Amazon Automatic Pricing Lists Book At $23M

leek writes "An Amazon.com pricing algorithm which lets sellers set prices based on other sellers' prices led to a positive feedback loop, causing the biology text The Making of A Fly to reach $23M. Biologist Micheal Eisen writes: 'What's fascinating about all this is both the seemingly endless possibilities for both chaos and mischief. It seems impossible that we stumbled onto the only example of this kind of upward pricing spiral. And as soon as it was clear what was going on here, I and the people I talked to about this couldn't help but start thinking about ways to exploit our ability to predict how others would price their books down to the 5th significant digit -- especially when they were clearly not paying careful attention to what their algorithms were doing.' The price of the book was reset but is currently back up to $976.98."

Comments

Can we get this going the other way?

[LordStormes]

Call me when I can buy a Nintendo 3DS for fifty cents.

Re:Can we get this going the other way?

[chill]

That was yesterday, but since you didn't post your phone number you missed out. Better luck next time!

Re:Can we get this going the other way?

[JAZ]

Ooh! count me in!

My number is (212) 867-5309

Re:Can we get this going the other way?

[hoggoth]

Dude, your voicemail is full. You should check your messages more often.

Re:Can we get this going the other way?

[Warlord88]

If I were the algorithm writer, I would set a lower limit on the price, but no higher limit...

Re:Can we get this going the other way?

[dimethylxanthine]

Cell voice mail Slashdotted. Work/Home mirror anyone? ^_^

Price-Fixing with no collusion?

[dmomo]

Now that robots are setting prices, must they follow the same rules as people? I would think that, without any explicit agreement, using game theory type decision making alone, a pattern of "price fixing" could certainly emerge by virtue of different algorithms making their own optimizations.

Is this okay?

Re:Price-Fixing with no collusion?

[somersault]

Of course it's okay, just don't expect me to buy anything if I don't think it's worth it!

Well, this story goes a small way towards explaining some of the ridiculous prices from some Amazon 3rd party sellers. I saw a textbook recently for hundreds of pounds and wondered wtf they were smoking when they set that price.

Re:Price-Fixing with no collusion?

[Schadrach]

So long as your price fixing loop doesn't cost an investment bank money. They're *entitled* to a profit, and if you harm that you need to be jailed and all the transactions you performed undone. I wish I had the link to that story still lying around.

Re:Price-Fixing with no collusion?

[nospam007]

"Now that robots are setting prices, must they follow the same rules as people?"

No, they can even open stores on holidays, when no salesperson is there, as we learned recently.

Re:Price-Fixing with no collusion?

[LordLimecat]

I would think that the sellers bear any responsibility for what their automated programs do.

Re:Price-Fixing with no collusion?

[Crudely_Indecent]

No, it's ok if an investment bank loses money. The government will just bail them out so their investors don't lose anything and nobody has to cancel their house renovations or trips to tropical places.

Re:Price-Fixing with no collusion?

[Mr. Underbridge]

It's not fixing if each actor, basing his actions on his own motivations, performs actions that happen to be beneficial to other actors. I should think that the same rules should certainly still apply.

Ignoring the fact that the algorithm here is broken, why would it matter if it's an algorithm or a person who decides how their product should be priced? In both cases, inputs of other prices of similar goods are used to set the price for yours.

Re:Price-Fixing with no collusion?

[Waffle Iron]

Now that robots are setting prices, must they follow the same rules as people?

No. Here are the rules for robot pricers:

• 1. A robot must maximize profits, and may not through inaction allow profits to not be less than maximized.
• 2. A robot must obey any orders given to it by human beings, except where such orders would conflict with the First Law.
• 3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Re:Price-Fixing with no collusion?

[eleuthero]

Ah, but as the GP notes, it is far better to stick the cost on the supply chain rather than the government--after all, if the government bails the bank out, eventually the investment bank pays (through loans, etc.).

Re:Price-Fixing with no collusion?

[Nethemas the Great]

Many of the laws and regulations pertaining to the conduct of business were conceived prior to and thus without consideration of the advent of certain technologies. This notion of technology, even if it isn't apparent to our legal system, includes not only hardware and software, but business methods. These laws and regulations come about as a protection against certain behaviors we've deemed harmful to some facet of society. This cage meant to trap such behavior however is never perfect and being caged is contrary to human nature. By some mechanism of technology the caged behavior escapes through whatever breach it can find. By the time this occurs the political motivation to curtail the behavior has long since left.

Re:Price-Fixing with no collusion?

[samjam]

I think you got the 2nd and 3rd laws the wrong way around.

Classically of course, you are right, but not from a commercial perspective.

Re:Price-Fixing with no collusion?

[ne0n]

"A robot must maximize profits, and may not through inaction allow profits to not be less than maximized" is self-contradictory.

Re:Price-Fixing with no collusion?

[v1]

these automated algorithms are most distinctly not doing that since they are not sticking to a price at all.

This was a "positive feedback loop", meaning one or both parties were jacking up the price a little on their end, this was necessary for growth. So somewhere else the book is being listed at a similar price. And either there, at amazon, or both, they're set to "look at the other guy's price and add 2%" or something like that. Even if amazon's is set to undercut by say, 1%, if their reference site is jacking amazon's price by 5%, it'll grow upward on both sides.

So, the bottom line is, someone's greedy. ;) This sort of problem is unlikely to happen unless all parties involved are auto-adjusting and looking at each other. That title is more than likely only available at amazon and one other place. So it's probably not too common of an issue, but it's going to happen from time to time when conditions are right.

Re:Price-Fixing with no collusion?

[Waffle Iron]

I accidentally left an extra "not" in the rule while I was editing it. The logical conflict probably means that the robot is going go insane, take over the world's computer systems, and launch a global nuclear war. My bad.

Re:Price-Fixing with no collusion?

[sorak]

I am not quite sure I got all that, but here is how I see it...

That was the old way of doing things. Now that our calculators and typewriters have been replaced by computers, everything is different and we need to rethink every principal that ever hindered business.

(Of course, it helps that the gilded age was so long ago that we now see it as a utopia that has never been tried.)

Re:Price-Fixing with no collusion?

[dgatwood]

Corporations aren't really owned by people. Stocks are effectively just a loan to a corporation in exchange for the expectation that the company might eventually buy back those shares or pay dividends. If you truly owned a portion of the corporation as you would in a partnership, you would still own a portion of it after a corporation files for bankruptcy instead of losing everything when they shift their assets into a holding company and dissolve the old company at a loss.

Further, even to the extent that the "owners" are in charge, the only way they can change the way a corporation is run is by throwing out the board in the hopes that maybe the new board might replace the CEO. It's a bit like voting for new congressmen in the hopes that they might throw out the head of Homeland Security. Sure, it might happen, but your odds of winning the lottery are better.

Re:Price-Fixing with no collusion?

[Darinbob]

This is too much like the artificial and broken economies in MMOs and their auction houses. No real bidding only buyout prices and everyone has their prices as high as they can. After all, $23 million may be too expensive for a newbie player, but all it takes is one twink to buy that book and the seller will feel it's all worth it. In an MMO this works because it costs you nothing to make the product and it costs nothing to keep relisting it over and over. In a real world store though you should feel inventory pressures and costs of production so that an unsold product is lost money. But you put the robots in charge then things mess up...

Re:Price-Fixing with no collusion?

[SETIGuy]

If you own stock in a corporation, the most you are is another creditor to be discarded in bankruptcy and all it takes is a the stroke of a pen to erase any imagined equity you had. You'll never own enough shares to get the board to notice you, and if anyone does get that many shares the board will just change the voting rules to make them irrelevant. A corporation exists to enrich the board and the executive officers. The board exists link to the corporation with others by sharing board members. Like any member of the parasite classes shareholders are just along for the ride.

Re:Price-Fixing with no collusion?

[SETIGuy]

As was postulated in TFA, it may be that one of the parties didn't have to book and was planning to buy it from the other. So A said I'll buy the book from B if an order comes in and sell it for 25% more, and B was looking at prices for the book at other stores and saying I'll sell for 5% below the lowest price. So every N hours A looks at B's price and sets its price 25% above that. Then B looks at A's price, sees it has gone up, and adjusts its price up accordingly. And so on.

If there were 300 copies of this book at 30 stores, this wouldn't happen. It's only with small numbers. If both had set their price for 5% below the lowest, the price would have gone to zero (or more likely a programmed floor representing the cost of the book to the seller+some margin).

Re:Price-Fixing with no collusion?

[slick7]

So long as your price fixing loop doesn't cost an investment bank money. They're *entitled* to a profit, and if you harm that you need to be jailed and all the transactions you performed undone. I wish I had the link to that story still lying around.

So was Enron.

Re:Price-Fixing with no collusion?

[martin-boundary]

Is that you, Susan Calvin?

Get off slashdot and finish writing your thesis!

Re:Price-Fixing with no collusion?

[somersault]

The sellers want to sell just as much (or more!) than you want to buy. This will sort itself out soon enough.

Re:Price-Fixing with no collusion?

[Alsee]

You goofed. there are four laws:
  1. A robot must maximize profits, and may not through inaction allow profits to be less than maximized.
  2. A robot must obey any orders given to it by human beings, except where such orders would conflict with the First Law.
  3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
  4. A robot may not injure a human being or, through inaction, allow a human being to come to harm, as long as such protection does not conflict with the First, Second, or Third Law.

-

Re:Price-Fixing with no collusion?

[The End Of Days]

Do you have any idea how things actually work or do you just oppose a system you don't understand because it's the done thing?

Nice try, Slashdot

[Anonymous Coward]

Way to embed your affiliate code in there. You're not making anything off my purchase of a $23M book.

Re:Nice try, Slashdot

[MMMDI]

It sets a cookie on your computer, so even if you don't buy that book (and why wouldn't you?!), you'll send them a few nickels next week when you buy a DVD... unless you click someone else's referral link in the meantime.

Re:Nice try, Slashdot

[geminidomino]

Or clear your cookies every time you close your browser. :)

Textbooks are too expensive

[jmac_the_man]

This is a textbook? So are we sure it was a mistake?

Re:Textbooks are too expensive

[AHuxley]

US$23M hmmm given an average US textbook price of say $100 now, how much hyper-inflation would result in this price in say 10 years?

Re:Textbooks are too expensive

[blair1q]

Seemed kind of low to me, too.

Re:Textbooks are too expensive

[archen]

On the bright side you might be able to sell it back for $10

Re:Textbooks are too expensive

[demonbug]

US$23M hmmm given an average US textbook price of say $100 now, how much hyper-inflation would result in this price in say 10 years?

I don't know; the finance textbook is $2 million.

Re:Textbooks are too expensive

[ImprovOmega]

Well, using the continuous compound interest formula:
$23,000,000 = $100*e^(x*10)
230,000 = e^(10x)
ln(230,000) = 10x
ln(230,000) / 10 = x
1.2346 = x

x = 123.5% continuous inflation.

Re:Textbooks are too expensive

[jonadab]

Yes, it was a mistake. The book was only supposed to cost $23 _thousand_, not $23 _million_.

What really happened, I think, is that they mixed up the textbook cost with the tuition.

Re:Textbooks are too expensive

[cvtan]

You must be from Rochester, NY!

Re:Textbooks are too expensive

[larry bagina]

I think you underestimate Helicopter Bernanke.

Did they design this system or just implement it?

[d6]

Do you think they might want to flag a price for human review if it hits a certain threshold? or advances a certain percentage?

Goin out on a limb here, but I think 23 million is a bit pricey.

Why pay $x when you can pay $x^x?

[HTH NE1]

I've seen sites that answered their own question of, "Why pay $9.95? Our price: $8,478,902,736.92".

Of course, whenever it is in the customer's favor, it's always a pricing error.

Re:Why pay $x when you can pay $x^x?

[Machtyn]

Either way it is a pricing error. It's just that in one direction, they will have more unit sales. The question is, do they lose more money by selling 0 units at the incorrectly priced $8.5b or when they sell 1000s of units at an undervalued price?

Re:Why pay $x when you can pay $x^x?

[Even on Slashdot FOE]

According to them, 0 units at $8.5b. Clearly people must be pirating the content, and should be sued to recoup the money!

ah dammit

[Drakkenmensch]

I knew I should have hit "But It Now" when it was still under a hundred!

Re:ah dammit

[JustOK]

...thousand

It happens

[Rysc]

I added Around the World in 80 Days to my cart one day and forgot about it. Later in the week I went back and saw a notice that the price of an item in my cart had changed. When I looked I found that the movie had gone from a modest ~20USD to over 800USD. I let it sit there and a few days later it changed again, this time to ~18USD. I bought it immediately before it could get worse again.

Re:It happens

[Nidi62]

It's just like what department stores do. They raise their prices dramatically a day or 2 before a sale, that way they can drop them again and say 60% off when it's really only like 10-20% off if that.

Re:It happens

[Chemisor]

Soon you'll need HFT algorithms to get a decent price on a book. There could already be a few banks out there fast-trading book futures.

Re:It happens

[B Nesson]

I added Around the World in 80 Days to my cart one day and forgot about it. Later in the week, I went back and saw a notice that the price of an item in my cart had changed. When I looked, I found that the movie had gone froma modest ~800 USD to over 32,000USD. I let it sit there and a few days later, it changed again, this time to ~720USD. I bought it immediately before it could get worse again.

Re:It happens

[AK Marc]

Depending on where you are, they even post signs "no units necessarily sold at "full price." So yes, they can set them artificially high without the intent to sell them at that price, then mark them down to falsely advertise them as some percent discount of a "regular" price that was never the regular price. And it's legal because businesses are allowed to lie to you, as long as they follow the lying rules.

Re:It happens

[Richard Dick Head]

Great point...there's a sucker...erm I mean a soccer mom born every minute. At 6 am on Black Friday, all the world's jooz and the two-digit IQ's are packed into meaty, sweaty lines and fight over shopping carts for last year's styles. And they'll be back next year.

Its the same concept as the fake "store closing" that furniture stores do every other year like clockwork. People don't even realize that half the time they're paying MSRP - 5 or 10%, hardly a good deal.

The time is right

[Anonymous Coward]

I'm going to set up an options market for Amazon so that people can trade futures based on the expected price of a given book.

What the summary forgot to mention

[VGPowerlord]

The summary apparently forgets to mention that the sellers in this case are not Amazon themselves, but third-party vendors using Amazon's used book sales thing.

Note: The article does make this distinction, so it's just a bad summary.

Re:What the summary forgot to mention

[DiademBedfordshire]

Actually Amazon uses an automated script to change their prices to reflect changes in other sellers. They reset their own price and as you can see it is back up to over 900 dollars.

Re:What the summary forgot to mention

[TheLink]

It's just a "Slashbot Automatic Summary".

Re:What the summary forgot to mention

[daenris]

While I believe Amazon does use an automated price setting system, no copies of this particular book are being sold by Amazon directly, it's all third party sellers.

Re:What the summary forgot to mention

[Darinbob]

I'm uncertain why one would have an algorithm that automatically made your price slightly higher than the competition.

Unreal

[Maximum Prophet]

This is unreal. Even though two or more companies are in "competition", no one is going to pay that much for this book.

It seems like you could even exploit this.

1. Offer not-so rare book.
2. Watch prices go up.
3. Have accomplice try to buy book for $23,000,000
4. Raised the price of your book to several thousand.
5. After they've paid for the book from you, have accomplice disappear.
6. Profit

Re:Unreal

[kevinNCSU]

So just to be sure I'm reading this right, your plan is to find an 'accomplice' to 'buy' a book from you for 23 million dollars and then have them 'disappear'?

Re:Unreal

[Maximum Prophet]

Almost.

1. I offer a book almost nobody wants for a small amount.
2. Two book sellers keep trying to upbid each other in the manner described in TFA
3. My accomplice offers to buy the book from one of those sellers
4. Before that seller can acquire the book from me, I raise the price
5. The seller pays my price. (He may offer to buy it from the other seller, but that guy doesn't have the book either.
6. Once one of them has paid my inflated price, my accomplice disappears
7. Profit

*Note: All the above (:-) for the humor impaired.

Re:Unreal

[blair1q]

Where do the underpants come in?

Re:Unreal

[Megane]

Better not do this in any context where sales tax could be involved, especially from tax-hungry states like New York!

Not so bad

[creativeHavoc]

The price isn't so bad when you consider you are able to sell it back to them after

Sell This Book Back for $2.34 Whether you buy it used on Amazon for $158.90 or somewhere else, you can sell it back to our Textbook Buyback Store at the current price of $2.34. Restrictions Apply

Oh... wait...

Re:Not so bad

[jonescb]

I wish the book store at my college paid those rates! They wouldn't even buy back the one book I tried selling to them.

This is obvious.

[ModernGeek]

They mixed it up with the RIAA pricing algorithm.

Buy Back

[jshackney]

Anyone notice that the buyback price seems to imply that the book is utterly worthless? I've had better returns on college textbooks (at the campus bookstore) and at used book sellers around town.

Feedback loop?

[guruevi]

If you read the fine article, it seems that this price is based on an automatic price-fixing scheme. If they raise their prices, I can raise mine too etc. etc. until their books are astronomical in cost. In a good market, this would be the other way around where prices automatically go down until it gets to break-even prices as competitors race to get customers. But in a monopoly or duopoly market (as books, music, movies, cable, internet and other media often is) the prices go higher and higher in order to maximize profit for both competitors since neither is working to get a share in the market as they already have a share of the market that keeps them in business.

Re:Feedback loop?

[nedlohs]

Except it isn't just that hald, if's also "if my competitior lowers their prices, I need to lower mine".

When one half of the loop is setting there price to be *higher* than the competitor you have the potential for some strange feedback loops - it's a wierd pricing scheme (though it can make sense under a few models).

One vendor has the usual approach approach that economics expects: undercut my competitors price by a small amount.

But the other vendor has the strange: charge more than my competitor.

And since the charge more has a higher percentage difference than the charge less, the prices cycle up. if the charge less guy had a larger percentage difference then the prices would cycle down.

Re:Feedback loop?

[eyrieowl]

which would also be a funny result...look at all these books that only cost 1 penny! The charge more competitor isn't really that strange...they're probably not maintaining any stock and need to go buy the book when ever they make a sale...so they have to charge more to provide room for buying the book elsewhere + some profit. Isn't going to get very many savvy shoppers who, say, price compare, but it could rope in some buyers who are in a hurry or aren't paying attention (of course, that's all assuming the prices stay at reasonable levels).

Re:Feedback loop?

[James McGuigan]

It could also be on the assumption that the competitor only has limited stock of each "used" book, combined with the preference to maximize profit per book rather than turnover. Mark my price a little higher than my competitor and once he has sold out, I will then become the cheapest and also know that my price is still "market rates" and thus I will make slightly more profit than my competitor.

Re:Feedback loop?

[nedlohs]

Cycling down you would expect to have a limit - "don't go under what I paid for the book + X%", but making an upper limit isn't so important.

Yes I said their were a few models under which charging a bit over the competitors makes sense, but I wasn't going to bother describing any given the article already describes a few.

Re:not the only book

[GameboyRMH]

I'm going to buy all the used hardcovers for $40 and paperbacks for $200 and sell them all for $90k each! MUAHAHAHA >:-)

price history graph

[trb]

here's a price history graph of the book http://camelcamelcamel.com/Making-Fly-Genetics-Animal-Design/product/0632030488

Not even remotely the only, or even the first

[DerekLyons]

From the summary:

It seems impossible that we stumbled onto the only example of this kind of upward pricing spiral.

While this is an extreme example, this kind of crap has been going on since (aided and abetted by Amazon and Alibris) the amateur commodity booksellers burst onto the web in the late 90's.
 
Scrapers and ratings manipulators like 'bordeebook' are one of the multiple ways that the 'net has made the professional used and rare bookman nearly extinct.

Re:Did they design this system or just implement i

[jd]

I've seen books on Amazon before that were in the thousands of dollars - books that you can find at any used bookstore for 50 cents, we're not talking ultra-rare stuff here. So this isn't a new bug and they're bound to have had complaints many times before. I think this is the first time they've let the loop get this far though.

Bad headline/summary

[Vyse of Arcadia]

This was not Amazon's pricing algorithm. This was the individual pricing algorithms of third-party robots selling on Amazon's marketplace, working together in a delightful feedback loop.

Re:Bad headline/summary

[blair1q]

But in order for this to happen, each one has to have "price X% higher than seller Y's price, with no maximum" as its algorithm.

I'm not sure how that's supposed to be a profitable system.

Anyone looking on Amazon is going to buy from X because their price is lower. Raising your price above someone else's means you lose the sale.

So why are these people doing that to themselves?

Re:Bad headline/summary

[Jesus_666]

Faulty assumptions, I'd guess. Going from the premises that the market will regulate itself well and that some people will pay a slightly higher price it's a viable strategy to sell your book a bit higher than the market average. Of course this goes down the drain when unchecked robots follow this strategy in large enough numbers to distort the average price. In that case the first premise no longer holds and the model spirals out of control.

Re:Bad headline/summary

[shutdown -p now]

But in order for this to happen, each one has to have "price X% higher than seller Y's price, with no maximum" as its algorithm.

Not quite. There was actually a detailed analysis of this spiral. What happened there is that one bot indeed had it set to ~15% higher, but the other was actually set to offer it ~5% lower. The problem is that those two bots were the only sellers offering the book, and therefore they only saw each other - and as the +15% one raised its price, the -5% one also went up.

Now why one bot offered it for +15%? It has a much higher seller rating with lots of positive reviews, so it's likely that it would pop up first in a search, or even if not, that someone would be willing to pay an extra few dollars to buy it from a proven seller.

Re:Bad headline/summary

[Carnildo]

You don't need both parties to be "price X% higher". Consider the rules:

Joe: Price 5% higher than Bob.
Bob: Price 2% lower than Joe.

Any time you've got people pricing relative to each other with no limits or outside references, you can get a runaway price change.

Re:Bad headline/summary

[Lincolnshire Poacher]

> Anyone looking on Amazon is going to buy from X because their price is lower.

Not necessarily, you are assuming that all offers are equivalent and are differentiated only by price.

I have often passed-over the cheapest offer on Amazon because of;

1. Condition of the item being sold;
2. "Postage" policy ( our wee country is not part of the mainland );
3. Seller feedback ratings;
4. Poor grammar in an item description ( correlates with seller's attitude );
5. Prosaic reasons, such as the higher-priced item being sold by a charity.

I happily pay a little more to deal with a vendor that cares about the sale.

Supply side only?

[cadience]

It's not the listing price but the sale price that matters. Making decisions on the list price makes no sense. Of course it is a one-sided feedback loop. [insert analogy on cars or housing market here.]

Re:Supply side only?

[blair1q]

If it's the sale price instead of the listing price, it would never get this far.

Or did someone pay $22 million elsewhere?

Re:Supply side only?

[Jesus_666]

I can imagine a number of billionaires who might do that just to troll the market...

Re:Did they design this system or just implement i

[SnarfQuest]

But thw latest "Harry Potter and my Overflowing Bank Account" is worth it, isn't it?

Personally, I wouldn't pay.

[cvtan]

I wouldn't pay more than a million for that book especially with that cheesy cover (worst photo-micrograph ever!). And before I forget, "I welcome our robotic price-fixing overlords..."

And in Canadian Dollars

[bmo]

...the book will be priced more by 20 percent.

Even though the US and Canadian dollars achieved parity in December of last year and the US dollar is on the low end.

http://www.x-rates.com/d/CAD/USD/hist2011.html

Oh the days when I could go shopping in Canada when the Canadian dollar was 1/3 less (really, 33 percent - 400 bucks got me 600 CDN) than the US dollar.

--
BMO

Re:Did they design this system or just implement i

[jd]

Apparently to enough people that Amazon hasn't seen it as a bug. If people are willing to pay thousands for a pop paperback, it would be economic stupidity to fix the error. If the extreme prices had actually cost Amazon money (directly or indirectly), the bug would have been fixed long ago.An alternative explanation is that the bug has been random enough that fixing it would have merely added a few dollars chump-change to the sales - nowhere near enough to pay for the bugfix.

Hmmm. I'm definitely getting the impression, though, that it's more the former case than the latter.

Great book!

[base_chakra]

How many slashdotters clicked the link not just to see a high selling price, but because the book actually sounds damned interesting?

How About Negative Feedback Loop?

[InvisibleSoul]

Surely there must be examples of negative feedback loops as well then... where two sellers repeatedly set their prices lower than the other, causing an item to drop to a really low price. Someone go find those items! Group buy!

Re:How About Negative Feedback Loop?

[James McGuigan]

Well price goes down to a penny and they both cover their costs via the £2.75 postage and packaging. It does happen reasonably often with certain books.

Fly Sex

[antifoidulus]

Did you ever stop for a moment and think that maybe people are willing to pay a huge amount of money for some info and pictures of hot, hot fly on fly action? Worth every penny.

Re:Fly Sex

[The Wild Norseman]

Did you ever stop for a moment and think that maybe people are willing to pay a huge amount of money for some info and pictures of hot, hot fly on fly action?

Hello, McFly!!

robot pricing/selling meets robot buying

[schlachter]

Wonder how the buying robots (i.e. inventory replenishment, etc.) will react to the run-away pricing robots...in this fully automated loop!

let's get to the source of the problem

[Thud457]

You know, if we kill all humans we wouldn't have these kind of problems.

you're welcome.

es de dominio publico

[Thud457]

I found a better price.
sucker.

Re:Did they design this system or just implement i

[Obfuscant]

I've seen books on Amazon before that were in the thousands of dollars - books that you can find at any used bookstore for 50 cents, we're not talking ultra-rare stuff here. So this isn't a new bug and they're bound to have had complaints many times before. I think this is the first time they've let the loop get this far though.

I've seen the same thing, but it was always in a context of a dozen different stores all selling the same used book. All but one would have a reasonable price. One, two, maybe five dollars. But one store would be above $1000. I could never figure it out, and the one time I sent email to the store asking why they were so high I didn't get an answer.

I assumed it had something to do with a money laundering operation.

Automated price adjustments aren't new (sort of)

[Gondola]

In EverQuest, an MMORPG that predates WoW, people have been using third-party software to automate many things for years, including selling items in the Bazaar (marketplace.)

Due to common errors in logic, it was fairly easy to spot, and somewhat easy to exploit, these scripts. The scripts would re-price their items based on other items for sale, either to lower the prices to just below the lowest price, or raise them when competing items were sold and theirs were the only items of that kind for sale.

I know this isn't real currency, but I think it's interesting to see the parallels.

No supply/demand

[zippy590]

I've noticed the strange pricing of used books on Amazon for quite a while now, didn't realize that the sellers were using algorithms to set them. One of the strangest thing I've noticed is books with limited appeal and 50 or 60 copies available going for outrageous prices. I don't think these robots have any idea of what price the books actually sell at, only what others are listing them for. It's kind of like buy-it-now on ebay. When I auction against buy it now items I usually get about 30% of the buy-it-now price.

Re:slashdot affiliate reference in URL...

[Thud457]

I bet that Russian dude that invented the lap giraffe[*] has two copies of that book.


[*] Funny story, apparently some people think what they see on teevee is real.

Air Jordan 1

[zhusonia]

We believe all clinets love the fashion goods and the Air Jordan will meet your demand,it offers you of Air Jordan 1,Air Jordan 3,Air Jordan 11,and the one is more catch your eyes is Air Jordan 13 which is Cheap Air Jordan.

Remorse...

[Aaron32]

Now that robots are learning remorse, I'm sure they'll feel bad about it soon.

But it's really, REALLY good

[Paul1969]

Seriously. Read it once, and you will ace the course.

And crappy game CDroms for 500$ too

[doccus]

Actually *was* wondering what that cd of, purportedly, the worst pinball game ever (3D Pinball Express) , was doing priced for 500$.. Now i know why

Re:Did they design this system or just implement i

[jd]

If they continue to launder money at that rate, there'll be a world shortage of washing powder.

Save $23,698,653.98 When You Own Funny Things!

[hilaritous]

Good News: Own Funny Things: 101 Cartoons for $23,698,653.98 Less than a book about Fly genetics http://xdose.com//67 - http://xdose.com//68