I have my first cover article in the 2016 November/December issue of Online Searcher. It's about Wild Colorado and change management. I would link out but they have it only in print right now.
I thought I had posted about the closing of British libraries earlier this year, but apparently I did not. Back in April the BBC had a rather bleak article on libraries closing in England. And it wasn't small numbers; if BBC numbers can be trusted, a full quarter of staff jobs have been lost and about 200 traditional library locations closed with another 100 closing this year. The article is depressing to say the least.
Pair that article with this Guardian article and it seems that British libraries are in widespread decline. It's hard to tell how much to read into two short articles, and the Chartered Institute of Library and Information Professionals (formerly British Library Association) is not acting like there is a crisis. In the US a 30% drop in adult patron usage would be a crisis. The quote from Mark Taylor seems to evince typical bureaucratic density, "these figures strengthen the case for rapid modernisation of library services across the country ". Well, that is what he would say if the numbers showed a 30% increase, or no increase... Statements like that only serve to lock people into their perspective camps and do not build allies.
But again, without a more detailed understanding of British libraries, their usage, funding, etc, it is hard to know what this means. There is also a significant difference to consider- there is no First Sale doctrine in England. Instead they have the Public Lending Rights Act of 1979 which actually allows authors to collect payment based on the number of times their books are checked out. It is hard to determine how much this changes the game for them. But my understanding is that it has a detrimental affect on their ability to create a large catalog of lending materials. As I understand it, they still have to buy each copy and then make payments for circulation on top of that price. Just imagine how much money they pay for Harry Potter and every other bestseller. My limited and anecdotal understanding of the practical effects of the PLR is that most British libraries cannot create the types of collections their American counterparts have, which probably augments the precipitous drop in library use documented above. We see quite the opposite in recent PEW reports where both lower and upper income Americans continue to utilize their public libraries. The British system is also much more centralized as significant funding for public libraries across the U.K. comes directly from the government, as do the PLR payments. At least that is my understanding of their system. There could be, and probably are, some fairly significant gaps in what I have communicated above, but the basic concept surrounding the PLR is accurate: British libraries make payments based on circulation.
So while it is distressing to see this happening in the U.K, the legal environment in the US would seem to indicate that US libraries have a chance to maintain their value.
I have been chewing over the avalanche of sensational news coverage of the fatal self driving Tesla accident. This USA Today column while published later than most coverage (ostensibly so the writer could do his journalistic research) is fairly typical of the shallow and misinformed nature of writing about the accident. The problem is not so much the facts, as it is the chaotic and unsettled nature of AI research. So you can't blame the journalists- the problem is that they are wading into an uncharted ocean.
But to put things into context I can offer this article about a combat AI beating a seasoned combat instructor every time. By default, a combat simulation is very different from a real world environment in the number of random variables that are presented to the AI. But the point I want to make is still relevant, because AI will continue to evolve at a far faster rate than human cognition. If AI continues to develop at even half the speed and capacity it is currently at, we can expect AI to be superior to human operators at some point in the future. The combat simulation demonstrates this fact quite nicely, as a real world environment is different from say a chess match with a finite number of variables. Machine intelligence is faster at analyzing data and interpreting it. This has been true for a long time. So it is only a matter of time until the fail rate of self driving cars is much better than human or machine assisted driven cars. AI simply has to improve and be able to match certain human cognitive abilities. Depending on who you are reading that could be just around the corner or a long way off. Which is again why it is so frustrating trying to make sense out of the Tesla crash, as there is simply too much difference of opinion on this emergent technology to communicate anything meaningful in layman's terms.
A 2 year ereader program in Ghana has produced dramatic results. Overall students in the test group (the ones given an ereader loaded with 240 books) produced remarkable results:
The blue line show significant literacy skills for students with ereaders. But before anyone jumps to a conclusion that this means ereaders themselves are the reason the students did better, remember that the study was in Ghana where access to books is really what matters. I think the study proves what we already know: access to books is what really matters for literacy development during those critical years. The key piece of the study is the 240 titles preloaded on the ereaders. The students outside the test group probably had access to a fraction of that number of books.
This has been a bit easier to predict, because the digital music industry is fairly mature, and it has a longish history of this kind of behavior. Artists are "windowing" their music on Spotify. Windowing is a deliberate strategy to release music in stages through different venues rather than all at once. So if you really want their music as soon as it is available you will have to pay up front. After 6-9 months, artists and labels then release their music to the much less revenue intensive streaming ecosystem. Or a streaming site like Spotify or Apple Music will pay for exclusive rights to stream an album. Currently, Tidal is a very popular source for exclusive streaming rights like this.
While the article focuses on the UK, Taylor Swift's dramatic and public exit from Spotify suggests that US artists are thinking the same things. Ultimately, the issue revolves around how to maximize profit on an artistic production. Artists and labels both feel that streaming revenues are too low to justify full and immediate release of new content to all streaming sites. Consumers on the other hand don't want to pay.
The bigger question for providers of content is whether or not the streaming model is sustainable or will fade out?
When Netflix first went digital, I started telling anyone who would listen that Netflix would die off as a company. Obviously I was wrong. They have become a behemoth. But I was right about the underlying parts of my argument: Netflix's catalog has shrunk and shrunk drastically for the same reasons libraries struggle to afford scalable digital materials. Licensing materials is expensive and there is no need for a middle man in the digital economy so studios are simply pulling their content or asking for so much money that no one can afford it.
Netflix has and had good leadership who got ahead of the studios and started producing quality, original materials, and locked in a highly coveted exclusive contract with Marvel that is paying dividends. It is still possible that my original prediction could prove correct, but because of fragmentation rather than a shrinking catalog and rapacious movie companies. This article and graphic on Netflix's shrinking catalog and their strategy for dealing with that are prescient for libraries clinging to older strategic models of legacy catalogs. Libraries are less nimble and quick to respond, but at least at my library we are consciously trying to learn from Netflix strategy to push the content our patrons want a little more as our catalog shrinks. As the gross numbers evince, Netflix's strategy is the polar opposite of libraries who have lots of content, but are more passive about getting that content to their patron. Netflix on the other hand never had more than roughly 8000 titles (even the smallest libraries have close to 8k titles), but made sure that the content is valuable to someone in their customer base. My library on the other hand has more than 8k non fiction titles alone, most of which never circulate. Netflix is still going strong and adding customers with only 5500 titles. Just amazing.
The "data guy" at Author Earnings has posted the slideshow for the 2016 Digital Bookworld presentation on the ebook market. They are one of the best sources for ebook data and do a better job than most at correcting bias and assumptions in the data.
One of the obvious takeaways is the small monopoly that seems to be developing for ebooks revenues: 95% of sales go through 5 online vendors with Amazon taking 71% of the market. It is also interesting to note that while nonfiction is the largest category as presented, it is only so because nonfiction does not get broken down into smaller categories like fiction. It is also worth noting that my hunch that publishers are taking cues from other media markets is clear in the presentation as the video game industry is presented as one industry that migrated revenues away from a dwindling market.
There is a lot to digest, but I leave you with the visually arresting slide on Amazon's market share:
I am at a sub conference of Computers in Libraries listening to Dave Snowden discuss "Complex Adaptive Theories Systems". What is interesting is that much of his theory seems to align well with my own ideas about the Library of Things and Information Ecology. His ideas are dense but seem full of potential. What I like is that he connects action and data in a new way. He argues that we need to let the people we work with interpret their own data and use a mixture of machine and human intelligence to analyze the data. Very interesting.
Nate Hoffelder has a great post critiquing the much ballyhooed "fact" that 92% of students prefer print over ebooks. When I first saw that factoid making the rounds it seemed obvious that it was lifted out of context, as I could tell from my own work with students that students interests, needs, and concerns about textbooks cover a wide range of variables, most of which are not scholarly, but matters of convenience and price. Like Nate I have a copy of Words Onscreen (the book the factoid was lifted from) and would highly recommend it. It is the type of nuanced and balanced examination of the digital transition that has been lacking from the conversation. Unfortunately media types don't often get the full context, and entirely miss the point of Dr Baron's findings. But Nate did a great job covering that.
What is interesting to me is that answering the question is not that difficult for the right scholar in the right place. Since I first started working with digital texts in academic settings (Kindle 2), it became apparent that we needed to clarify more than quantify a lot of different variables around digital text. I ran a number of informal surveys that gave the general impression that we also needed to separate light reading from heavy reading, as most readers found little difference in reading something light and easy on screen rather than paper. But it gets messy when we read more complex books. Dr Baron delves into this in her book. I had the same conversation with my Mass Communications undergrads this semester about their textbook and their opinions were all over the map. This is because their motivations were all over the map. We need to separate students into those that read and those that don't, those that are motivated to succeed and those with less motivation. It even seems that taller and more muscular students care less about the weight variable than smaller students. The point is that there are lots of variables affecting student responses and surveys are the wrong methodology for studying the question.
While I tend to believe that print has some advantages over digital when it comes to absorption and retention, I am not convinced that those advantages are endemic. Millennials brains seem to be wired differently than mine, and it may be possible to develop tech that mimics the advantages of print. In my mind the jury is out and we should always be wary of attention grabbing headlines. They tend to be wrong 92% of the time...
This is receiving widespread coverage, so there is not much to say other than to agree that it really stretches the concept behind the term "book". The entrances and exits one seems very interesting. The sample is enjoyable. Before librarians hyperventilate about this, I would remind them that this is actually a good thing for traditional print books, because the more differentiation there is between ebooks and print books the better. Consumers will see them as two distinct products rather than interchangeable ones. As someone who believes that print books maintain a number of unique advantages to ebooks this is a good thing.