Saturday, September 24, 2011

Are Neutrinos Superluminal? Judge for Yourself.

Yesterday, I couldn't restrain the physicist inside me. I just had to watch the seminar from CERN where Dario Autiero presented the result from CERN and CNGS of a measurement of the neutrino's velocity. The measurement is a tour de force of modern experimental physics, which harnesses amazing technologies such as the global positioning system, large scale data processing, picosecond lasers, ultrafast digital electronics, billion-volt particle beams, and highway tunnels a mile beneath a mountain in Italy. The scientific communication system is also state-of-the-art. The preprint (with 174 authors) appeared in ArXiv.org on thursday night: "Measurement of the neutrino velocity with the OPERA detector in the CNGS beam",  arXiv:1109.4897v1 [hep-ex]. And CERN webcasted the presentation live around the world, with not even a hiccup in the video feed.

What the news reports have failed to convey is that despite the impressive effort, the accuracy of the velocity measurement is teased out of the data with statistical procedures that are sure to come under intense scrutiny.

The way the experiment works is this. Pulses of neutrinos lasting 10.5 microseconds are generated by the accelerator at CERN, pointed at neutrino detectors 730km away under the mountain in Italy. Neutrinos are incredibly hard to detect (they have no difficulty traveling through 450 miles of rock), so only a tiny fraction of them are detected. Over 3 years, 16,111 of the CERN neutrinos were detected in Italy. The shape and timing of each generated pulse is measured and stored to be compared later with the timing of the detected neutrinos. The nub of the matter is shown in this graph, which shows only the leading and trailing edges of the accumulated data:
As you can see, the leading edge of the neutrino pulse is about 500 nsec wide. The red line is the cumulative shape of the generated pulse, the data show the counts and relative timing of the detected neutrinos.

The evidence for superluminal neutrinos is that the red curves at on the bottom, shifted by 60.7 nsec faster than the speed of light, are a better fit to the data than the red curves at the top. The claim is made that the bottom fit is 6 sigma away from the fit at the top. What do you think? Isn't physics fun?
Enhanced by Zemanta

Wednesday, September 21, 2011

Can JSTOR Solve the Course-Assigned eBook Problem?


Just imagine if libraries were like airlines. Your flight to Information could cost $20 or $2,000 depending on how far in advance you book, whether you're connecting through Seattle or going direct, how full the flight is, the day of the week, the phase of the moon, and who knows what else. Airlines employ operations research Ph.D.s who create and maintain elaborate systems to "manage the load". These systems make sure that people who have little choice of when and where to fly pay a lot more than others who have flexibility to seek low fares.

You don't have to imagine very hard to realize that academic libraries are part of a system that's a lot like the airline industry. Consider college students. Once they've signed up for a class, they have little flexibility with their assigned reading. Even a college library with a 10 million volume collection won't be of much use to them, because someone else will have checked out the assigned reading – if it's not on reserve. The student is sent to the college bookstore, where it's a pretty good bet the book won't be on sale. If the student has planned in advance, they might bought a copy used on Amazon, or borrowed it from an upperclassman.

Academic book "load management" was not devised by financial engineers, it arose as a by-product of a paper-based distribution system. Students don't perceive librarians as evil just because they haven't bought a hundred copies to supply the whole class. But the environment is very different when the books are digital. Scarcity of digital books is a manufactured fiction, so it's easy for a student to perceive the library as being complicit in a money-extraction machine, never mind the benefits that accrue globally when publishers are rewarded for producing books worthy of being assigned in a course.

Libraries hate saying “No” to their users, so ebook models that allow unlimited simultaneous access are very attractive to them. Even limited simultaneous access would help meet the needs of students. The same models are very scary for publishers, because they count on profits from selling many copies of course-assigned texts to cover their losses on slow-selling titles. Academic libraries don't have the money to replace that revenue, because the embedded practice is for students to pay for their own textbooks.

A few universities, notably Indiana University, have started to reexamine the role of academic libraries in the digital provision of course-assigned textbooks. Libraries have a better bargaining position with publishers than students, and are in a unique position to integrate digital texts with databases and other types of information available from the library. It makes little sense to tell students to come to the library for quality information- unless they need it for a class!

Publishers also have incentives to find library-based digital solutions for course-assigned reading. If the library can be charged based on the number of students enrolled in a course, no revenue is lost to used-book sellers, peer-to-peer networks or students who don't bother to read the texts. Solutions that work have nonetheless been elusive, partly because it's very difficult to have print and digital models coexisting, but also because budgets are tight all around.

At Ithaka's Sustainable Scholarship meeting this week, JSTOR's Bruce Heterick told the gathered publishers and librarians that JSTOR's ebook program, to be launched in 3rd quarter of 2012, would include a pilot program to address the course-assigned title problem. This news was warmly received by the entire audience. Books at JSTOR will make over 15,000 books available on multiple platforms (including ebrary and NetLibrary) and on multiple devices (iOS and Android). At launch, the ebooks will be offered on a sales model, and all the ebooks will be archived in Ithaka's Portico service so that libraries can be assured that their access will really be perpetual. The icing on the cake is that the ebooks will be integrated with JSTOR's discovery and crosslinking platform, which is very popular with the students and faculty at subscribing institutions.

Although the details of the course-assigned title pilot were not yet decided, JSTOR Managing Director Laura Brown responded to questions by saying that as many as 3 different models would be tried in the pilot. JSTOR's subscriber outreach has suggested that different approaches are needed in different sorts of institutions. I suppose you could think of these models as “charter flights” that can be added to package tours.

Regular readers of the Go To Hellman blog will know that I'm working on a new approach to selling ebooks. I view every market problem as a possible opportunity for the new model, which we call "unglued ebooks". Course-assigned titles are no exception.

My analysis indicates that the market dynamics for some books may favor the ungluing model. It works by aggregating donations by many people and institutions with a stake in a particular book and then paying the book's rights holders who “name their own price” to issue a Creative Commons licensed digital edition. The ebook can then be used without limit by everyone, everywhere. (OK, it’s backwards from Priceline, but we totally have to get Leonard Nimoy as our spokesperson!)

In situations where two titles compete with each other to be put on a course list, the ungluing model introduces a severe form of price competition. A title that is successfully unglued, even at a price equal to the present value of its entire future revenue stream, would have a huge advantage over a competing title that remained on the per-copy selling model. It remains to be seen, of course, whether students and libraries will be able to organize an ungluing campaign to meet the high ungluing prices commanded by books with steady recurring sales. Still, there is a huge variety of courses and books, and thus a reasonable chance that the model will work in at least a few cases.

There’s a great need for experimentation and risk-taking by libraries and publishers in the transition to the digital environment. It's good to see JSTOR step up to that plate. But don’t take my airline industry analogy too seriously. When developing new models for ebooks and libraries, let’s omit the pat-down security searches and checked-baggage fees, OK?
Enhanced by Zemanta

Monday, September 19, 2011

Crowd-Finding the Orphan Books

A solution to the orphan books problem won't be delivered by a settlement of the Google Books lawsuit anytime soon, we learned on Thursday. "Orphan" works are ones subject to copyright restrictions, but for which rights holders can't be found or determined.  A proposal to open them up via settlement of a class action was rejected by Judge Denny Chin in March of this year. There was still hope, however, that a settlement between the parties would find a way to implement a "Book Rights Registry" that, while falling short of eliminating the orphan works problem, could go a long way towards making rights holders easier to find. I was present at last week's status conference where the parties met with Judge Chin to report their progress towards a new settlement addressing Judge Chins concerns, or failing that, a timetable for renewed prosecution of the lawsuit.  Although the lawyers for the publishers seemed to be optimistic about a settlement, Michael Boni, the lead attorney for the "authors" talked as though the their part of the lawsuit would go to trial.
No orphans here, just books being deeply
discounted at a Borders going-out-of-business sale.
Note the blurb on the book at the bottom.

While there are important copyright issues at stake in the lawsuit, I've been most interested in the proposed book rights registry. A book rights registry would compile and maintain a database of rights holders, making it easy (and cheap!) to contact, query and pay rights holders for uses that are would not be allowed without their permission or provided for by fair use. In the current environment, it can cost hundreds of dollars to clear the usage rights for an in-copyright book, even without paying anything to rights holders. With a publisher-only settlement and continued prosecution now likely, my guess is that the book rights registry is either dead or years from being a reality.

It's ironic that the past week demonstrated how badly a book rights registry is needed. HathiTrust, the research library consortium formed to manage the scans of library books generated by the Google Books program, announced plans to expand access to a small number of works that they deemed to be orphans. The Authors' Guild then sued HathiTrust to block the expanded access.

The legal experts thought the suit "borders on the frivolous", due to serious problems with the "standing to sue" issue. Only copyright owners have standing to sue to enforce copyrights in the US, and none of the rights holders were parties to the Authors' Guild suit.

A funny thing happened on the way to the courtroom, though. Under intense scrutiny, one of the purported orphans turned out to have easy-to-find parents. The Authors Guild gleefully reported that the day before filing the suit, they were able to locate a rights holder "with a few minutes of googling". This report seemed to indicate incompetence all around, HathiTrust for neglecting the few minutes of googling in their orphan-identification workflow, and the Authors' Guild legal team, for failing to address their "standing to sue" problem with "a few minutes of googling".

A subsequent blog post from the Authors Guild was a revelation. It turns out that the Authors' Guild didn't really need a registry to find rights holders. By asking for help from their large number of blog readers, they were able to identify rights holders for many more of the purported orphan works. It seems that when large numbers of people are interested to find a rights holder, it's not so much of a problem. Of course a rights registry would help, but even the best registry can't fix everything.

I started following the Google Books case over two years ago because I thought it was important to increase access to all sorts of in-copyright works. What I learned in the process made me realize that a cookie-cutter, one-size-fits-all solution would never work, let alone pass judicial scrutiny. An equitable arrangement for academic authors would treat authors who write for a living unfairly; and vice versa. Authors in other countries would be ill-served by a process devised with American authors in mind. I realized that access to the works most important to book lovers would only happen with lots of reader support. And that realization has led to the work at Gluejar on Unglue.it, currently in the implementation stage.

Unglue.it will address "orphan works" the same way that the Authors' Guild has done in its recent blog posts. In addition to working with rights holders that want to offer creative commons licensing of ebooks to the public (ungluing them, in our parlance), we'll give users the opportunity to "wish" for the ungluing of any book that's been published. If a lot of users wish for a book, we'll check into who owns the rights, and give them a chance to make an offering. If we can't find the rights holder, we'll ask the people doing the wishing for help. If 10,000 people ( or even a dozen, for that matter) care about J.R. Salamanca's The Lost Country, they'll do a lot more than a few minutes of googling. They'll be knocking on his door and sending him postcards from Fiji.

When a lot of people care about a book, they'll have the combined economic power to do a lot more than opening a book to snippeting and search. We'll ask the rights holders for their price to give their book as an ebook to the world under a non-commercial creative commons license (CC BY-NC-ND). That will make it possible for everyone, everywhere, now and long into the future, to use the book the way the creator always intended- to read, to learn and to enjoy.

Sunday, September 11, 2011

The Smell of a Book

There's one part of the human brain that seems programmed to never forget things. It's somewhere in the limbic system, and it connects smells to emotions. This past week, deep in the bowels of New York Penn Station, that part or my brain was momentarily triggered by an acrid smell. Perhaps it was smoking train brakes or hot diesel oil, but it evoked a sad memory from ten years ago.

Smells connect us across decades, maybe across millennia. Some smells are hardwired to be pleasant or noxious, other smells are neutral and imprintable. Think of the smell of a new-born baby or the smell of your grandmother. Think of the smell of Starbucks, or of bread baking in the oven at home. Imagine being in a damp cave, or a medieval cathedral.

Scientists have studied this. It's now thought that the primal connection between smell and memory is a result of direct connections between our olfactory lobes and the hippocampus. Some scientists in Israel used functional MRI to see directly the involvement of the hippocampus in memories imprinted with strong smells. (Notes 1 and 2 and the picture.)

It's odd that so many people claim to love the smell of books. It's even stranger that people claim to love the smell of libraries or used bookstores. It's just old glue, ink, dust, mold, and decay. Odd, until you think about the time-travel aspects of smell.

In preparation for the upcoming launch of Unglue.it, I've been talking to a lot of people about the books that they love. "Love" in this context is not the "love" people might use casually to describe their relationship with a product for sale. Instead, people seem to relate to books the way they relate to people. There's the love for a teacher who makes a difference in your life. Love for a friend you helps you feel joy. The thrill of discovering a soul mate. And among authors, there's the blind love for a child that goes beyond all rationality.

The intensity of these emotions must get bound up with smells in the hippocampus to create a lasting impression on book lovers. When we smell a book all of these feelings resonate across time and they comfort us. Even in the future when all our reading is done on ebook readers or other screens, we'll keep real books around us like the clothing of a spouse or a parent lost to a tragedy, left in the bed to warm and comfort. And then we'll find strength to move on, but the spirit of the book will remain.


Notes:
  1. Jonah Lehrer's article on the Israeli fMRI study is very accessible.
  2. That study, "The Privileged Brain Representation of First Olfactory Associations" was written by Yaara Yeshurun, Hadas Lapid, Yadin Dudai and Noam Sobel in Current Biology 19(21), 1869-1874, (9 November 2009) and is available at http://www.cell.com/current-biology/abstract/S0960-9822(09)01857-0
  3. Another human sensation mediated by the hippocampus is laughter. I suffered repeated bouts of this affliction upon reading a website claiming to promote an aerosol spray. I was almost unable to finish this post.

Wednesday, September 7, 2011

To Honor Project Gutenberg's Founder, Dedicate Something to the Public Domain

The following is an obituary Of Michael Hart, founder of Project Gutenberg, who died yesterday. Its author has dedicated it to the public domain, which allows me to reprint it here:
Michael Stern Hart was born in Tacoma, Washington on March 8, 1947. He died on September 6, 2011 in his home in Urbana, Illinois, at the age of 64. His is survived by his mother, Alice, and brother, Bennett. Michael was an Eagle Scout (Urbana Troop 6 and Explorer Post 12), and served in the Army in Korea during the Vietnam era.
Hart was best known for his 1971 invention of electronic books, or eBooks. He founded Project Gutenberg, which is recognized as one of the earliest and longest-lasting online literary projects. He often told this story of how he had the idea for eBooks. He had been granted access to significant computing power at the University of Illinois at Urbana-Champaign. On July 4 1971, after being inspired by a free printed copy of the U.S. Declaration of Independence, he decided to type the text into a computer, and to transmit it to other users on the computer network. From this beginning, the digitization and distribution of literature was to be Hart's life's work, spanning over 40 years.

Hart was an ardent technologist and futurist. A lifetime tinkerer, he acquired hands-on expertise with the technologies of the day: radio, hi-fi stereo, video equipment, and of course computers. He constantly looked into the future, to anticipate technological advances. One of his favorite speculations was that someday, everyone would be able to have their own copy of the Project Gutenberg collection or whatever subset desired. This vision came true, thanks to the advent of large inexpensive computer disk drives, and to the ubiquity of portable mobile devices, such as cell phones.

Hart also predicted the enhancement of automatic translation, which would provide all of the world's literature in over a hundred languages. While this goal has not yet been reached, by the time of his death Project Gutenberg hosted eBooks in 60 different languages, and was frequently highlighted as one of the best Internet-based resources.

A lifetime intellectual, Hart was inspired by his parents, both professors at the University of Illinois, to seek truth and to question authority. One of his favorite recent quotes, credited to George Bernard Shaw, is characteristic of his approach to life:
 "Reasonable people adapt themselves to the world.  Unreasonablepeople attempt to adapt the world to themselves.  All progress,therefore, depends on unreasonable people."
Michael prided himself on being unreasonable, and only in the later years of life did he mellow sufficiently to occasionally refrain from debate. Yet, his passion for life, and all the things in it, never abated.

Frugal to a fault, Michael glided through life with many possessions and friends, but very few expenses. He used home remedies rather than seeing doctors. He fixed his own house and car. He built many computers, stereos, and other gear, often from discarded components.

Michael S. Hart left a major mark on the world. The invention of eBooks was not simply a technological innovation or precursor to the modern information environment. A more correct understanding is that eBooks are an efficient and effective way of unlimited free distribution of literature. Access to eBooks can thus provide opportunity for increased literacy. Literacy, the ideas contained in literature, creates opportunity.

In July 2011, Michael wrote these words, which summarize his goals and his lasting legacy: “One thing about eBooks that most people haven't thought much is that eBooks are the very first thing that we’re all able to have as much as we want other than air. Think about that for a moment and you realize we are in the right job.” He had this advice for those seeking to make literature available to all people, especially children:
 "Learning is its own reward.  Nothing I cansay is better than that."
Michael is remembered as a dear friend, who sacrificed personal luxury to fight for literacy, and for preservation of public domain rights and resources, towards the greater good.

This obituary is granted to the public domain by its author, Dr. Gregory B. Newby.

The best way to honor Hart's life, I think, is to dedicate one or more works to the public domain. I have just dedicated my Ph. D. Dissertation to the public domain: Hot Electron Resistance and Magnetoresistance in High Purity Gallium Arsenide. And this blog post, too, is hereby dedicated to the public domain. That means you can take it apart, copy pieces of it, remix it, make a Youtube video from it, perform it as a dance, even pretend that you wrote it, instead of me. And that would be OK with me and Michael Hart. Go crazy.

Update (9/9): I've been reminded that the easiest way to do a public domain dedication is to upload the work to Internet Archive. As I did.