04:03

mystic fred wrote: Ratings should be reserved for albums listed in prog categories only - one can hardly give a 5 star "masterpiece of progressive rock" to a prog- related or non-prog album, can they?     gold ratings for albums in prog categories only -     Essential: a masterpiece of progressive music           Excellent addition to any prog music collection                 Good, but  non-essential                       Collectors/fans only                             Poor. Only for completionists       5 stars are clearly enough, but the ratings should only be for prog albums only as clearly described.      Essential: a masterpiece of progressive music           Excellent addition to your  prog music collection                 Good, contains some prog elements and has prog leanings                       Not really a prog album, no prog elements                             Definitely not a prog album,         or change the descriptions....     Essential: a  masterpiece of great  music and historically important           Excellent addition to any  great music collection                 Good, but non-essential                       Collectors/fans only                             Poor. Only for completionists     ..then what would be the point of a progressive rock site?  the answer would be to exclude prog related and proto prog ratings, then there would remain albums such as "90125" which are hardly prog but require ratings.     my answer when reviewing prog related is to give two ratings - a prog rating and a musical content/merit rating, so therefore Led Zeppelin's "In Through the Out Door" could ber rated thus-   MUSIC RATING 4.5   PROG RATING 3/5

Or there's the option to change the wording ONLY for prog-related and proto-prog albums. So you'd have the status quo for albums in a prog sub-genre, and have a more generic description ('a masterpiece of great music', as you suggested) for proto and related. Makes sense to me.

To my mind, reviews function as recommendations. If I see a two-star rating, I might have to read closely to realise the reviewer considers it a four-star album but is only giving it two stars because it's on a prog site. Better if the rating descriptions reflect this tendency, I think.

http://www.progarchives.com/forum/forum_posts.asp?TID=46404&KW=star
http://www.progarchives.com/forum/forum_posts.asp?TID=40798&KW=star

A subject that comes up as new threads it seems every 4 months - and I notice a number of the old suspects who respond to each do so without reference to what has gone before or what they wrote before. We have been discussing this at least from the time I joined, WITHOUT any change and as ever there are questions to the apparent disproportionate numbers of 5 star albums.

I suggest we cobble all the various correspondence on the subject over the years, into one large master file, bang this a prominent place i.e. the appendix to the guidance on star rating, to demonstrate some stubborn lack of vision locally and that we seemed to have been wasting our collective breathes for at least 5 years.

  Essential: a  masterpiece of great  music and historically important


Historical importance? Imply an album has to undergo the test of time, thereby eliminating any brand new release: I can go for that. But what  length of time before upgrading or not to 5 star?

Dick Heath wrote: http://www.progarchives.com/forum/forum_posts.asp?TID=46404&KW=star http://www.progarchives.com/forum/forum_posts.asp?TID=40798&KW=star A subject that comes up as new threads it seems every 4 months - and I notice a number of the old suspects who respond to each do so without reference to what has gone before or what they wrote before. We have been discussing this at least from the time I joined, WITHOUT any change and as ever there are questions to the apparent disproportionate numbers of 5 star albums. I suggest we cobble all the various correspondence on the subject over the years, into one large master file, bang this a prominent place i.e. the appendix to the guidance on star rating, to demonstrate some stubborn lack of vision locally and that we seemed to have been wasting our collective breathes for at least 5 years.

hahhaha.. you are priceless Richard ... and came to your post, which I missed earlier, because of a link posted in another thread that has been recently started directing them here. 

Yo!... collabs....  any of you see what is going on in the forums these days... . M@X doesn't have time for this crap.. and even before he didn't...  the half stars were apparantly shot down. Reviews.. not ratings matter... for those who write them at least hahha and ratings without reviews are wanted by the owner and that is probably not going to change. Even if they are worthless and pointless for any of us.   Why don't let people know that instead of directing them to old threads.  The logistics of doing it are a hard one to deal with in the best of times... and you all know that M@X has slightly more important things to deal with.