Booktrope

When a publisher sells an ebook, they receive wholesales not retail price. Wholesale prices in the book businss range from about 35% of list (if you use a full-service distributor) to about 5o% of list (if the publisher sells directly to the retailer). You present an example where the list price is about $28, which means the publisher will get at best $14. But your calculations are based on the supposition that the publisher would get the full list price.

Here's another huge cost you don't include for the publisher with print books -- they are returnable and often get returned. If a publisher gets 25% of the books sold returned, effectively the publisher is getting only about $10 for a list price of $28 / wholesale $14. (Return rates can be higher or lower, but there are always returns.) Using the cost figures you use on your referenced example, the publisher loses money on the print book. And, you omit other costs incurred by traditional print book publishing, including shipping, warehousing books and fulfilling orders.

With ebooks there are no returns (a huge factor in the analysis), no printing, no shipping, no warehousing. Wholesale prices are a higher percentage of the list price. We publish ebooks and get about 65% of list from Amazon, Apple and Barnes and Noble (it's a bit higher or lower depending on which vendor) So it's obvous that our costs are lower for ebooks and the share of revenue we get is higher. In other words, list price of ebooks indeed should be far less than print, and ours are.

I think you're misreading that study. It compares the number of women estimated to have their life prolonged because of a mammogram, to the number of women estimated to have been treated unnecessarily because of a mammogram. The issue with whether mammograms prolong life is different than the issue of whether mammograms accurately predict cancer. Depending what research you look at, other approaches are sometimes seen as just as effective as mammograms. This is very controversial right now among people who do cancer treatment.

The point is, however, the article you cite does not say for every correct treatment there are 10 unneeded treatments; it suggests that for every life saved there may be 10 unneeded treatments. The summary doesn't say what they mean by a life saved. Mammograms find a certain number of cancers. Other diagnostic techniques also find a certain number of cancers. These sets overlap. It may be that the authors consider the number of lives saved by mammograms to be the delta between cancers detected/treated due to mammograms and due to other diagnostic techniques. This seems to me to be a valid issue, but it has nothing to do with how accurate mammograms are in detecting actual cancers, which seems to be 80%. (So, for example, what if other diagnostic techniques have a 70% chance of detecting cancer -- this might lead to the 10 unneeded treatments per one life saved statistic)

The issue I have with the article posted above is that it claims the probability of a woman having cancer after a positive mammogram is 7.8% when actual results show this to be 80%.

I did contact a statistics site that had a similar statement about AIDS tests and their response was, Baysian statistics are correct and they saw no problem with telling people that a positive AIDS test only indicated a very low chance of having AIDS.

I am troubled by the mammogram analysis. These sorts of health care examples (mammograms, AIDs testing, etc.) are often used to explain Baysian statistics and obviously make an interesting point. But as someone below points out, these tools are meant to be used in situations where we have incomplete information. We have direct experimental evidence of the accuracy of a positive mammogram in predicting cancer, and the accuracy is far, far higher than 7.8% From direct studies of patients who get positives, it looks like about 80% of them actually have breast cancer. See, for exmaple, http://www.cancerresearchuk.org/health-professional/cancer-s...

It may be that the author of this article simply go mixed up -- he reports than 1% of the population that 80% of the women who have cancer get positive mammograms, while from what I can see, evidence actually shows that 80% of the women who have positive mammograms have cancer. So the problem with the example might just be a matter of garbage in garbage out.

Mammograms, incidentally, are still controversial, especially doing them on an annual basis, because if you,for example, do a test with 20% false positives 10 times, you're pretty likely to get a false positive in at least one. But that probability is still not 92.2%, which is what the article suggests is the false positive rate for a single mammogram.

But it's disturbing to see statisticians flippantly saying things like, only 7.8% of positive mammograms represent actual cancer, when evidence shows, 80% of positive mammograms represent actual cancer. Survival rates for breast cancer have gone up and most everyone agrees early detection of cancer plays a role. I certainly hope that any woman who reads this article recognizes that, if she has a positive mammogram, the changes are much, much higher than 7.8% that she has cancer.

I've seen the same arguments made about AIDS tests. Why is it that statisticians like to use examples of life threatening illnesses and present a Baysian model that vastly underestimates the effectiveness of tests that could be crucial in saving lives?

I don't understand how you can say that, even if you were to search all non-black objects and none were ravens, this would prove that all ravens are black. It also might prove, there aren't any ravens.

And, even if you were to find a black raven, it might just mean that you were mistaken and it either wasn't black or wasn't a raven. Thousands of years ago, the Chinese nominalist philosophers built an entire school out of saying, "A white horse is not a horse", and if you read what they said, they were right. You can say, "A white raven is not a raven" but I think you'd be wrong, because it's easy to paint a white raven, especially if you use a dark background. And please, don't tell me, it's the same with a white horse, without reading a bit of Chinese nominalist philosophy.

And besides, what is black, actually?

I'm the 66 year old founder, Ken Shear, and our company is Booktrope (my co-founders are 54 and 45). I've been around the block a few times but YC has still been the most valuable educational experience I've ever had. I can vouch for this, YC is no respecter of age, in the best sense. The partners might be quite a bit younger than I am, but years beyond me in experience of starting and growing a business so I've learned an enormous amount from talking with them and from their insights into how startups are done. For my own part, a while back I was working for a judge and one of his colleagues, amused or unhappy at something I'd done, asked me how old I was (in that tone, you know). I asked him, did he mean, intellectually, emotionally or chronologically? He wouldn't say, and neither would I. So, the other thing about YC is probably, there's incredible value to be gotten from it and it doesn't depend on your age, but on how you enjoy it.

Well, if he was clueless then two very large and successful tech companies had a clueless guy running their AB testing and showing great results in each context.

I'm certainly not looking for "One Method for Interconnectedness Correction" (especially not, as you put it, with each word capitalized). I'm looking for studies or papers that might have addressed anything like the effect of interconnectedness of web data on AB testing. I think you're saying, you don't know of any, and also that you personally don't think it's a real issue.

I understand if you think this is a non-issue, though I don't agree. The speaker I referenced about asked the statisticians at his company about this and they said it was a non-issue because things balanced out. He thought that was an idealization and claimed to have tested it building in some real world data, and reported that interconnected data of this kind drastically affected confidence levels. He didn't get into the details of how he measured interconnectedness, however.

The example you give seems to me to oversimplify the issue of complex interconnections between data points, as if the traffic on a real website came from one set of referrals, while in reality its much more complex, with referrers inducing other referrers and a variety of campaigns, postings, etc. influencing each other, and over time, overlaid in a fairly complex pattern. In other words, a bunch of interrelated data, very little of which is actually independent of other items.

I'm not really asking for an explanation of this in the comment thread here; what I'd like to know is, if there are any studies or other publications that deal with the issue of how to evaluate tests run on interconnected data of this kind.

The presenter I mentioned did not go into details about what interconnectedness he found, but I think it's quite obvious that people visiting a site do have influence on other visitors, which is at least a part of the underlying issue. On the simplest level, most web sites have share buttons to make it as easy as possible for visitors to influence other traffic. Or, other examples, a trending tweet can massively influence patterns of usage of a website or Facebook page (or many tweets with small reach can cause many small influences); or an RSS feed might influence patterns of tweeting or posting elsewhere. I think there are myriad other interconnections within web traffic. We do a great deal of work to drive traffic that is based on the premise that different visitors to websites are mostly interconnected. It's these factors that give me pause when I think about statistical measures that are premised on an assumption that we're measuring independent events.

If the stats aren't based on samples of real-world data, rather than very rigorously designed experimental comparisons, I think there's another reason to be wary of p-values. [Note, in what follows I probably use some terminology wrong, because I'm not a statistician, but I do think the point is important and I don't see much written about it.] In the real world data is not a bunch of independent events, but events (or data points) that are interconnected in ways that are difficult to quantify. I once heard a presentation by an expert in AB testing at a leading tech company who became wary of his results and brought his concerns about non-independence of events that were being tested to corporate statisticians. (He was concerned that, even though the AB testing procedure supposedly randomized the test, interconnectedness within website traffic was not being accounted for.) By his account, the statisticians agreed it was a problem but recommended that he assume that variations due to non-independence would more or less balance each other out. He wasn't satisfied with this and said that he went ahead and did some more measurements, then ran out all the binomial expansions rather than relying on approximations. When he did this more detailed work, he found out that with at least some web-based AB tests, where conventional statistical formulas showed a p-value of .05, he was getting a confidence level of more like 30%. (I don't think you could call his measurement a p-value because he wasn't using the formulas normally used to compute p-values, but I think what he was saying was more or less that the p-value from the forumalae was .05 when it should have been .30 from a more rigorous look)

As I'm not a statistician by trade I don't keep up on the literature very well, but interconnectedness of data does seem to me to be a very important issue. I'm wondering if anyone can point me to some helpful reading to understand this side of the issue better. In particular, is there any approach to AB testing that can reliably address the issue of data interconnectedness in the kind of situation described above?

Same experience here with a MacBook Pro circa 2012. Often seems slow to re-establish network connection after sleep. I've seen this on a variety of networks. Really just an annoyance, but better management of this on my Mac would be nice.

The problem with Le Guin's position is that it's so Manichean, a world composed of good guys and bad guys with no ambiguities. To her, it's either sell books like deodorant or some ideal system where it's quality, not marketability, that makes books sell. This sounds good, but it's tripe. As you say, the Internet in general and Amazon in particular have opened up publishing -- Amazon getting huge credit because it not only allows freer publishing, but has set up a system that allows many thousands of new authors to actually connect with readers. Yet, Amazon is also a business focused on selling products at high volume with tight margins. These two goals aren't opposed, they're different aspects of a complex business. Amazon can be both a great contributor and profit-motivated, it's not that hard to see how.

Who defines quality, anyway? The tastemakers, like newspaper reviewers, who nobody much pays attention to anymore? Or Le Guin -- she seems to think she knows what's good. I imagine, what if John Kennedy O'Toole had the opportunity to self-publish A Confederacy of Dunces, instead of facing despair when all the traditional publishers turned him down. What if he'd been able to sell his work like toothpaste on Amazon, even if the tastemakers thought it wasn't worth publishing. To Le Guin, this would somehow violate the ideals the publishing world ought to aspire to. To me, I wish the opportunities offered by Amazon had been around back then.