POO: AMG vs Pitchfork vs RYM

Message Bookmarked
Bookmark Removed
Not all messages are displayed: show all messages (71 of them)

completely wrong poll results now haha

imago, Monday, 16 April 2018 17:16 (six years ago) link

all due respect to ned etc etc

imago, Monday, 16 April 2018 17:17 (six years ago) link

xxxp Maybe it's that music, like everything else, is increasingly polarised, and there's a tendency for some users to give everything that mildly displeases them a score of .5 or something like that. Another thing is that there's a lot of unquestioning adherance to THE CANON and the belief that anything that's already established and well-known is objectively better but that's hardly exclusive to RYM.

As someone who's pretty active on that site I rarely take general ratings into account tbh, I'm more interested in what certain users think even if I don't always agree with them.

obnoxious pun (ultros ultros-ghali), Monday, 16 April 2018 17:20 (six years ago) link

lower ratings = combination of

- widespread reluctance to give out the highest ratings to new albums until they've been 'proven by time' or w/e

- lamentable increase in use of the dreaded 'positive' rating system where morons curate like a tiny handful of high ratings and loads of low ones to show how sophisticated they are

- shitlords (see also previous)

nonetheless it is a far, far more useful site for the discerning music nerd than a million pitchforks or AMGs. also Fantano cops a lot more hatred there than he does here. the xxxtentacion album is on like a 1.5 out of 5 average rating atm fyi

imago, Monday, 16 April 2018 17:35 (six years ago) link

also ALSO some young artists get Roy high average ratings, for instance frank ocean, the last tyler album etc

imago, Monday, 16 April 2018 17:37 (six years ago) link

sometimes a new artist who wasn't in odd future does well too

imago, Monday, 16 April 2018 17:37 (six years ago) link

'Mt' might be their worst song, otoh, it is an odd album

imago, Monday, 16 April 2018 17:40 (six years ago) link

Wrong thread FFS but posting about SFA accidentally in here feels apt

imago, Monday, 16 April 2018 17:40 (six years ago) link

I also think there are a lot of users who monitor the charts and downvote anything which climbs the list too fast. A lot of cool stuff gets rated in the 3.9-4.1 range within the first week and then freefalls to like, 3.2 over the next month as the album starts to get 'hyped'. imago knows what I'm talking about :)

frogbs, Monday, 16 April 2018 17:50 (six years ago) link

yeah, hence 'shitlords'

imago, Monday, 16 April 2018 17:51 (six years ago) link

fantano can be a bit of a kingmaker as well, regrettably

although him giving 8/10 to the richard dawson album was a big boost to its profile, so there is some occasional smooth to go with the rough. his taste is highly suspect though obv and his personality more so . actually the thing that gets me the most about him is how bland he is. but the kidz who tag around after him are all about the memes and the branding, fuck the actual content

imago, Monday, 16 April 2018 17:56 (six years ago) link

glad we could have this grown-up chat about RYM without whiney getting too het up

imago, Monday, 16 April 2018 17:58 (six years ago) link

i never use RYM, but i'll go ahead and suggest a silly "solution": they should provide statistical scores(maybe Z-scores?) that take into account how the individual users tend to rate albums.

for example, let's say i give "I Get Wet" a 4.0. As others have mentioned, that doesn't really tell anyone much, because maybe i'm a person that generally likes everything ("person A") and gives everything a 4.0, with exceptional albums receiving a 5.0 and the worst album of all time a 2.5. But maybe i'm a more critical person ("person B") who very rarely gives a 4.0 to anything, and usually my scores hover around the 2.0 range. it would be much more useful to know how that 4.0 stands relative to my other ratings. that's easy to do with a z-score (a measure of standard deviations from the mean). let's say both examples have rated 10 albums on RYM:

Person A: 4.0, 4.0, 4.0, 4.5, 4.0, 5.0, 4.0, 3.0, 2.5, 4.0
Average: 3.9
Z-score for a 4.0 album review = 0.14 (positive but very low, meaning that a 4.0 is just barely higher than what the user typically gives an average album)

Person B: 2.0, 1.5, 2.0, 2.5, 3.0, 2.0, 2.0, 4.0, 2.5, 2.0
Average: 2.35
Z-score for a 4.0 album review = 2.33 (positive and very high, meaning that a 4.0 is more than 2 standard deviations above the average for that user; this person is stingy with 4.0s but loves "I Get Wet"

this isn't going to catch on is it

Karl Malone, Monday, 16 April 2018 18:07 (six years ago) link

they do have an algorithm which weights individual user's ratings and fudges the rating to be more than just a straight up average. but they won't reveal what it is because they don't want people to game the system. I don't think it's quite that but that's a good idea.

frogbs, Monday, 16 April 2018 18:09 (six years ago) link

What if approximately 90% of the ratings you give are 0.5s? As is this invaluable contributor's wont:

https://rateyourmusic.com/~Freeloader

pomenitul, Monday, 16 April 2018 18:10 (six years ago) link

Every member like that either has some ambient shit or something by the Arditti String Quartet as their highest score

imago, Monday, 16 April 2018 18:12 (six years ago) link

I remember some explanation of the weighting--like your ratings have more weight if you've written x number of reviews of the site, and most 5 star and 0.5 star ratings are ignored etc. But this was from awhile back.

President Keyes, Monday, 16 April 2018 18:13 (six years ago) link

reviews on the site

President Keyes, Monday, 16 April 2018 18:14 (six years ago) link

xpost
lol! well, the highest rating ~Freeloader has ever given is a 3.0, for The Caretaker - Everywhere at the End of Time - Stage 1 (2016).

379 0.5 reviews
30 1.0
12 1.5
8 2.0
2 2.5
1 3.0 review

The Z-score for 3.0, given that sample, is an off the charts 7.25. so basically ~Freeloader IS the caretaker, or else the caretaker's #1 fan of all time

Karl Malone, Monday, 16 April 2018 18:16 (six years ago) link

Speaking of the Ardittis, RYM is godawful for contemporary classical. You have to submit referenced data re: uncatalogued works to a specific thread where they get approved 24h+ later by select moderators. And adding the albums said works appear on is an ordeal in and of itself as you have to link each track to its respective movement or whatever. I only use it for non-classical music as a result.

pomenitul, Monday, 16 April 2018 18:18 (six years ago) link

all of this z-score chat is reminding me that I have to help tt with her statistics exam revision

when it's over I will never want to see another chi-square, another standard error

imago, Monday, 16 April 2018 18:20 (six years ago) link

AMG for me

omar little, Monday, 16 April 2018 18:24 (six years ago) link

I also think there are a lot of users who monitor the charts and downvote anything which climbs the list too fast. A lot of cool stuff gets rated in the 3.9-4.1 range within the first week and then freefalls to like, 3.2 over the next month as the album starts to get 'hyped'.

I've been online for long enough that this sort of thing shouldn't bum me out but it still does.

Gerald McBoing-Boing, Monday, 16 April 2018 20:46 (six years ago) link

AMG - too much javascript
Pitchfork - reviews are way too long
RYM - some fun lists but way too much garbage to wade through

brimstead, Monday, 16 April 2018 22:37 (six years ago) link

My least favorite thing about RYM is that they added those weird little comments sections for each release which sort of pulls back the curtain and lets you see all the shitlords (stealing imago's word) in action.

cwkiii, Tuesday, 17 April 2018 00:47 (six years ago) link

Speaking of the Ardittis, RYM is godawful for contemporary classical. You have to submit referenced data re: uncatalogued works to a specific thread where they get approved 24h+ later by select moderators. And adding the albums said works appear on is an ordeal in and of itself as you have to link each track to its respective movement or whatever. I only use it for non-classical music as a result.

― pomenitul

i figure it's just a culture thing. rym is also shit for electronic music, which is much better represented on discogs.

re: shit ratings for more recent records - i do find that more recent records have way more ratings than the older "canon" stuff. and there are a _lot_ of people who are just following /mu/p4k/fantano in their ratings, and a _lot_ of people who treat music like some sort of game theory exercise (which i used to do back when i first got on the site so i can't complain too much). mind you a couple years back i more or less gave up on ratings entirely (i still throw one in occasionally just for shits) and do all my aggregation via lists.

the thing i like most about rym is that for whatever reason it's packed to the gills with trans people. that's cool.

ziggy the ginhead (rushomancy), Tuesday, 17 April 2018 00:59 (six years ago) link

I mainly use it for randomness—like, Hey what’s the 791st highest rated album of 1978? Why don’t I listen to that? (It’s Robert Palmer- Double Fun btw)

President Keyes, Tuesday, 17 April 2018 01:08 (six years ago) link

I don’t really mind the comment box...it’s easy to ignore and it cuts down on the number of jokey one-liner reviews. Some of which are pretty funny mind you.

frogbs, Tuesday, 17 April 2018 01:14 (six years ago) link

i figure it's just a culture thing

For sure, I just wish there was a database to rule them all. AMG isn't half bad for classical music but every now and then they'll be missing something like Grumiaux/Gendron/Bourdin/etc.'s set of Debussy's late sonatas and it makes my inner documentalist despair.

pomenitul, Tuesday, 17 April 2018 01:46 (six years ago) link

I haven't been on AMG in a while but one thing I clearly remember is how odd it is to page through an artist's catalogue (as the site is set up to do) and read a bunch of reviews that were written by very different people, at least one of whom was clearly not a fan and was just assigned to clean up loose ends or whatever. Not to mention the fact that their star ratings are seemingly pulled out of a hat. Don't know if it's still like that though

frogbs, Tuesday, 17 April 2018 12:33 (six years ago) link

never forgive, never forget

https://www.allmusic.com/artist/cardiacs-mn0000531346

imago, Tuesday, 17 April 2018 14:12 (six years ago) link

idk those Dean Carlson reviews read like a pretty strong endorsement to me

frogbs, Tuesday, 17 April 2018 14:23 (six years ago) link

lol

imago, Tuesday, 17 April 2018 14:24 (six years ago) link

I don’t really mind the comment box...it’s easy to ignore and it cuts down on the number of jokey one-liner reviews. Some of which are pretty funny mind you.

I wish it was easy for me to ignore! But I can't help but get myself worked up over reading these people petitioning to put certain releases in bold even though the ratings aren't particularly high, or insisting that a 90 second demo of fucking around with a synth posted to a Soundcloud counts as an Aphex Twin release. One for the IA thread I guess.

The real reason I hate those fuckers is for Lulu's rating, who am I kidding

cwkiii, Tuesday, 17 April 2018 19:38 (six years ago) link

justice 4 lulu

flamenco drop (BradNelson), Tuesday, 17 April 2018 19:51 (six years ago) link

:)

cwkiii, Tuesday, 17 April 2018 19:54 (six years ago) link

i have a hard time ranking these bc a significant difference between rym and the other two is that the writing is almost completely worthless as opposed to marginally/mostly worthless

flamenco drop (BradNelson), Tuesday, 17 April 2018 20:02 (six years ago) link

rym and amg are v useful when approaching completely discographies and weirdly ime tend to pick up each other's slack? what's received wisdom at one isn't always received wisdom at the other. pfork is useless for this but it's not built that way

flamenco drop (BradNelson), Tuesday, 17 April 2018 20:03 (six years ago) link

but then i use rym exclusively for personal use and cannot even imagine using it socially

flamenco drop (BradNelson), Tuesday, 17 April 2018 20:05 (six years ago) link

i have a hard time ranking these bc a significant difference between rym and the other two is that the writing is almost completely worthless as opposed to marginally/mostly worthless

― flamenco drop (BradNelson)

the great thing about writing for free is that i don't need to give a shit about whether or not my writing is "worthless"

ziggy the ginhead (rushomancy), Tuesday, 17 April 2018 20:35 (six years ago) link


You must be logged in to post. Please either login here, or if you are not registered, you may register here.