Author |
Topic Search Topic Options
|
bl6464
Forum Newbie
Joined: September 14 2012
Location: FL
Status: Offline
Points: 11
|
Topic: Ratings Posted: November 11 2016 at 14:57 |
Ok, I get your point and the reasoning behind.
On another topic, who classifies the type of genre of a band or album ?
This band has been classified as Crossover Prog ? This is definitely not Crossover, I would qualify it as Neo-Prog, though they have their own personality.
|
|
siLLy puPPy
Special Collaborator
PSIKE, JRF/Canterbury, P Metal, Eclectic
Joined: October 05 2013
Location: SFcaUsA
Status: Offline
Points: 15254
|
Posted: November 08 2016 at 10:50 |
Maybe we should demand DNA samples and ID checks in order to open an account and if anyone abuses the ratings they will be awoken in the wee hours of the night by the prog police. Seriously, how about more scrutiny for opening accounts in the first place so that it's not so easy to open up 1000 accounts. Damn, who has that kinda time? Do these people not have lives?
|
|
Dean
Special Collaborator
Retired Admin and Amateur Layabout
Joined: May 13 2007
Location: Europe
Status: Offline
Points: 37575
|
Posted: November 08 2016 at 09:44 |
bl6464 wrote:
I understand your point, however if the goal is to motivate and recognize people that write a review, there are other ways to reach that goal without biasing the ratings. For example, give an incentive for the first reviewers. You can do a crowd funding effort to collect funds for this purpose. Give away a CD from one of their favorite bands, gift cards, etc. You can count on me to donate for this.
Also, if you believe your weighing system is motivating people to write reviews, then show both results. General Rating and Reviewed Rating (Counting only people that wrote reviews) and explain to the audience what is each.
I love this site, it is my go-to source for anything related to prog rock. I decide what to buy based on these reviews. That is why I want it to be better and reflect the unbiased opinions of the community.
Hope this helps
Best, Ben
|
That isn't understanding my point at all. In fact it's missed it by a country mile. The purpose of the weighting is not to encourage reviews or discourage rating-only. It's simply to limit the damage caused by stupid idiots who think they can manipulate the ratings of their favourite artists by creating multiple accounts or by manipulating the chart rankings in the Top 100 charts by maliciously giving 1-star reviews to hundreds of albums they've never listened to. As I said, we'd rather not have to do that but these muppets (of which sadly there are many) give us very little choice in the matter.
If we differentiated between review-rating average and rating-only average it would actually render the rating-only result as a meaningless number that no one in their right mind would ever believe.
~~~~~~~~~~~~~~~~~~~~~~~
Trust me on this, when I was an admin I spent hours on end manually computing averages every which way and can assure you the rating-only values would be completely unrepresentative and misleading because of deliberate rating manipulation. To give you some idea of how bad that can be on one evening I deleted 1000s of ratings from one clown who had used over 150 fake accounts in an attempt to manipulate their favourite band's latest album to the top of that year's album chart.
~~~~~~~~~~~~~~~~~~~~~~~
Deciding what to buy based in reviews - great. Deciding what to buy based on average ratings without reading a single review - not so great, but frankly what's the difference between 4.16 and 4.67? It's still a 4-star album and this ain't an exact science.
|
What?
|
|
Guldbamsen
Special Collaborator
Retired Admin
Joined: January 22 2009
Location: Magic Theatre
Status: Offline
Points: 23104
|
Posted: November 08 2016 at 09:33 |
Any "fair" system you care to propose will always be ruined by folks who think music is a sport. "Let's win the gold on PA!!! 5 stars to my boys!!!....and 1 to the rest of you feckers hoho" he says while making several similar accounts on PA where he continues this rating procedure. Nothing will ever stop people doing silly things on the internet.
|
“The Guide says there is an art to flying or rather a knack. The knack lies in learning how to throw yourself at the ground and miss.”
- Douglas Adams
|
|
Nogbad_The_Bad
Forum & Site Admin Group
RIO/Avant/Zeuhl & Eclectic Team
Joined: March 16 2007
Location: Boston
Status: Offline
Points: 20880
|
Posted: November 08 2016 at 07:43 |
It wouldn't be unbiased, it would just be more reflective of fanboys, trolls & serial spammers. This is the best approach arrived at over a large number of years based on all the factors impacting it. The least worst option.
|
Ian
Host of the Post-Avant Jazzcore Happy Hour on Progrock.com
https://podcasts.progrock.com/post-avant-jazzcore-happy-hour/
|
|
bl6464
Forum Newbie
Joined: September 14 2012
Location: FL
Status: Offline
Points: 11
|
Posted: November 08 2016 at 05:49 |
I understand your point, however if the goal is to motivate and recognize people that write a review, there are other ways to reach that goal without biasing the ratings. For example, give an incentive for the first reviewers. You can do a crowd funding effort to collect funds for this purpose. Give away a CD from one of their favorite bands, gift cards, etc. You can count on me to donate for this.
Also, if you believe your weighing system is motivating people to write reviews, then show both results. General Rating and Reviewed Rating (Counting only people that wrote reviews) and explain to the audience what is each.
I love this site, it is my go-to source for anything related to prog rock. I decide what to buy based on these reviews. That is why I want it to be better and reflect the unbiased opinions of the community.
Hope this helps
Best, Ben
|
|
Dean
Special Collaborator
Retired Admin and Amateur Layabout
Joined: May 13 2007
Location: Europe
Status: Offline
Points: 37575
|
Posted: November 08 2016 at 05:07 |
bl6464 wrote:
I didn't know the ratings were weighted. Sorry, but I think this average is biased. Not everybody has the time to write a review or belong to the collab team. In my opinion, the rating should reflect the plain average of the person evaluating as in most sites that take surveys.
To Meltdowner, I posted twice because wasn't sure which was the right place to do it. To Kingsnake, I believe you are reading my comment wrong. I don't want to be the best album ever, I just want to reflect the opinion of the community, without any type of weighting or biases.
Hope this clears my point,
|
This is primarily a review site (it may not seem like it at first glance, but it is), the ratings are a bit of fun for amusement only and should never be taken too seriously.
All "rating" sites use weightings but because the samples sizes (i.e. number of ratings per album, book or movie) are generally much higher than anything we get here that isn't so visibly apparent on those sites. This is to prevent deliberate skewing of the results by unscrupulous miscreants who abuse the rating-only system - we would much rather this practice never existed so we could use simple averaging but unfortunately rating abuse is very prevalent across the internet because people are basically arseholes.
Weighting is a necessary approach to combat unnecessary fraudulent ratings. Other sites (IMDb for example) are even stricter than us and actually ignore all rating only votes and any rating+review votes from people with a low number of reviews completely when computing their averages.
This site has been here since 2004 and this is the fairest method we have found in that time.
|
What?
|
|
Meltdowner
Special Collaborator
Honorary Collaborator
Joined: June 25 2013
Location: Portugal
Status: Offline
Points: 10232
|
Posted: November 08 2016 at 04:53 |
What is the site without reviews? Why shouldn't the site value the time and effort people spend reviewing?
I don't believe you can't come up with 100 words to describe an album you consider a masterpiece.
Edited by Meltdowner - November 08 2016 at 05:00
|
|
bl6464
Forum Newbie
Joined: September 14 2012
Location: FL
Status: Offline
Points: 11
|
Posted: November 08 2016 at 04:46 |
I didn't know the ratings were weighted. Sorry, but I think this average is biased. Not everybody has the time to write a review or belong to the collab team. In my opinion, the rating should reflect the plain average of the person evaluating as in most sites that take surveys.
To Meltdowner, I posted twice because wasn't sure which was the right place to do it. To Kingsnake, I believe you are reading my comment wrong. I don't want to be the best album ever, I just want to reflect the opinion of the community, without any type of weighting or biases.
Hope this clears my point,
|
|
someone_else
Forum Senior Member
VIP Member
Joined: May 02 2008
Location: Going Bananas
Status: Offline
Points: 24310
|
Posted: November 08 2016 at 04:08 |
Just a rating without a review has a weight of 1. A rating accompanied by a review by a non-collab has a weight of 10. A rating accompanied by a review by a collab has a weight of 20.
I am fine with this, except for one point: I have something against the rating-only feature, but that's my own problem. If you want the average rating cranked up a bit, just put some effort in a review and it will be 4.38 for the moment.
|
|
|
Kingsnake
Forum Senior Member
Joined: November 03 2006
Location: Rockpommelland
Status: Offline
Points: 1578
|
Posted: November 08 2016 at 03:50 |
I think the guy who started this post, wants the album to be the best album ever made. A typical fanboy reaction.
|
|
Meltdowner
Special Collaborator
Honorary Collaborator
Joined: June 25 2013
Location: Portugal
Status: Offline
Points: 10232
|
Posted: November 08 2016 at 03:26 |
It's correct, one of the ratings is a four star review from a Prog Reviewer so it's worth way more than a simple rating. Does it bother you so much to the point of creating two threads?
|
|
bl6464
Forum Newbie
Joined: September 14 2012
Location: FL
Status: Offline
Points: 11
|
Posted: November 08 2016 at 03:12 |
I rated Kyros lastest album Vox Humana and noticed that it has 67% 5's and 33% 4's showing and avg of 4.16 instead of the correct 4.67. Suggest revising this formula. Thanks, Ben
|
|
Donate monthly and keep PA fast-loading and ad-free forever.