Wasn't this because of rating spiking? They'd buy tons of these, then after rating went up return them to spike ratings on the charts as a form of marketing.

Is hacker news now the place SEO spammers go to get support?

> Is hacker news now the place SEO spammers go to get support?

I doubt the OP is a SEO spammer. Buying 4 different CDs in a week is a normal operation for a music aficionado. Back when I regularly bought CDs I would buy maybe 10 or 15 new albums in one go (and then nothing for a while). I think it's a perfectly reasonable and legitimate thing to do.

I understand Amazon's predicament with SEO spammers, but essentially it's a somewhat user-hostile "fix" for their broken rating system. I don't understand why they don't limit the amount of returns, which also makes this type of spam harder. Or maybe rethink the entire "crowd-sourced rating" thing as it keeps being problematic in various ways.

Either way, there are many things you can do better than "fuck you legitimate customer".

> rethink the entire "crowd-sourced rating" thing

The real fix is to not have a single global rating for a product, but instead have 'personal ratings'. Ie. Every product rating I see is either a rating I have given, or Amazon's best prediction of the rating I would give if I had bought that product.

Such predictions are inherently hard to game, since if you start artificially rating your friends products highly, then the system will show high ratings only for your friends products to you.

Then they can do things like "I see you bought this product thinking it would be 4 stars, but you actually gave it 2 stars. We apologize for getting it wrong this time - have a refund of half the purchase price". That gives the AI team plenty of financial motivation to get the ratings correct.

The issue with your proposed model is that it assumes Amazon is a neutral actor who are attempting to accurately predict what you would like.

In reality, Amazon are incentivized to sell you more stuff, so allowing them to calculate ratings through some mysterious black box would introduce perverse incentives (i.e. Amazon is incentivized to sell more higher-margin stuff - would they be more tempted to inflate a product that is actually a '4 star' to a '5 star' to close the deal? Also how can we be sure that their black-box algorithm doesn't generally inflate Fire tablets and deflate iPads? This can happen even if the algorithm is entirely open).

Secondly, would probably reduce trust in the product rating system (as it will be necessarily opaque and hard to understand compared to the existing approach of averaging the reviews).

Amazon won't even fix basic review gaming - not because they can't (otherwise they would have acquired Fakespot) - but presumably because fake reviews help with sales.

You say Amazon is incentivized to sell you more stuff, but this case clearly says they care more about objective ratings than about sales :)

I don't think that logically follows :)

If Amazon had a ratings black box, they would not be incentivised to be fair with it.

But a simple-to-understand ratings system may help sales more than an untrustworthy-black-box calculated rating (particularly considering reviews are about trust).

I feel like the returns are the problem here.

In a healthy free market a competitor to Amazon might fix the problem and grab serious marketshare in the process. I have some hope. Not much though, due to the sheer size of the behemoth.

Criticker does the rating prediction thing really well.

Rate-limit the ratings, not the purchases.

In this case this is about spiking the CD in the top seller charts / search results. Amazon is limiting purchases due to the increased returns from random buyers who buy it and immediately return. They could probably limit the number of returns to 4 instead of purchases, but it might be complicated legally due to their return policy (i.e. they'd need to update the return policy and inform the customer.)

I buy CDs because I want to own my music, instead of renting low-quality versions from a company that can unilaterally block my access on a whim, degrades quality with watermarks, includes DRM so I can't listen where I want, and forces me to use their shitty software that only works on half my devices. CDs are often cheaper than renting the vendor-locked versions, and it only takes a couple minutes to rip a high-quality version.

That said, I'll pay even more if they offer high-quality downloads on Bandcamp.

There are services that let you buy music as uncompressed, DRM-free downloads. Qobuz comes to mind.

There isnt much reason to buy a physical disc unless you want the disc itself.

And why wouldn't you want the disc? Not only is it (as noted) often cheaper, it's also a durable read-only archival-quality backup. Come to think of it, a pressed CD is probably the best ever consumer technology for that so far. What other medium has safely borne nearly a gigabyte across nearly 40 years, and can still be read in the average office?

Right! I try though to buy the CDs directly from the artists - when I catch them live even better - in the hope more of that money goes in their pockets like that.

> and it only takes a couple minutes to rip a high-quality version.

Which drive do you use? My specially-selected-for-the-task drive (selected 10 year ago, to be honest, and I don't know what I will do when it dies, all modern drives looks to be crap) with firmware tuned for audio grabbing takes ~1.1x to grab CD, so 45 minutes CD will be grabbed in 1h20m (because you need check pass, of course, to have high-quality version!).

On a well-made, new-out-of-the-box CD, a modern cheap drive should have no problems ripping at 48x with full CDparanoia checking.

It is much more likely that you will be getting a loudness-war mastering problem than a bad copy of the data. That you can only fix with better source data.

It is completely opposite to my experience, Exact Audio Copy with Secure (not Paranoid) mode, with enabled "defeat audio cache" feature grabs modern new CD on proven SATA LiteOn drive slightly faster than realtime, and slim USB attached drive could spent 3-4x time (0.25x) and fail at the end!

And, yes, old Audio CD (new old stock, 10-15 years old) can be grabbed at 6-8x speed. Looks like new Audio CD pressed much worse than old ones.

I don't believe in 48x exact extraction, sorry.

But you can buy more than one pair of Wolverine meat shredder claws despite the fact that you only have two hands.

https://www.amazon.com/dp/B01N5U14FA

Streaming services can delete music from your library, but if you have the CD in your house, the only thing you need to worry about is fire and theft — and if you have remote backups, you needn't worry about that either.

Plus FLAC is audibly better than streaming formats, and disk space is cheap enough these days to store an entire library in FLAC.

Maybe you want to buy someone a gift, maybe you never buy books but know someone that loves books and for their birthday you want to gift them a book series, same with CDs, even if that person does not have a CD player I think they can appreciate a CD/DVD with all artwork and stuff that comes with that.

I do. I buy CD of the albums I like to collect them.

There are lots of people who still purchase CDs. At least four a week is my normal haul from charity shops!

Looking through the poster's profile https://news.ycombinator.com/submitted?id=erichocean wow, he sure put in the work to let his SEO spam get through! but you caught them anyway, good job!

imagine, all those years, since 2011 https://news.ycombinator.com/user?id=erichocean just to be caught out in the end! that's just heartbreaking.

Maybe that's the reason, but it happens if you try to buy any five CDs -- including used CDs. I just verified it by adding five different CDs by five different artists to my cart and then clicking "proceed to checkout".

> Wasn't this because of rating spiking?

Interesting idea, but who the hell buys a used Audio CD due to its Amazon bestseller rank? I sure don't.

I can maybe see SEO spamming (or whatever) for something like USB chargers or HDMI cables or something. A commodity item that people simply want cheap products with good ratings, and otherwise don't care.

But not music. Literally no one buys old used Audio CDs because of the Amazon best seller rank. It's not even on the list of considerations.

Why doesn't Amazon just remove ratings from users that return the items rated if they are good ratings?

I'd highly suspect support folk are saying it's an error because they have no idea what is happening so that counts as an 'error', and "it will get fixed" is a guess/hopeful assumption to fob the user off.

It's not about Amazon ratings (stars) on the site, it's about album sales numbers for the "top 100" kind of lists. So they can't really take that back after an album has been top seller in a certain week.

People review bomb competitors, so you buy a product from a competitor and rate it a zero, say it burst into fire etc (was popular when battery fires were in the news). The popularity spiking is to get folks higher up on charts etc. Away from the top few it's pretty easy to get up (for a few days).

I though that about support too, that's why I posted here.

I'm certain it is unintended, but who knows if anyone in a position to fix it at Amazon is even aware of the issue. The front page of HN is a great way to give it some visibility…

How could 1 person buying cds do that? You can only leave 1 review per purchase. That only works if multiple people buy the same cd and even then, it would not hit the limit.

You have a "best sellers" category, so I assume they're targetting that(?)

https://www.amazon.co.uk/gp/bestsellers/music

Literally no one on Earth cares about the current value of the best seller rank of a used CD for some random 80s band.

The whole SEO spammer idea for used music CDs doesn't pass the smell test. There's only a few copies available to buy for most used CDs anyway (if you're lucky), so there's no way to pull off the scam.

But why is it 4 cd's in total, rather than 4 of one kind?

Well, honestly, most used Audio CDs don't even have four copies to buy. I'm happy if there's one.

The SEO spammer idea sounds good for other products (e.g. USB chargers, HDMI cables, etc.) but used Audio CDs ain't one of them. You can't even pull of the scam because there isn't enough product to do it.

i have never considered or heard of this. the more you know...

I think it worked because you could buy a lot of CD's each week on behalf of folks who needed the ratings bump, then when the 30 days were up you could return them. So amazon was covering labor and shipping both ways on large quantities of these - stock, then pick and ship, then dealing with the returns and restock etc.

Of course I hope the account would get canceled, but based on other areas these folks seem able to create reasonable numbers of accounts.

The problem is that no one on Earth buys used music CDs due to the Amazon.com best seller ranks. The whole SEO spammer idea is nonsensical in this case, and I highly doubt Amazon is trying to target SEO spammers buying thousands of copies of used Audio CDs. I'm like 99.9% certain it's a mistake.

USB chargers? HDMI cables? Sure, I'd buy your SEO spammer theory that people are messing with the product rankings somehow.

But not used music CDs. You search Amazon for the used CD you already know you want, pray they have even one copy, then buy it. I don't think in 20+ years of buying CDs online, I've ever looked at a best seller rank.

But the real problem with your theory is that there aren't many copies of these things! Certainly not "thousands" as you implied. There's like…two (if you're lucky, a lot of times Amazon doesn't have any 3rd party sellers and you have to check eBay or whatever). I don't see how buying two used CDs, then returning them 30 days later to a 3rd party seller, helps anything with the best seller ranking.

Sorry, if this was just used / out of production CD's then I'm probably wrong.

However, this IS an issue in everything from book lists (which predate Amazon actually and happen outside of amazon with bulk book buys) to I thought regular music "sales" which have become easier to manipulate because total unit sales are low (given streaming etc etc).

> Wasn't this because of rating spiking?

No. It's obviously a bug.

Perhaps related to ratings (There's also chart spiking which is uniquely CDs, which would also have high Amazon returns) . A 40 limit accidently got set to 4 perhaps on Friday. I can't see a customer returning 100% of their CD's lasting long so perhaps it's a limit normally on new customers. Often the same CD has limits around 4.

Or maybe the bug is related to a different limiting mechanism. An engineer set a limit to all CDs not just the one that's in demand perhaps, which they do set for limited stock.

> Is hacker news now the place SEO spammers go to get support?

Why not?

Someone wanting to buy more than 4 CDs is enough for you to call them an SEO spammer?

If I want to buy a 6 disc boxed set, does that qualify me as an SEO spammer too?

wtf is wrong with you?

For the moment at least, there's still bandcamp.

This is likely to avoid buying-for-ratings and then returning; and the solution is to factor it in rating only after the return window expires (or the user explicitly says “I like it, giving up my right to return it”. (True fans will do that happily, SEO staffers……. Not so much)

I tested this just now, sure enough when I get past four CDs I can no longer checkout.

I did another quick test, and it appears it affects Vinyl as well.

Seems to be a region specific issue; I’ve just tried to proceed to checkout with 7 CDs in the cart on Amazon.de and the only warning I got was a changed price notification.

Maybe to stop mass buying gaming sales rankings?

If you were limited from buying more than 4 of the same CD that might make sense, but this seems to be hit after any 4 CDs are bought.

Someone running a r(ating)-aas business might have 1000 accounts and order 1 cd per customer per account. Limiting per account forces them to make more accounts, which means more phone #s, more cc numbers, more cost, and makes them easier to spot.

It's completely bizarre to me that Amazon, with all that data and ML cannot detect this and hammer down.

Then again their best recommendations to me are extremely simplistic "you bought a washing machine, maybe you want a washing machine?" attempts, so I guess not.

Or they just don't care that much because that costs money and this is just easier and cheaper.

I've seen this brought up a lot, but if you split all your customers into people who have ordered washing machines in the last month and people who haven't, the first group is almost certainly more likely to be ordering a washing machine in the near future. Keep in mind that Amazon also knows the exact return rate for every item.

Even if 99% of washing machine buyers won't be in the market for a new one, that's probably still a better ratio than you'd get from advertising to random people who haven't bought one lately.

>but if you split all your customers into people who have ordered washing machines in the last month and people who haven't, the first group is almost certainly more likely to be ordering a washing machine in the near future

If, after 30 years of putting our best minds on the problem, this is the best we can do, then the entire advertising industry deserves to be burnt to the ground for making zero progress while wasting untold riches

I prefer to see it as a reminder of the omnipresent power of Bayesian inference :)

> Someone running a r(ating)-aas business might have 1000 accounts and order 1 cd per customer per account.

These are used Audio CDs. You're lucky if Amazon has one 3rd party seller with a copy to buy. There sure as hell aren't thousands available!

If they have 1000+ accounts ordering 1 CD per customer then this won't affect them until they have more than 4 customers at once.

It's a bug. There is no ulterior motive.

Yup. It's also impossible to "mass buy" used Audio CDs. Amazon itself doesn't sell them, so it's just 3rd party sellers. And for most CDs, there's almost no inventory available. You can't even pull off the supposed SEO scam even if you wanted to.

It's 100% a bug.

To fix rating spiking could they simply remove reviews if the customer returned the item?

Or they could just not count the purchases beyond 4 CDs in a week, when computing the ratings, while still allowing the customer to buy those CDs.

If what's going on is really what people here seem to think, one might conclude that Amazon is run by total idiots. So maybe it's not what people think? Hard to believe that Amazon foregoes real income this way... On the other hand, I suppose it could be misaligned internal incentives - that it somehow makes sense from some manager's point of view, though not from the stockholder's point of view.

Latest update from the Steve Hoffman thread:

> Jeff Bezos team just responded to my email by phone. The problem is a glitch not policy, and is being worked on. Everyone can calm down.

Update: issue appears to be fixed now.

I guess I'll just buy tapes then?

That's weird. I wonder why that is?

Customer service already acknowledged it's an error and should be fixed soon. This thread has huge amounts of speculation on why Amazon did this, when the likely answer is "oops crap I need to rollback that change"

Do you always trust what customer service say? What if they just told him that to make him go away and be next week's problem for some other call-taker? Either because they don't give a crap or because corporate told them to say this...

I'd like to live in your world where we can trust what others, and especially what a big corporation that's been shown to be a lying, thieving, scumbaggy one, say.

Because I’ve worked at a video game company for years, and gotten a good perspective on the value of speculation.

This one time our DC does network maintenance, we alert players of possible larger latencies on the coming weekend, and the playerbase is absolutely certain the latency comes from some mythical AI-based anticheat - and that we’ll announce a massive round of bans soon.

I’m not sure whose razor it should be, but essentially ”what if there is no conspiracy” should IMHO be the default mode. If you spot weird behaviour and the vendor acknowledges it as a bug, why presume it’s actually something nefarious?

I miss AimeStreet. It was a site for independent artists that started tracks at free and then slowly increased them to $0.99 as they sold more copies. Amazon gobbled them up and shuttered the site. I’ve never found a replacement.

Even better, if you reviewed a song, and the price went up, you received the difference as store credits. So reviewing a song early could potentially net you some credit if it became popular. That was one of the most interesting features I've seen in a music store.

How did it compare to something like Bandcamp? Many of the artists I enjoy sell their music there.

I think the complete absence of "me too" or "not me" posts shows the state of the market for CDs.

Well, four in a single week is also a lot. Not absurd by any stretch, but high enough that I'd expect few people to hit it.

How many people buy more than four hard cover books in a week? What about AAA video games?

If you were asking how many people buy four CDs every week, you would have a point, but people often bunch up their purchases for many different reasons. Buying five CDs once a year shouldn't be impossible.

I've bought 11 CDs in the past week, with another 3 in the mail. Might be something of an outlier though.

CDs are a lot less expensive than hardcovers or newly-released video games, especially if you buy them used.

They really don't want you to be able to actually own music anymore.