Cathy Marshall, Microsoft Research, Silicon Valley – Reading and Collaboration in a Digital Age

From Cyborg Anthropology
Jump to: navigation, search

On July 8th, the Computer-Human Interaction Forum of Oregon (CHIFOO) hosted Cathy Marshall of Microsoft Research at Jive Software (CHIFOO’s new location). Marshall’s presentation, titled Reading and Collaboration in a Digital Age: or How I learned to Stop Worrying and Love the Screen, was a mental tour de force that reexamined assumptions of how we read, annotate, and look at text.

Approximately 60 people were in attendance, and the audience and speaker discussion was lively and relevant. There was never a dull moment or boring segment. I sat there furiously trying to capture every piece, as you will see evidenced below.

A Short History of eBooks

Marshall: I know lots of you are thinking, “what does reading have to do with collaboration?”.

Generation 1

eBooks have really been around for a long time, since around the 1980’s. The first generation was really about hypermedia and multimedia. Kind of the excitement of having these things on the screen, to be able to do things that you couldn’t do before. Peruses was a site about ancient Greece — the reason people loved it was that you were able to look up words in Greek and have them available immediately.

Generation 2

Generation 2 had P-books, or portable books. This turned out to be a bad name. There were multiple jokes about it. There was even a Zippy comic that made fun of it.

The comic shows Zippy and his friend flying through the city on the back of a book. Zippy’s friend says, “I head that the E-book trend never really took off, sales of the things are tanking.” and zippy says, E-books will never replace P-Book!”.

There’s some more text discussing the comparative values of books over electronic media, and the cartoon ends with Zippy saying, “E-books are spineless”.

Marshall: I think there’s a real sort of cultural anxiety about the end of books, and the death of text. And there was also skepticism about reading on computers, Like Sven Birkerts, Richard Harper, who wrote about how paperless offices didn’t work. There were also people in library science who said that these things wouldn’t work out well eighteen.

Marshall brings out a slide of an old cell phone displaying a partial sentence from Moby Dick on its tiny, pixilated screen.

Marshall: For many people, their worst fear was of having to read something on a cell phone while being trapped in the airport.

But there is no reason to laugh about this anymore because people in Japan are actually reading and writing novels on cell phones.

In Family Circus…by the way….does anyone think Family Circus is funny? I think they must have some hidden message or something , and that’s why people keep publishing them.

Audience: I have some friends who carefully cut out Family Circus every day…and then replace the captions with something else. Then they’re funny. A Family Circus comic shows up on the screen. The kid is talking to his mother. “I’m never going to start reading eBooks,” he says, “it’s too hard to curl up with a monitor”.

And one last point was from Clifford Lynch in the battle to define the future of the book in the digital world. He said, “Try to think of eBooks as personal libraries instead of books” First Monday, 2001. “>First Monday 2001.

Generation 3 – 2006-2008

By the time Generation 3 happened, the generations were getting closer and closer together (as they say in future shock). In this generation, we asked ourselves, will eBooks somehow renew the social side of reading?

Why was it so hard to see what’s coming?

Reason 1: Changes aren’t always in technology

There was a very famous article written by Vannevar Bush about a system he called a Memex (portmanteau of “memory extender”). It’s heralded as the introduction to the hyperlink, that you could go from one place to another and record that hyperlink.

“The advanced arithmetic machines of the future…will have enormous appetites. One of them will take instructions form a whole roomful of girls armed with simple keyboard punches and will deliver sheets of complicated results every few minutes”. – Vannevar Bush in As We May Think, 1945.

I took typing class too, on those big clunky computers. And there were no boys in the class. You weren’t a boy in my class unless you were in drag.

An audience member nods. “Were you in drag?” Marshall asks. “Depends,” he responds, “what year was that again?”

Why is it hard to answer this question?

Answer: Because it is often difficult to see the whole cost/benefit analysis side of the picture, like this panel I cut out from the back of a box of Shredded Wheat that says, “Dear NABISCO Shredded Wheat Users”.

Reason 3: Reading is Invisible

“Nothing is more commonplace than the reading experience, and yet nothing is more unknown. Reading is such a matter so common that at first glance, it seems there is nothing to say about it. ”

Tzvetan Todorov, quoted by Nicholas Howe in The Ethnography of Reading. Marshall: I’m kind of a feral Ethnographer. Sarah has worked with me and knows that I like to have principles.

I was sitting there on the airplane and I was sitting there watching this man read his magazine. There he was, reading this magazine. I thought I was so discreet. And at some point he got up and went to the restroom. And he looked over at me and said, “you stole my magazine”. and I said, “I did not!” and he said, “Let me look in your briefcase”. And so reading is invisible. And it’s very dangerous to watch people read. And people think it’s creepy!

But in this talk I’m trying to summarize 15 years of studies on cooperation, and reading tech, to really find out what reading is. So you’ll have to bear with me as I tease out a definition.

I starting looking at intelligence analysts – how people gathered and collected things, and then how people annotated things, and found that they aren’t quite the scholarly things people see in the margins, and then looked at it in law offices and law school. Those also who came in and talked to the Vice President and President and briefed them every morning. And I actually got to be there when President Bush got the Osama bin Laden briefing.

I went to work at Microsoft and looked a Microsoft reader, and then I looked at shared annotations, and then how people clipped things out of magazines and how they read. So we looked at reading in some detail. Then I worked with some people t Microsoft at the New York Times Reader application. Does anyone have one of those?

One audience member raised his hand.

Well then, it was a tremendous success! The photos in it are really nice. You don’t really notice how nice the photos are in the Times until you view them in that reader.

Then she showed a photos of a guy sitting on the subway reading a newspaper seated next to a guy who was sitting there with a tremendous cathode ray tube monitor and keyboard on his lap, the computer unit on the ground underneath his feet. It was making fun of Reading, of course.

How Do Most People Think About Reading?

We think it’s private, individual, stationary and passive. We think it’s something as immersive, and sometimes soggy (she shows a picture of a guy reading a newspaper in the bathtub).

But what we found instead was that reading is mobile. That’s why reading on a screen was so dismal at first, because nobody wanted want to carry around a screen with them everywhere. Because reading was so mobile. What we found at first was that mobility overwhelmed many things at first.

“If I’m going home to Colorado, I have to be really sure I’m going to read something if I’m going to bring it. Otherwise, why should I bring it [if it's large, heavy]. [The Pocket PC] is small, it’s handy.”. Quote from a college student talking about a Pocket PC with his course texts.

Marshall: Note that he actually didn’t end up reading his coursework on over the break.

So reading is mobile, material, passive.

In The Places of Books in the Age of Electronic Reproduction, Geoffrey Nunberg of Xerox Palo Alto Research Center and Stanford University said this about eBooks:

“Reading what people have had to say about the future of knowledge in an electronic world, you sometimes have the picture of somebody holding all the books in the library by their spines and shaking them until the sentences fall out loose in space” (Representations 24, Spring, 1993). Also in Howard Bloch and Carla Hesse, eds., Future Libraries, University of California Press, 1994.

“You get this little screen, so you get no sense of even how long the work is…but you have 600 pages, which means what? No one knows. So I definitely don’t see it as a literary experience”. An English Lit Grad student talking about reading on the Jordana Pocket PC.

(Note from Amber Case: This is what I continually think about when I encounter a computer, because no matter how much data I stuff into it, it never gets heavier. A book weighs the same as a leaflet – nothing). Marshall: Navigation is fundamental to the material of paper.

“Something else that I think I sometimes do when reading an article: I’ll be like, ‘boy this has been going on a long time, and sometimes I’ll even flip ahead and think, how many more pages do I have? And if it’s going to end on this page, then I may just read it. But if I see it’s three more pages, the…I may just either give up. Or just go into scan mode, where I just flip, you know, see what grabs my attention” Marshall: Reading has a basic physicality.

(Note from Amber Case: Here, the materiality allows scanning, weight, and thickness).

“I usually read in one of the chairs in the living room. That’s partly because I don’t have a desk in here. The chairs are very comfortable. There’s a occasionally much too comfortable, that’s why I have blankets around every chair in the house, so I can always be prepared to go to sleep.” – An English Lit major talking about where she reads.

Then Marshall shows a quote from the French philosopher Jean Baudrillard. “I can’t read this without a French accent,” she says, “else I can’t get away with it. Does anyone have a French accent?”.

No one in the a audience had one.

“The compact disc,” says Baudrilliard, “It doesn’t wear out, even if you use it. Terrifying. It’s as through you’d never used it. So it’s as through you iddn’t exist. IF things dont’ get old anymore, then that’s because it’s you who are dead”. Jean Baudrilliard, Cool Memories II.

Marshall: Maybe you don’t want the pristine copy – you want the one that is like the one you first bought in the 70’s. The one that is used. The one that is well read.

You think about how interact with books online – you don’t have to think about that with a paper book. You don’t have to think about how to annotate.

Audience: The medium of the book is to have it be as transparent as possible. But when you have these different mediums that have types of media placed, you can’t read them anymore. You’re inhibited by the medium. You notice it.

We’ll get back to that later – I have a big rant about that too.

People interact with text far more than they own up to. People don’t remember making the annotations, they idealize them, they make far more than they actually remember. And when you show someone their annotations from a few days back, they don’t know what many of the annotations were referring to.

Audience Member: Have you ever heard of the book as a sacred object?

Because I’m a librarian and I can’t annotate a book. I buy one copy for me and another to annotate.

Marshall: And what about the Ebook? Do you value the Ebook?

Audience Member: There’s nothing sacred about an Ebook because it doesn’t have a material embodiment. And I know I’m not going to pass it along to anyone else.

Marshall: Not unless you violate the DRM you won’t!

Audience Member: Is that sacredness of the book genetic, do you think?

Audience Member: Well I don’t know.

Librarian: Well, I was one of those, “Matchbox car collectors, a ‘never open the package’ kind of person.

Audience: What about the notes taken by college students?

Marshall shows the image of a page that’s been completely highlighted.

Like this? Or some people carefully save all of their college notes and them look at them later, or think they will look at them later. Or value them highly, but never look at them.

Literally, though, this highlighting goes on for pages. If you find that at the beginning of a math book, it means that the person’s going to drop the class.

Audience: I could never buy a book that was already annotated, because I’d go through the book and be like, “that’s not worthy of being annotated! or that section is not important enough to be highlighted!”.

Audience: Can you tell me the context of this study? How it was formed? Where you got the information?

Marshall: I’m smushing together many years of research here, but I can tell you about a few experiments.

For instance, for the highlighting, annotation one, I staged myself in the Stanford bookstore and pretended that I worked there, and I stayed there 2-3 weeks, looking through 1000’s of textbooks, watching people buy used and new textbooks, eavesdropping on whether or not they would buy what kind of book, and interviewed them about f they would by And a lot of them would look through books to see what had been outlined before they decided on purchasing them.

This was a study I did a dozen or so years ago. It was one of the first studies I did, and it was just to get an idea of what people did when they purchased textbooks.

Audience: Did you ever find out the answer, “why did you highlight this entire text? Like why so much?

Marshall: Well, I think it happens in instances where there’s really complex information placed in front of someone who doesn’t understand it. The highlighting becomes more of a tracing of general attention. Sometimes it is from multiple readings.

In Mortimer Adler and Charles Van Doren’s How to Read a Book, there’s a whole section on active learning. Sometimes I see those results. One time I saw a book with multiple different colors and I found a student who said, “Oh, I do that!”. I asked why, and she said, “Oh, I just change colors when I get bored”. Evidence of why it is important to ask.

Annotations May Quickly Lose Their Value or Be Forgotten

“Some of them are absolutely ridiculous and I can’t believe that I actually wrote this in pen in the book. Some of them are – I have no idea what I’m talking about. Some of them are really interesting, and it’s something I’ve forgotten. It just depends on the notes….when I did Milton, we were doing the epithets about Satan or something, so I underlined all of them. And when I was going back through it, I’m like “what on Earth!?” A grad student talks about annotations she made as an undergrad.

Marshall: The reason I found out about the subconscious stuff is that I’d go back with them through their notes a week after they’d done it and ask them about it, the notes, the diagrams, and some of them would say, “I’m sure it had some meaning at the time”. So annotations have more meaning than we think.

Reading is Interrupted and Variable

I think this is at the root of “what is reading”. It’s not this image of a little girl in the window seat and she’s totally engrossed in a book, uninterrupted.

“We do not read everything with the same intensity of reading; a rhythm is established, casual, unconcerned with the integrity of the text; our very avidity for knowledge impels us to skim or to skip certain passages (anticipated as ‘boring’) in order to get more quickly to the warmer parts of the anecdote…” – Barthes, The Pleasure of the Text.

Marshall: Reading is not a single, undistracted stream of concentration. Has anyone read all the words of Proust, or War and Peace?

Audience member: Yes. But it was not normal circumstances.

Marshall: Right, most of the time, reading is fragmented.

Turning a Page as a Complex of Lightweight Navigational Acts

A series of actions: Constance is reading the first page of a review, but halfway through the article she turns the page halfway over, so she can see the next article while still reading the first one.

She looks at the cartoon before she goes to the next page because she thinks it’s funny.

She goes through the next page, which looks like a lengthy review, looks at the ads, because the likes to look at the ads.

She successfully flips over the magazine so that she can read the next article.

She changes the orientation of her hands so that she can comfortably read again.

I have so many videos of people moving their hands to their face or moving them when they’re reading.

I’m going to claim that reading is social. Not that it is intensely individual, as many people may think.

“It is also worth noting that solitary reading always was, and still is, inherently social: how we read is ultimately determined by social convention and community membership”. -David Levy, Scrolling Forward: Making Sense of Documents in the Digital Age

Marshall: Now I’m going to bring up our old friend, the CSCW matrix.

When, Where, same time, different time, Same place, different place. It’s been around so long that I couldn’t figure out how to source it. Audience: Stands for Computer Supported Cooperative Work. In the upper left: reading together, same time, same place.

We were watching students read on the web and we seated each one in front of a computer. We just told them to ‘go and browse the web’. And very quickly they had organized themselves in twos or threes around the computers instead of individually exploring the web alone.

And then we did studies with an early Web TV, and I thought, “ ‘ho hum!’ big deal, the Web on your TV!”

But then I watched as one kid was messing around with the Web TV, and another kid joined him. Before long, they were negotiating about where to go next on the web.

And then there were situations designed to read socially, like reading groups.

One of the things I noticed is how people stayed together while reading together. One of the problems with some books is that people go to the used bookstore and buy different editions, and people all have to align in class on the same class. They’re all different ways people use to get to the same page. Chapters, indexes, page numbers, ect. What we noticed is that people can be productively engaged in the discussion but not actually on the same page. This sort of things people would get punished for.

Audience: Was it established why it was important to be on the same page? Reading together: on-the-spot research enhancing discussion or digression?

Marshall: Well, we did some studies where there would be a line in the reading like “Did they really hang dogs a witches?” This was an interesting quote so all the kids reading on their pocket PC’s began to look it up.

Some teachers found it to be good, and others a distraction.

But a problem with sharing reading materials occurs when one tries to share them electronically, especially with a Kindle.

Audience: You can share books on a Kindle!

Marshall: Even DRM ones?

Audience: You can share them if they’re in the public domain.

But that’s not the same as sharing a book. The problem is that you have to have an ID or account to share that data. You can’t just pass it to the next person, like you would with an analog book. You can’t share the data itself, or annotations, or things you’ve torn out.

Speaking of tearing out data; we all have experienced this. Tearing out data makes us this of our mothers, our mothers or brothers or sisters, tearing something out and mailing it to us.

A Few Questions About Sharing Encountered Information

How important/ubiquitous is the information? Do people cut out things to annoy people?

It’s kind of like, you buy a magazine because of the things you might find in there. But you don’t know what’s going to be in there.

Audience: I now look at people’s Twitter feeds to see what I should look at.

At this point, @brampitoyo said (on Twitter) “@caseorganic Twitter is made for sharing artifacts encountered everywhere else. RT is one of the forms.”

Marshall: What are some of the reasons people share?

  • Sharing for mutual awareness.
  • At work, in customer-focused jobs.
  • At home, keeping up with friends and family

short of a way to keep in synch.

  • Sharing to educate or raise consciousness. Valued by sender — perhaps not by receiver.

Mostly occurred for personal topics/home

Audience: I was thinking with Twitter how funny it is, how the more boring Twitter users just send out links, and we don’t get to know them as person.

Audience: Well, I like those people!

Sharing to strengthen social ties

  • “I’m thinking of you”
  • “We have common concerns”
  • “We have the same sense of humor”.

Audience: Or sometimes you’re sharing to make people think you’re smart

Yes, we just notice it because it’s so obnoxious, but it’s rally not that prevalent. Just sharing knowledge to show off.

Audience: Or sharing to “hint”, like “I’m thinking about getting a camera”.

P2, a high school student, receives links to online article from her dad sometimes as often as 2 or 3 times a day. She usually reads he article son the screen and doesn’t keep them. For example her dad recently sent her an article from the NY Times. “Sending these articles is nice. I don’t know how we started doing it, but it feels nice to know people are thinking about you. It’s our way of keeping in touch.

Marshall: Here’s an example of sharing to educate.

P15 has a pre-adolescent son has been diagnosed with Asperger’s Syndrome, a form of high functioning autism. The mother found an article on educating children with Asperger’s syndrome and photocopied the “really good” article from Time. Then she told the her son’s teacher that she should read it. The Social Role of Sharing: Myth Busting

All four participants in our study shared information. None of them dominated in sharing the inormation, and none of them were the single sharers of information. This busts the idea of people setting themselves up as “information brokers’ not many people just send out completely, or one-way. Everyone sends out a few links. Audience: There are some people on Twitter who Retweet. I don’t really like that. Audience: Tell them! Marshall: I’m worried about you and Twitter. We should talk later. Audience: I work alone, so it’s my water cooler that I check every few hours.

Marshall: Still, I think you’re spending too much time on it.

It’s more complicated than that!

Riox looked at why people share or don’t share data.

  • Do I have the recipients email address at hand?
  • What will it look like?
  • Will this seem impersonal?
  • Will the Email look like spam? (Riox, 2000).
  • Form is important.

A technological solution for sharing should:

  • Present a sense of layout and article boundaries.
  • Allow the sender to limit or expand scope or context (compare sending a photo plus text vs. part of text).

Modes of Sharing are Important

“My plan is to actually give a hardcopy of an article from nature to him and talk to him about it, rather than just put it in his inbox because he’d kind of wonder where it came from or why he was getting it. And I’d rather say, hey, I saw this online and it’s pretty interesting. Check it out”.

Because he wants to get this higher into another person’s attention instead of the low attention the recipient might give the article should he receive it through a digital source.

“I have come to view margins as a literary commons with grazing room from everyone – the more, the merrier”. – Anne Fadiman, Ex Libris : Confessions of a Common Reader, London : Penguin Books, 1998.

Of course, sharing annotations is more complicated than it looks. See, for example, Shipman et al., ECDL 2003.

I was working at Microsoft Research and a guy on my team said, “wouldn’t it be cool if the annotations you wrote would be sent to the author of the book?” and I said, “No! I’d be dead!”.

But, I thought, is there a way to take multiple highlighting, annotations of multiple copies of the same book and see commonalities between them, in order to deduct the most useful pieces of text — a sort of wisdom of crowds sort of boil-down?

Annotations in the Aggregate

Consensus is significantly more common than predicted by strict probabilistic calculations of overlap.

Annotators converge on important text that is different than the text that the authors and publisher designate as important.

Annotation; collective effects. If you had dozens and dozens of books, could you use a ‘wisdom of crowds approach to zoom in on something that was important? Something that many different people underlined across all of the books? Some essential passage?

Audience: The Folksonomy of Cliffnotes? Is that what you’re getting at?

Marshall: Maybe… Kind of.

Audience: Or like a Wiki?

  • Collaboration and reading technologies; What of displays – are we thinking enough about “looking on” or shared focus?
  • How do social expectations interact with restrictions introduced by Digital Rights Management?
  • Which collaboration architectures will work for people using the same collections (i.e…annotation, reading rooms, bookmark servers)?

Are there new modes of collaboration enabled by digital devices? Collaborators: XLibris studies: Morgan price, Bill Schilit, and Gene Golovchinsky at FXPAL.

About Cathy Marshall

Cathy Marshall is a Senior Researcher at Microsoft Research, Silicon Valley; she has knocked around in both the product and research divisions at Microsoft. Cathy has long worked in the disciplinary interstices of computer science, information science, and the humanities, with occasional collaborations in the arts and the sciences. She was a long-time member of the research staff at Xerox PARC and is an affiliate of the Center for the Study of Digital Libraries at Texas A&M University. Cathy won the ACM Hypertext conference’s best paper award in 1998 and 1999, and the best paper award at the IEEE/ACM Joint Conference on Digital Libraries in 1998 and 2008. She has delivered keynotes at WWW, Hypertext, Usenix FAST, CNI, VALA, ACH-ALLC, and a variety of other CS and LIS venues. MS Reader study: Contact info:

cathymr [at] Microsoft [dot] com.

Source: [1]