Hacker News new | past | comments | ask | show | jobs | submit login
Amazon’s $23M book about flies (2011) (michaeleisen.org)
186 points by YeGoblynQueenne on Jan 27, 2018 | hide | past | favorite | 106 comments

There was someone who saw his own product being resold at a higher price. So he placed an order with the reseller while substantially raising the price of his own product. They had to buy from him at a big loss or take a hit on their reputation.

This is a very common scam in Eve Online.

A buy order for an extremely rare item in one place, and slightly cheaper sell order elsewhere.

People notice this, and try to buy the item, intending to move it and sell, but as soon as it's sold, the buy order disappears.

Clever. Would only work if the offending seller didn't already have inventory, but the downside (temporarily depressed sales) is not very large.

Such speculation happens in a lot of resale marketplaces. Stubhub, for example, allows speculative selling (selling tickets you don't have). There are "brokers" who see tickets listed elsewhere (like Ticketexchange), create mirror listings on Stubhub at a higher price, then buy the tickets from the other site if they get an order. That's why you can see +/-40% of listings vanish from a Stubhub event page at the 4-hour-before-start mark -- sellers generally have to pre-fill listings after that.

If an order is placed and the tickets don't exist (or the seller can't get them at a price that allows profit after all and chooses to flake out), Stubhub can just refund the buyer and wash their hands of it. Easy peasy.

That's neat -- similar to a short squeeze.

I feel like this is illegal... not sure specifically for what though. maybe racketeering?

Which part do you think is illegal? Reselling at a higher price like that is called arbitrage, and it is perfectly legal.

I'm an Amazon Seller myself - the reason that other sellers add to the listing with products for sale (with crazy price), is they get access to all the metrics on the listing, traffic, conversion rate etc etc. It's a ploy to get access to the sales data.

The author makes the assumption that some buyers will pick the higher rated seller. This can certainly happen, but almost everyone buys from whichever seller Amazon puts in the buy box and so this would lead to virtually zero sales. However the buy box does not always list the cheapest price. Amazon uses seller rating as a large portion of the decision of who gets that buy box. By having a better seller rating and a higher price, Amazon will sometimes give you the buy box and you will get the sale. So in the end this method works not because the buyer is shrewd with regards to seller ratings, but because Amazon doesn't always suggest the lowest price. If you always want the best price you need to click through to look at all sellers.

I found the first discussion before posting, but I figured after 7 years there would be enough people who hadn't seen it the first time around to warrant a new post.

Is that not OK?

I didn't notice the second conversation. If I had, I'd probably not have re-posted it so soon.

Edit: Actually, I can't find the 2015 conversation in the search.

No it's fine, I just post the links because people like to look at previous discussions.

To be precise, it's fine once more than a year has gone by since the last significant discussion. Otherwise we treat reposts as dupes: https://news.ycombinator.com/newsfaq.html. But if a story hasn't had significant discussion yet, and it's particularly good, reposts are ok. Allowing good stories to have multiple chances at getting attention is a way to mitigate the randomness of what gets seen on the /newest page.

OK, thanks!

fwiw I wouldn't worry: GP is just adding the previous links so that HN readers can see those previous discussions. If it were a problem you'd perhaps not be on the front page. dang is a moderator.

Re-posting is fine here really, as long as it's innocent (!). The system the site runs on actually has an automated re-poster AIUI that puts stories back on to the front page if it looks like they missed out on the attention they deserved.

It seems the couple of mods have the power to change the placement of stories at will too, not sure, moderation is opaque here.

>> Re-posting is fine here really, as long as it's innocent (!).

Well, I mean. Obviously I noticed that the first posting of the story had a very high upvote count. But I was going to post it anyway by that point (I got the story from another article I read today on HN, discussed in this thread: https://news.ycombinator.com/item?id=16243067, and thought it would definitely be interesting).

I can see that being an exploitable tactic (search for very old articles with high number of votes) but not "by hand"- one would have to scrape HN for that sort of thing to work.

I digress- I don't know how I'd prove my post was "innocent". I do try to maximise my karma so the high upvote count on the original story was definitely a motive. I try to use karma as a guide for what to post. Not that it's working :)

>I do try to maximise my karma //


It's like a kind of game?

Enjoyed this read, reading the comments section makes me wonder how many people have silently made money over the years arb'ing out-of-whack algorithms such as this.

Also reminds me of local competitor retailers exploiting Amazon's policies + algorithms to incur cost at Amazon: Amazon promises the lowest price for game pre-orders, and retailers would deliberately drop the price for a game hours before launch, forcing Amazon to honour the low price for all pre-orderers.

> arb'ing out-of-whack algorithms

How would you arb this?

If there's only two copies on offer and you believe the higher priced seller is planning to buy from the other: Buy the book from the lower priced seller, list it for sale at a much higher price, and order it from the higher priced seller before they change the price.

I wouldn't risk a million dollars on that strategy, though.

Would a good tip-off be that the higher-priced seller won't have two-day shipping enabled?

Not necessarily -- if the first seller has it enabled, the second seller would just order it to be shipped directly from the first seller to the buyer, and never have it touch their inventory.

Fun story! A family member wanted to buy an air-fryer online. She found it on Walmart's site, then found it for $20 more on amazon. At first she tried to buy it from Walmart, but got frustrated with the process (setting up an account and stuff) and so she just went and paid the extra to order it from amazon. A few days later, she gets an air fryer in a Walmart box! At first she's super confused, and thought she'd mistakenly bought it from Walmart, so she went to cancel it from Amazon, and saw that that one had already arrived. What had happened was the Amazon seller was getting orders on their page, then immediately going to Walmart, ordering the fryer, then shipping it directly to the recipient. Kinda genius. Basically no overhead, no stock, just acting as a hidden walmart portal and pocketing a percentage.

Wait, the second seller wouldn't be able to set the buyer's address for the first seller at checkout. At least, Amazon doesn't have such features -- the seller would have to manually add the buyer's address to their own list of addresses and so on. There is no easy "drop shipping" between sellers like that.

I am saying this because I wanted to do drop shipping with sellers on Amazon, to arbitrary addresses via an api. How can it be done?

zinc api

I really want to know how much that person banks because I will quit my day job right now.

arb means arbitrage ?


I think this no longer happens. Amazon seems to have implemented an algorithm that suspends listings whose price is too far out of range. Even by a hundred or so dollars in some cases.

As a seller:

Yes, sometimes they'll suspend an offer if they think it's a mistake, they send a pricing notification. However, it's easy to relist at that price if you really want to, you just need to put a min and Max price in their system, and as long as your price is in between they'll accept it. The reason for this is if sellers make a fat finger mistake, so it makes sense they have a way to override.

Separately, they sometimes suppress the buy box if the price is over MSRP or more than other retailers. For a current example, see https://www.amazon.com/CollEGGtibles-12-Pack-Egg-Carton-Seas...

You can still buy, it just shows up as other sellers and needs an extra step to purchase. And it definitely hurts conversions, as I have seen on products I sell.

Oh I've always wondered by the other sellers thing happens!

I'm curious how you found this example. I don't expect Amazon make it easy to query/list this sort of thing.

I knew hatchimals were a popular toy that sometimes had more demand than supply, so I searched for it and went through the results till I found one that was suppressed.

I actually have a database with tens of millions of asins that I have used in the past to generate lists of such items, but I just needed one example so didn't bother using it for this.

And if you check now, you'll see Amazon is back in stock and it's no longer a good example.

For anyone wondering where that weird 1.27059 factor comes from: Amazon takes 15% commission (+ a fixed commission fee) for book sales. 1.27059 * 0.85 = 1.08(00015), so (practically) exactly 8% price difference (or possible profit margin).

I've noticed this on Amazon. I searched for a fiber supplement which I buy at Walgreens for 22$. One day I saw it for sale on Amazon for 15$! So I put it in a wishlist and waited. When it came time to buy it the price had changed to 30$, which was of course more expensive that going down the street. It was really strange. I have kept it in my wishlist and the price has gone between $20-$30 and the price seems different every time I look, sometimes up, sometimes down. It's kind of mind boggling but I think I'll stick with my local Walgreens.

There are probably all sorts of arbitrage scams that make things look weird. For example I could imagine you could get in to baskets and lists by having a low price whilst not having product available to ship; then increase the price: sufficient buyers don't want to search again so you sell at the higher price fulfilling the order by purchasing from the lower priced seller ..? Amazon probably can't pick up all such systems, possibly they don't even try for low grossing sellers??

Who says it's a scam? Most things fluctuate in price.

Amazon themselves sold me this sapphire ring ( https://www.amazon.com/gp/product/B00OJ2BS10/ ) for $6. As I type this, they're charging $35 for it. It's the same product and the same vendor; the price varies intentionally.

I just noticed something simmilar to this [0], with three sellers offering a book for $1,499.95 (new), $1,634.06 (used), $3,361.91 (used) (plus shipping, of course).

The book itself has other listings, such as [1] for as low as $14.76 used, or [2] for $37.28 new [0] https://www.amazon.com/gp/offer-listing/B015X4RCD0/ref=sr_1_...

[1] https://www.amazon.com/English-Words-Linguistic-Introduction...

[2] https://www.amazon.com/English-Words-Introduction-Heidi-Harl...

This makes me wonder if it might be possible, for certain commodities, to buy out the entire market (across all retail websites) and immediately offer the items for resale at a higher price in order to expose naked offers. Though unlike doing the same on commodities exchanges, it would probably never be a profitable enterprise, as these naked sellers will simply cancel orders they can't fill.

I've bought Amazon out on some items when they have a sale and then sold later for a profit.

The best items to do this on is discontinued items that will still be in demand, as you can sure that eventually there will be no stock. Or when they have a significant sale on items that still sell reasonably well at regular price. The significant shipping fees for reselling mean you need to have a large discount. You also have costs of returns, storage, etc. You typically want a minimum of 15-20% gross ROI after those fees.

For example, I've bought shavers at $180 and resold at $290-$300, from Amazon and on Amazon.

I appreciate the frankness, but this is a lot like the usurious trade of concert tickets: not a very noble thing to do, to make an understatement.

If concert halls / theatres actually cared about scalping they’d require ID on admission. But they don’t, so they don’t. They care about concerts selling out, and they want to outsource the risk management of optimising the price to the market. It’s like a futures market they can hedge in. All the crocodile tears in the world won’t change that.

I will maybe, maybe accept that ID’ing is a barrier in the states, but in Europe everybody and their dog has id. Yet: scalping persists.

I don’t blame the resellers anymore, frankly.

I think amelius is more concerned about the prices concert-goers have to pay for tickets, rather than about venues' profits.

Yes, that’s my point. Event organisers cry incessantly about scalpers and all their purported anti-scalping rules. “Oh no, scalpers! They make it so expensive for the fans to get our reasonably priced, accessible tickets! Which we totally priced that way because we want you to have cheap access—not because we want to exchange potential profits for the guarantee of selling out!” /s.

Print a name on the ticket, check ID at entry. Zero scalpers, “problem” solved. They know that. Yet here we are. Why oh why?

Edit: to clarify, I’m trying to say that an event organiser has value from the concept of “selling out” beyond simply ticket revenue. It looks good, it’s a label that evokes exclusivity and popularity in audiences , which is important for future events. Big festivals are in the news every year with a new “sell out record time! Five femtoseconds omg! We are electrolytes, you are plants! Crave us!” Conveniently glossing over the fact that scalpers now own half the tickets. When pressed, “oh no, we hate scalpers! Boo. If we catch you, ho boy! We’ll give you a heck of a diddly talking to! Bad scalpers.” — “will you do anything to stop them?” — “yes, we all spend five minutes each day farting in their general direction.”

Give me a break.

I get what you're saying- but that's not absolving the scalpers, it's just apportioning some of the blame for the exploitation of concert-goers to concert organisers and venues.

It's like having a mafia and a corrupt police force, both. It's not like because the cops are corrupt, the mafia gets away scott free. They're both to blame that things are going down the drain.

In fact they're the same kind of people, following the same kind of tactics, just from different positions in society, like.

It's fairly simple to show that trading in concert tickets is Kaldor-Hicks improving given reasonable assumptions. I've made the argument in the past.

In this situation, it's arbitrage across time, and provides a benefit to customers buying later. It's meeting demand, which is good.

It's only a tiny minority of my business, because it's a lot easier to buy in bulk from wholesale in many cases.

I have talked to people that do millions just using this model, there's nothing wrong with it. Generally when Amazon has a sale they strictly limit how many you can buy to 3 or 27 or something, so it's not like you're buying out the market - just taking savings from when it's on sale and meeting demand at a different time.

You can't increase tickets for a popular event, but an optimal strategy will not sell out every ticket to every concert. Thus ticket resellers are a net loss of sales.

Ticket resellers only exist when the people selling the tickets directly price them below the market clearing price.

Ticket resellers move tickets to people willing to pay more, which means they have a higher utility.

If there were enough tickets for everyone that wanted to pay the set price there would be no room for profit for resellers.

Oh and in general you can increase tickets by booking a larger place or doing whatever it is on multiple dates.

Measuring utility of luxury goods by price is saying that rich people enjoy the concert more than poor people, which is clearly untrue.

Rich people pay more not because the concert offers more utility but because the same number of dollars offers less utility.

Well, it's an extension of thinking people should be able to make voluntary transactions. If someone is willing to give up their ticket for $X and someone else is willing to buy the ticket for $X then it is a pareto improvement for them to trade, since they both get something they value more than what they have.

As an extreme example of why market rationality is useless outside of academic debates, people who buy heroin are making a voluntary transaction - so it's a pareto improvement for them to trade their dosh for junk.

That's assuming that "making a voluntary transaction" is really and uncontroversially a "pareto improvement". Personally, I have no idea what that means, but you can replace it with any term:

  If someone is willing to give up their ticket for $X and   
  someone else is willing to buy the ticket for $X then it is 
  a <gobbledegook gobledegook> for them to trade, since they 
  both get something they value more than what they have.

  People who buy heroin are making a voluntary transaction so 
  it's a <gobbledegook gobledegook> for them to trade their   
  dosh for junk.

Heroin produces externalities, which is the economic justification for its regulation, and why voluntary trades involving it aren't always pareto improvements.

You've made it clear you don't have any economic training, so I don't see the point in continuing this. You have all the terms you need to look it up if you want to learn, or just take some econ classes somewhere.

You're sure you don't want to dumb the economic language down a bit for the poor untrained person, then? This is a forum for technologists, after all, not economists.

But, if you don't, I'll just assume that what you say would be obviously wrong, had I "economic training" and you're just wrapping it up in misused technical language to obscure that fact.

In my experience that's usually what's up when people introduce technical terms in everyday conversation outside academic fora, and where ordinary language is fully adequate to debate an issue.

I apologise for any condescension, but basic economic theory really is important to analysing this.

This really is stuff taught in econ 101, it's not like you need a very advanced framework. All of the terms have a fairly easy to understand Wikipedia page.

Here's the simple definitions of the terms used:

Pareto improvement is a change made that benefits at least some people without making anyone worse off.

Kaldor-Hicks improvement is a change that, if it were combined with a transfer of money, would be a pareto improvement. In other words, the people who gain from the change gain more than the people who lose from it lose.

Externalities are effects of some action that happen to someone else - the classic example is pollution. If something produces externalities, that can be a reason to ban voluntary transactions or actions.

Ah, thank you for clarifying. I must confess that the above were exactly my attempts at understanding the terms you used from the context of your comments- so of course my comments were spot on.

For instance, in the case of scalping, those who are worse off are the concert-goers who don't (go to a concert) because either they can't afford the artificially inflated prices, or there is not enough space for them in the venue. etc.

If there's not enough space in the venue that is obviously not the scalpers fault.

Regarding prices, assuming a ticket ends up going to someone willing to pay the price the reseller charges, then that state of a affairs is a Kaldor-Hicks improvement over the ticket going to someone else who paid less for it. Because the person paying more has more consumer surplus from going, and therefore it is more efficient if they go - any money changing hands is irrelevant to a Kaldor-Hicks analysis.

Scalpers can only exist if there are more people wanting to pay list price than seats. Scalpers ensure those seats go to those who assign higher value to them, rather than randomly. This improves Kaldor-Hicks efficiency.

Congrats on finishing exon 101. Now learn econ 201 where they teach about monopoly and monopsony and why someone having access to more money doesn't mean that person derives higher utility from their purchases.

Can you point me to an analysis using more advanced tools that concludes that "scalpers" are not welfare promoting?

I may have been a bit sloppy talking about utility interchangeably with willingness-to-pay, but it remains true that someone willing to pay more should get it instead of someone willing to pay less, or else the situation admits a kaldor hicks improvement. (There are weird cases with weird demand curves where this may not be true, which is why I specified "reasonable assumptions".)

You are right. The person hiding behind big words is arguing from definitions without understanding the context for applying them.

I think cutting the poor out of going to concerts produces externalities too; financial profit seeking doesn't care about the cultural fulfillment of the population AFAICT.

I mean, if you think poor people attending concerts is a public good, have the government subsidize it. Which to some extent they do with public television stations like PBS or the BBC.

This crosses into a personal swipe, which is against the rules here. Please be civil when commenting.


You're right of course. I myself get very uncomfortable when discussion becomes personal and now I did it myself.

I'm really sorry, ikeboy.

In addition to what others have commented, that's just saying that the venues exploit the concert-goers anyway, so why not just join in and help yourself to some of the loot?

There's no exploitation, just market clearing prices. Or could you define exploitation?

Why do I need to "define exploitation"? We're only discussing the behaviour of venues and ticket resellers. What you say the venues do and what the resellers do is exploitation and that should be enough for the purposes of the conversation.

The way I understand the term exploitation, neither the venues nor the resellers are exploiting anyone. So you must be using it in a different meaning, much like I've seen Marxists use the term to encompass all employers.

I don't know any Marxist theory (it's economic theory, after all), but, maybe,

Is there something wrong with the Marxist definition of exploitation?

Well my main complaint with how it's used in practice is that it carries a negative connotation, so if they successfully convince you that every corporation with employees is exploiting by definition then they have a moral argument against capitalism, an ought derived from an is.

Is there any way to frame "exploitation" that doesn't carry negative connotations?

I don't really have anything to do with Marxists ("they") and I don't see how their attempts to convince of anything are relevant to our discussion.

I think Ikeboy may have actually nailed your objection: you appear to dislike touts, resellers, etc., because they increase ticket prices whilst appearing to add no value to the system.

The same objection is levelled at the notion of employees who are restricted from accessing the financial gains that the value of their labour leads to, that the owner of a company - a rich person say who inherited their money - is possibly getting more money just because they are rich whilst the workers, when classed as employees, are doing all the work that creates the value represented in the financial profit that the owner walks away with.

So, Ikeboy is saying you appear to object because the resellers are interjecting themselves and extracting the value without creating any value themselves, much as a corporations stock holders might. He appears to be saying "what's wrong with that", it is at the heart of capitalism after all.

Not quite. I'm making the claim that they are adding value to the system, by correcting inefficiencies.

"Marxist" is the new indicator for Godwin's law. It's a term free market fundamentalists who don't understand the real world trot out to show that they have nothing substantive to say anymore.

You are ignoring transaction costs. If I get 20 tickets for 50$ sell 8 of them for 150$ then that's a net win for me and it may not be worth trying to then sell the final 2 tickets at the last minute.

Both the band and fans may prefer selling out the concert at use a price that maximizes utilization. A scalper is trying to maximize returns and does not care about 'selling out' just maximizing profit while minimizing effort.

If 8 sold for $150 then there is surely a price between $50 and $150 that would sell at 20. $50 cannot be the market clearing price given reasonable assumptions about demand curves.

Yep, but just because they 'could' sell them does not mean it's worth the effort to actually sell them.

Hell having a reputation as the guy that always has tickets may be worth regular lost sales. In the same way premium brands may sometimes destroy inventory vs offering discounts. EX: http://www.slate.com/blogs/xx_factor/2010/01/07/why_hm_destr.... Note the recommendations to use a textile recycler and zero recommendations to sell at a lower price.

Remember models are fun, but the real world is complex and you need to model reality not play with math.

I am recommending the list price be raised, which would get rid of the need for resellers. The resellers are capturing some value from an inefficient system, and the system became efficient there would be no value left to add and therefore no resellers.

Again bands and resellers are making different optimizations.

A percentage of people showing up will buy extra merchandise which increases revenue. This captures value from people willing to pay higher ticket prices. So, if the band picks a price to sell out our reseller can buy all tickets raise the price and not sell out for a net profit. Meanwhile the band makes less money and fewer fans get to see the show.

Remember, in the current system both the venue and bands are maximizing their profit.

I've seen cases where the scalper bots bought all of the tickets, and re sold them for two to three times the list price. And then I get there to a show that's maybe between 50% and 60% of capacity that's been 'sold out' since quite literally 30 seconds after the tickets went on sale.

If scalpers are buying 40% of the tickets and not finding buyers they'll quickly go bankrupt.

If they are finding buyers at the higher prices, that means the market clearing price is well above the list price, and scalping increases total surplus (again, by moving tickets towards those who value them more).

If the ticket sellers are artificially reducing the cost then this is what will happen.

They're buying all of the tickets, increasing them by more than double and only selling between 50% and 60% of them, so they still come out ahead.

> more than double

That's optimistic. Ben Folds at the Washington Center for Performing Arts, tickets from $15-29...

After scalping bots? Cheapest ticket was $130.

Seems like nonoptimal behavior on their part - once they've sold all the marked up tickets they can, they should sell the rest for a smaller markup until they're out.

And, it's also nonoptimal behavior on the events part, because they could increase total revenue by decreasing seats and increasing price.

Or maybe maximum money changing hands isn't always the best proxy for maximum utility.

No, you measure total surplus which is supplier surplus plus consumer surplus.

On the other hand, isn’t the alternative the product being impossible to find? Storage ain’t free.

As Canada Bill Jones said, it’s immoral to let a sucker keep his money.

Two futures traders actuall managed to buy every onion in America in 1955. They tried to sell them all in Chicago which lead to a massive price crash their and shortages everywhere else. Congress then passed the Onion Futures Act.


That's fantastic. It can sit alongside the https://en.wikipedia.org/wiki/Video_Privacy_Protection_Act as an example of fixing the narrow problem while carefully avoiding fixing or even considering the wider perspective.

> these naked sellers will simply cancel orders they can't fill.

And risk the negative feedback?

They won’t get negative feedback but run a very real risk of getting booted off the platform. Preshipment Order Cancellation is a big part of the Order Defect Rate. Let that get too high and you won’t be selling on Amazon again. In most cases it’s worth taking a loss on an order to avoid canceling it.

So if the one seller counts on being able to buy the other seller's book if the first seller actually makes a sale, what happens if someone buys both listings at the same time and the higher priced listing is unable to source the item? Can the seller just cancel the sale for no fee? Why wouldn't everyone with a little time on their hands make 1.2x-priced listings for every single item available for sale on amazon and dropship from the cheaper listings?

I wrote a hymnal[1], and currently Amazon has a copy for sale at double the price of a new copy from my website. That put me in the unusual position of leaving a three-star review of my own book. I sell the book as nearly at cost+shipping as possible, as I don't want to profit off a religious book — but that doesn't mean I want someone else to profit off it, either.

[1]: http://olpbhtb.com

Based on the other replies in this subthread, I think it's safe to say the following link won't be used to order.

But I was very curious what the Amazon listing contained nonetheless, so: https://www.amazon.com/B06W56HNPQ

could you sell it on amazon yourself for the same price you have it on your website? That would discourage anyone to sell it for a profit

Honestly, it's not worth the time for me to do that; it mostly sells by word of mouth. I'd rather say, "I wouldn't recommend buying it this way," and let people carry on doing as they wish. I'm not that worked up about controlling others' behavior.

You have no ISBN, right? It's fascinating that this bookseller took the time to create by hand an entry for your book in Amazon's product database.

In a way, not even not having an ISBN is a sure way to "withdraw from the international book trade"[1].

[1]: http://www.anagrambooks.com/no-isbn

Yes, that was part of the reason I didn't bother with it. I was frankly surprised to see that anyone bothered to list a copy at all, since the market is very small. (Primitive Baptists of any sort are a tiny, tiny fraction of a percent of American Christianity, and the appeal of my book, if any, is to an even smaller subset. But de gustibus non est disputandum.)

[EDIT: close paren. I couldn't take it!]

I like how they were still charging $3.99 for shipping. Maybe Amazon can implement Excel's useful circular reference warning (/s)

To be fair, the Kindle version is only $4.3M.


What does his have to do with the article other than driving traffic to that blog?

Amazon has had private label products for years, like many retailers do.

> But Peter Lawrence can now comfortably boast that one of the biggest and most respected companies on Earth valued his great book at $23,698,655.93 (plus $3.99 shipping).

I was wondering if pricing on Amazon would be sufficient evidence to get into a Guinness World Records for the most expensive book offered in public??

If there is such a category, then I think they'd have to actually sell it at that price, otherwise an author could just put his book up on eBay with a billion dollar starting price.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact