Besides time, what's to keep a collective of authors from publishing $3.99 and below copies of things like college textbooks? I don't mean bit for bit copies, but newly written works. Also, such books wouldn't be accredited, but in the long run, that might not matter as much as their being good.
This gets me to thinking: should classes like English still be about disseminating culture? Or should it be about examining it? I don't think culture needs disseminating as much as it needs examination anymore.
I have a hard time distinguishing some of the colours, it would be nice if hashing or other marking was also used.
Yes, colour confused. Why do you ask? (I perceive some pinks and purples as grey and my favourite brown shirt is actually green, according to those who can see all colours properly. I also have a problem with faint red, but bright red isn't a problem: I've never seen Mars as red, but stop signs and stop lights are quite a fine, bright red, thank you.)
The data only show that DRM is weakly correlated with lower ebook sales; or, if you like, that foregoing DRM is weakly correlated with higher ebook sales. While one explanation would be that the presence (or lack of) DRM is the causative factor, I think it's a bit more likely that thoughtful authors both forego DRM and write better books.
1. Buyers prefer non-DRM, and this influences their buying decision.
2. Buyers of non-DRM ebooks are more likely to lend them or give away copies, and the recipients them sometimes buy a copy for themselves.
3. Buyers of non-DRM ebooks are more likely to lend them or give away copies, and the recipients of them discover a new author that way, and then they buy OTHER books from that author. Since the decision to use DRM or not tends to be made by the publisher, the other books by that author are likely to also not have DRM.
I don't think #1 is likely. I'm sure buyers do prefer non-DRM, but at Amazon it is not generally obvious if a given book has DRM or not. There are hints, sometimes. In the product details it sometimes has an entry for how many copies you can read simultaneously on different devices. If that says "unlimited", it generally means the book has no DRM. If it says a definite number, or says that it is publisher-determined, it means DRM.
If you don't know whether or not the book has DRM, it is kind of hard for it to make the difference in whether or not you buy.
The other reason I'm skeptical about #1 is that I don't think most Amazon customers have had occasion to run into a problem with DRM, and hence probably do not even think about whether or not a book has DRM. Amazon has Kindle reader apps for iOS, Android, and Windows Phone, and for OS X and for all Windows from XP forward. They also have a browser-based Kindle reader that works in Linux and allows offline reading. This means that most people are not going to run into one of the primary DRM annoyances--not being able to access their content on all their devices.
Another DRM annoyance is that it can prevent lending. With physical books, it was common to borrow books from friends, and people want to do that with ebooks. Kindle has support for this. The publisher decides whether lending through the built-in Kindle lending system is supported or not. If they want to disallow it, they have to use DRM (I think). If the book does not have DRM, I think it automatically works with the lending system.
I can see that customers might run into this, so that might provide some reason for them to care about DRM. there might be some preference for non-DRM on the part of buyers indirectly because the buyers want books that they can lend. This is the only way I can think of offhand that most users would (indirectly) notice that a book they are buying has DRM--by seeing that lending is disallowed. In this case, it is the lending vs. non-lending that influences the buying decision, not the DRM vs. non-DRM, but since non-DRM is 100% lending enabled, whereas the DRM books are sometimes lending enabled, and sometimes not, there is a positive correlation between lending enabled and non-DRM.
They should do an analysis of their data that breaks it down by lending enabled vs. lending not enabled instead of by non-DRM vs. DRM.
You forgot 4. Sample size is too small to draw meaningful inference from statistical properties.
The money quote right here: "… the only two price points that appear to buck the general trend [were] due to 3 outlier DRM titles published by only two authors."
My personal guess that it's the combination of at least the three points you mentioned. Discovery via grapevine, combined with the ability to move the purchase between devices - regardless of vendor.
It is possible that a "once burnt, twice shy" effect is amplified by the 3/11 rule. It takes one bad experience with DRM'd goods for a person to realise that something is wrong, and once they learn that DRM was the reason for their problem they will broadcast their discovery about the anti end-user experience.
And as any news reporter knows, bad news travel fast.
This gets me to thinking: should classes like English still be about disseminating culture? Or should it be about examining it? I don't think culture needs disseminating as much as it needs examination anymore.