Way too many five stars review out there!

erlenst wrote: Sean Trane wrote: Ideally of all albums in a given genre: 5* masterpieces should be 5% tops 4* excellent should be 25% 3* good albums should come to 25% 2* average albums to 25% and 1* bad albums the remaining 20%   We are far from this  in our case.   the debate about too many 5* ratings is hardly new (thisis why there is this warning about using 5* as you click it, but it does not deter people from using it) and all collabs are aware of this!! This is why most collabs try to rate more in the middle   Unfortunately the younger members always are more enthusisatic and often have a fanboy attitude and certainly lack the sense of perspective an older proghead!! Please explain why there should be 20% one star reviews and 5% five star reviews ... Makes no sense.

You will agree that the five-star scale is non-linear since good is three stars (the middle grade in a scale of 5), and that good usually means that it should be above the middle mark>

 

Therefore with three different stars ratings allocated from that 5 * scale to break down good category into outstanding 5* (which means exceptional and therefore rare> hence less than 5%) and very good 4* (or excellent) and simply the good 3* albums (worth having but nothing to write home about)

 

On the opposite side of that scale , we only allow one rating (2*) to speak of average 2* albums which we must agree should be the majority of all albums>  2* should still be a strong minority of album but given the fact that we are on a specialized site rating a genre that has many good or better albums, I reduced this part to roughly half of what it should be (hence the 25%)

This leaves the bad 1* albums (which on a linear scale would represent the lower 40% of albums) all allocating on that sole 1* rating. This sokle rating should be very crowded since it all less than average records should be in this category >>> hence I am quite conservative by stating that it should represent roughly 20% of the mass of albums.

 

Hope I could bring that concept up.

 

IMHO on a linear scale , our star system reads as such

5* should be 9,5 /10 and more

4* should be 8/10 and more

3* should be between 6/10 to 8/10

2* should be vetween 4/10 and 6/10

and 1* should be less than 4/10

I am not sure I can be more clear.

If we all agree the ratings system was a good idea and may have even worked in the early days of the site, when I daresay the members were more purist prog lovers, but has now become flawed to the point of uselessness, we can ignore the ratings system altogether and stop these continuous threads about trying to fix up the flawed system, when the owners of the site are obviously not interested in doing anything about it. Nowadays, just a simple I like, I hate scale would probably work better, but changing the system now is improbable and impractical, without a complete rebuild. No wonder the owners want nothing to do with it

Of course you can't encourage fringe dwellers into the site and expect to keep a purist rating system, but I think it is better to promote and encourage the music to new listeners, than to keep the purist rating system. Take note, that the two are at complete odds with each other- how can someone new to prog rate a work that uses a scale based on progressiveness

In my opinion writing reviews is a very subjective matter but you have to keep in mind that writing reviews for this site should not be too subjective. This sounds almost paradoxal but I notice many 5 star reviews about albums that are good but the subjective euphoric state of the reveiw lifts it to a 5 star level. At that moment you have to compare your 5 star awarding with albums like Foxtrot from Genesis, Close To The Edge from Yes, Dark Side Of The Moon ffrom Pink Floyd and In The Cour Of The Crimson King from King Crimson, then ask yourself or your 5 star rating is the best rating. I remember that I was often euphoric about prog when I started to listen to prog (in the mid-Seventies). When Marillion was my band I also bought everything that was similar to early Marillion like Deyss, later I felt a bit ashamed that I was so enthousiastic about bands like Deyss. I think that it is a bit too often the subjective enthousiasm from progheads to give 5 star ratings instead of analysing those albums more objective.

If it was up to me, I'd have separate ratings for different aspects of the album: Performance, songwriting, overall experience...

Sean Trane wrote: Blacksword wrote: Thats the thing about art of any kind, is that there is no such thing as good or bad art. The whole point of art, including music, is to express an artists ideas and feelings, and to hopefully evoke somekind of response in the listener/viewer/reader etc. If it capures an audience of just one person then it has succeeded in its aim. What constitiutes a 'masterpiece' is entirely subjective for this reason. Very true about which album IS a masterpiece, but what is a masterpiece. This comes from Meisterwerk meaning an exceptional piece of work , a submit in a artist's career and there should be at most three of four of them for a lifetime achievement. The exceptional quality factor is really the notion to be thought of when considering something to be a masterpiece. This is why I say that masterpieces should be less than 5% and even less in a general art . OR, arguably the terminology used by PA to define each star rating is misleading, I do not really think the star ratings descriptions are flawed, but the over enthusiastic reviewer is!!!, but how else do you review a piece of art if not by what you, as the critic personally thinks.

What constitutes 'exceptional quality' ??

^the star rating descriptions are flawed in that they only work IF all reviewers follow the guidelines. Obviously only a small part of them do. That will always be the case if you want to keep the open nature of the system.

On my own website I will do the following:

• People can rate reviewers. Each user can only issue one rating for each reviewer, and the ratings are very simple: ignore, neutral, agree.
• The reviewer ratings are not shown individually, but maybe you will be able to see a general hint as to the reviewer's status.
• In the calculation of the avg. ratings of songs and albums the avg. rating of the reviewers will affect the weight of their ratings.

I think that this will work really well. The general Top N lists for the genres will not be manipulated by single persons who submit irrational ratings, because most others will choose to ignore them. But they will increase the weight of reviewers which take their job seriously.

I tend to give an album 5 stars  if, for example I'm still listening to it after 20 years, and it still blows me away each time. That is a masterpiece to me. My rating has to be a reflection of the pleasure it gives me personally as the reviewer.

I would give Dark Side of the Moon, for example, 3 stars (3.5 perhaps ) as it's good, but not a masterpiece IN MY OPINION. It happens to be the album that made Floyds popularity go through the roof. For that we should be thankfull for that album. On the other hand did Floyd have to compromise on something to gain that popularity? Is the fact that millions of people happend to tune into that album, a reflection of its 'quality' or it's 'connectivity' with the masses. If the latter, then is that a good thing artistically, and if so why? In short I would never give an album a five star rating because the popular consensus is that it is a masterpiece. If it doesn't touch me in any way, then I shall review and rate it accordingly.

There are clearly WAY too many 5-star reviews on this site, but that has been the case since the beginning.  Part of the problem is the availability of only five stars; part of the problem is uncritical fanboyism.  When I rate an album, I don't review it -- I just give it a star rating.  This way, I don't have to worry about spending time writing a review for an album that doesn't excite me. 

Nevertheless, the arguments that ratings must represent a bell curve are completely erroneous.  This should only be the case -- and even then, it's just theoretical -- if the reviewer reviewed a *random* sampling of all albums released within the relevant time period, and the sample was of statistically adequate size.  Considering the tens of thousands of albums that fit the "progressive" standards of this site that have ever been released (Gnosis, for example, currently has over 52,000 albums in its database)  an adequate sample size would be fairly large -- significantly larger, I suspect, than total albums reviewed by most reviewers.  And since the albums reviewed on the site are hardly random (they were, in fact, largely heard by the individual reviewers because it was probable that they would have a positive opinion of them), I don't think that there is anything improper about raters not distributing their ratings along such a curve.   

I think the problem with the rating system is that people are either fanboys to a ludicrus extreem or have found something here that holds similaraties to what they were looking for. In the either case theres not a lot that can be done, fanboys that are tha extreem arnt going to be swayed any time soon and newbies like that may never have heard a progressive band before so they might get their minds blown by it but wont be able to compare it to anything else.

Personally when I review albums, I have to think of what is the best album(s) in that genre and compare to that, after all you cant compare CTTE with Scenes..., its just not possible. It also helps if you know several albums from different bands in that genre as well.

 

erlenst wrote: Sean Trane wrote: erlenst wrote: Sean Trane wrote: Ideally of all albums in a given genre: 5* masterpieces should be 5% tops 4* excellent should be 25% 3* good albums should come to 25% 2* average albums to 25% and 1* bad albums the remaining 20%   We are far from this  in our case.   the debate about too many 5* ratings is hardly new (thisis why there is this warning about using 5* as you click it, but it does not deter people from using it) and all collabs are aware of this!! This is why most collabs try to rate more in the middle   Unfortunately the younger members always are more enthusisatic and often have a fanboy attitude and certainly lack the sense of perspective an older proghead!! Please explain why there should be 20% one star reviews and 5% five star reviews ... Makes no sense. You will agree that the five-star scale is non-linear since good is three stars (the middle grade in a scale of 5), and that good usually means that it should be above the middle mark>   Therefore with three different stars ratings allocated from that 5 * scale to break down good category into outstanding 5* (which means exceptional and therefore rare> hence less than 5%) and very good 4* (or excellent) and simply the good 3* albums (worth having but nothing to write home about)   On the opposite side of that scale , we only allow one rating (2*) to speak of average 2* albums which we must agree should be the majority of all albums>  2* should still be a strong minority of album but given the fact that we are on a specialized site rating a genre that has many good or better albums, I reduced this part to roughly half of what it should be (hence the 25%) This leaves the bad 1* albums (which on a linear scale would represent the lower 40% of albums) all allocating on that sole 1* rating. This sokle rating should be very crowded since it all less than average records should be in this category >>> hence I am quite conservative by stating that it should represent roughly 20% of the mass of albums.   Hope I could bring that concept up.   IMHO on a linear scale , our star system reads as such 5* should be 9,5 /10 and more 4* should be 8/10 and more 3* should be between 6/10 to 8/10 2* should be vetween 4/10 and 6/10 and 1* should be less than 4/10 I am not sure I can be more clear. Ok, that was an acceptable explanation.. BUT then we can reflect over wether or not there are an unproportional big amount of masterpieces in the prog scene of the early 70s. I truly believe that there is. Therefore, as far as the classic albums of the golden era go, there should should be unproportional large amount of 4 and 5 star ratings. Of course, this is easily outweighed by the enormous amount of crap that were released in the late 70s / 80s , but I don't believe this is the problem in this topic ! Makes sense ?

I have to agree with you on that, I mean look at Genesis- up to '76 a lot of 4-5* albums. After '76 the less said the better.