Friday, September 11, 2015

Library of Cards


On Thursday, September 10th, I had the honor and the pleasure to present at Access 2015. I've been to many Access Conferences over the years and each one has been a joyful experience. Thank you so much to those organizing Access YYZ for all of us.

Have you ever listened to the podcast 99% Invisible?

99% Invisible is a weekly podcast dedicated to design and architecture and the 99% percent of the invisible activity that shapes our experiences in the world.

They’ve done episodes on the design of city flags, barbed wire, lawn enforcement, baseball mascot costumes, and it’s already influenced my life in pretty odd ways.

If I was able to pitch a library-related design to host, Roman Mars for an episode, I would suggest the history of the humble 3 x 5 index card.

That being said, for this presentation, I am not going to present to you the history of the humble 3 x 5 index card.

That’s what this book is for, Markus Krajewski’s 2011 Paper Machines published by MIT Press.

Now, before I had read his book, I had believed that the index card was invented by Melvil Dewey and commercialized by his company, The Library Bureau. But Krajewski makes the case that the origin of the index card should be considered to go as far back as 1548 with Konrad Gessner who described a new method of processing data: to cut up a sheet of handwritten notes into slips of paper, with one fact or topic per slip, and arrange as desired.

According to Krajewski, when this technique goes from provisional to permanent – when the slips that describe the contents of a library are fixed in a book, an unintended and yet consequential turn takes place: it gives rise to the first card catalog in library history in Vienna around 1780.

Most histories of the card catalog begin just slightly later in time -- in 1789 to be precise -- during the French Revolution. The situation at hand was that the French revolutionary government had just claimed ownership of all Church property, including its substantial library holdings. It order to better understand what it now owned, the French revolutionaries started to inventory all of these newly acquired books. The instructions for how this inventory would conducted is known as the French Cataloging Code of 1791.

The code instructed that first, all the books were to be numbered. Next, the number of each book as well as the bibliographic information of each work were to be written on the back of two playing cards - and this was possible because at that time the backs of playing cards were blank. The two sets of cards are then put into alphabetical order and fastened together. One set of cards were to be sent to Paris, while a copy remains in each library.

On the screen behind me, you can see two records for the same book.

Again, my talk isn’t about bibliographic history, but I want to return back to the 16th century to Gessner for some important context. The reason why Gessner was making all those slips in the first place was to construct this, the Bibliotheca Universalis which consists of a bibliography of around 3,000 authors in alphabetical order, describing over 10,000 texts in terms of content and form, and offering textual excerpts. As such, Gessner is considered the first modern bibliographer.

And you can find his work on the Internet Archive.

Gessner’s Biblioteca Universalis wasn’t just a bibliography. According to Krajewski, the book provides instructions to scholars how to properly organize their studies through the keeping excerpted material in useful order. Gessner was describing an already established practice. Scholars kept slips or cards in boxes, and when they had the need to write or give a lecture or sermon, they would take the cards that fit their theme, and would arrange those thoughts and would temporarily fix them in order using devices such as the one pictured. This hybrid book has guiding threads that stretch over the page so that two rows of paper slips can be inserted and supported by paper rails.

Until the Romantics came around and made everyone feel embarrassed about taking inspiration from other people, it was not uncommon for scholars to use Scholar’s Boxes. Gottfried Leibniz actively used what was known as an excerpt cabinet to store and organize quotations and references.

Leibniz's method of the scholar's box combines a classification system with a permanent storage facility, the cabinet. So in a way this is similar to the use of Zotero or other citation management systems, but instead uses loose sheets of paper on hooks. The strips are hung on poles or placed into hybrid books

And that’s the reason why I wanted to start my talk with a brief history lesson. To remind us that there is a common ancestor to the library catalog and the scholar’s bibliography, and that is the index card.

So as we’ve learned, from as far back as Gessner’s 16th Century, writers have been using cards and slips of paper to rearrange ideas and quotations into texts, citations into bibliographies, and bibliographic descriptions into card catalogues.

You can still buy index cards and card boxes at my local campus bookstore. That’s because there are still authors today, who still use index cards to piece together and re-sort parts of their paper or novel, or they use and rearrange digital cards inside of such writing software tools such as Scrviner to generate new works.

Now, I don’t write this way myself. But I do use Zotero as one of the tools that I use to keep track of citations, book marks,  saved quotations, and excerpts of text that I have used or might use in my own work as a writer and academic librarian.

Zotero acts as an extension of your web reading experience and it operates best as an add-on to the Firefox browser. If you use Zotero, you can usually easily capture citations that one finds on a page either because someone who supports Zotero has already developed a custom text scraper (called a translator) for the database or website that you are looking at or that citation has been marked up with text that’s invisible to the human eye but can be found in the span HTML tags that surround the citation using a microformat called COinS.

Zotero also allows scholars to backup their citations to their server and in doing so, share their citations by making one’s library public on  Alternatively, scholars can share  bibliographies on their own website using the Zotero API which is so simple and powerful you can embed a bibliography styled with APA with a single line of code.

One of my favourite features of Zotero is not widely known. Zotero out of the box allows the scholar to generate ‘cards’ which are called ‘reports’ from your bibliography.  When I have a stack of books that I need to locate in my library, I sometimes find it’s easier for me to select and generate a report of cards from my Zotero collection rather than to search, select and print the items using my library’s expensive ILS system.

There is a terrible irony to this situation. As I learned from the Library Journal column of Dorothea Salo, the design problem given to Henriette Avram’s, the inventor of the MARC records was to have “computers print catalog cards.”

As Salo says in her piece, “Avram was not asked to design a computer-optimized data structure for information about library materials, so, naturally enough, that is not what MARC is at heart. Avram was asked solely to make computers print a record intended purely for human consumption according to the best card-construction practices of the 1960s.”

Let’s recall that one of the reasons why Zotero is able to import citations easily is because of the invisible text of COinS and translators. 

The metadata that comes into Zotero is captured as strings of text. Which is great - a name is now tagged with the code AU to designate that the text should go in the Author field. But this functionality is not enough if you want to produce linked data.

Dan Scott has kindly shared the code to RIS2WEB that allows you to run it on an export of a bibliography from Zotero in doing so create and serve a citation database that also generates of linked data using Schema. Afterwards, you can add available URIs.

You can see the results of this work at

When I showed this to a co-worker of mine, she couldn’t understand why I was so impressed by this. I had to hit Control-U on a citation to show her that this citation database contained identifiers such as from VIAF: The Virtual International Authority File. I explained to her that by using these numeric identifiers - invisible to the human eye - computers will be able not only find matching text in the author field, they will be better able to find that one particular author.

So can we call Zotero a Scholar’s Box or Paper Machine for the digital age?

I think we can, but that being said, I think we need to recognize that the citations that we have are still stuck in a box, in still so many ways.

We can’t grab citations from library database and drop them into a word processor without using bibliographic manager like Zotero as an intermediary to the capture structured data that might be useful to my computer when I need format a bibliography. Likewise, I can’t easy grab linked data from sites like the Labour Studies bibliography page.

And we still really don’t share citations in emails or social media.

Instead, we share the URL web addresses that point to the publisher or third party server that host said paper.  Or we share PDFs that should contain all the elements needed to construct a citation and yet somehow still requires the manual re-keying and control c-ing and v-ing of data into fields when we want to do such necessary things as add an article to an Institutional Repository.

Common Web tools and techniques cannot easily manipulate library resources. While photo sharing, link logging, and Web logging sites make it easy to use and reuse content, barriers still exist that limit the reuse of library resources within new Web services. To support the reuse of library information in Web 2.0-style services, we need to allow many types of applications to connect with our information resources more easily. One such connection is a universal method to copy any resource of interest. Because the copy-and-paste paradigm resonates with both users and Web developers, it makes sense that users should be able to copy items they see online and paste them into desktop applications or other Web applications. Recent developments proposed in weblogs and discussed at technical conferences suggest exactly this: extending the 'clipboard' copy-and-paste paradigm onto the Web. To fit this new, extended paradigm, we need to provide a uniform, simple method for copying rich digital objects out of any Web application.

Now, those aren’t my words. That’s from this paper Introducing unAPI written by Daniel Chudnov, Peter Binkley, Jeremy Frumkin, Michael J. Giarlo, Mike Rylander, Ross Singer and Ed Summers.

This paper, I should stress, was written in 2006.

Within the paper, the authors outline the many reasons why cutting and pasting data is so infuriatingly difficult in our sphere of tools and data.

But what if there was another paradigm we could try?

In order to see how we might be possibly break out of the scholar’s box, I’m going to talk about a very speculative possibility. And in order to set us up for this possibility, I first need to talk about how cards are already used on the web and on our tablets and smart phones.

If you look around the most popular websites and pay particular attention to the design patterns used, you will quickly notice that many of the sites that we visit every day (Twitter, Facebook, Trello, Instagram, Pinterest) they all use cards as a user interface design pattern.

The use of cards as a design pattern rose up along with the use of mobile devices largely because a single card fits nicely on a mobile screen...

...while on larger surfaces, such as tablets and desktops, cards can be arranged in a vertical feed, like in Facebook or Twitter, or arranged as a board like Pinterest, or like a like a stack, such as Google Now or Trello.

This slide is from a slidedeck of designer and technologist, Chris Tse. The rest of this section is largely an exploration of Chris’ work and ideas about cards.

Case in point, Chris Tse states, the most important quality of ‘cards’ is that of movement. But by movement, he isn’t referring to the design’s apparent affordances that makes swiping or scrolling intuitive.

The movement of cards that’s really important is how they feed into content creation and content sharing and how cards feed into discussions and workflow.

(How cards fit into kaban boards and shared workflow software like Trello, is a whooooole other presentation)

Social media is collectively made up of individuals sharing objects - objects of text, of photos, of video, of slideshows - and they share these objects with friends and to a larger public. Each of these objects are framed - by and large - within cards.

It’s important to realize that the cards on the web are fundamentally more than just a just a design hack.  If you are familiar with Twitter, you may have started to see cards that don’t just feature 140 characters - you see playable music (such as from Soundcloud), slideshows that you can read through without leaving Twitter, and you can even download 3rd party apps from Twitter advertising cards. When the business press say that Twitter is a platform, it’s not just marketing hype.

As Chris Tse says, cards are more than just glorified widgets.  “When done right”, he says, “a card looks like responsive web content, works like a focused mobile app, and feels like a saved file that you can share and reuse". As “cards” become more interactive, he believe they will go from being just concentrated bits of content and turn into mini-apps that can be embedded, can capture and manipulate data, or even process transactions.
But why isn’t this more obvious to people? I think the reason why is that cards don’t really feel this way is that most cards can only move within their own self-contained apps or websites.

For example, Google Now cards work with your Google applications – such as your calendar - but doesn't know about the events that you've RSVPed in Facebook.

That being said, Google and Apple are working on ways into integrate more services into their services. In Google Now, I’m regularly offered news story based on my recent searches as well as stories that are popular to other readers who read similar stories using the Feedly RSS reader.

And this is a problem because it’s Google who is deciding whose services I can choose from for such card notifications.

The apps on your smart phone live in a walled garden where things are more beautiful and more cultivated, but it is a place that is cut off from the open web.

The fall of the open web and the rise of the walled garden is not a trivial problem. We should not forget that if you want your app to be available on an iPhone it must be in the Apple Store and the content of your app will be subject to the Apple Review process and Apple will take a 30% cut of what your app sells for.  Content within apps curtail various forms of free and freedoms.

To bridge this split of the open web and the walled app garden, Chris Tse founded The mission of Cardstack is to “To build a card ecosystem based on open web technologies and open source ethos that fights back against lock-in.”

CardStack wraps single-page JavaScript applications as a reusable ‘card’ that can be embedded in native apps and other web apps. According to Chris, HTML5 cards will be able to move between apps, between devices, between users and between services.

CardStack itself is comprised of other JavaScript libraries, most notably Conductor.js and Oasis.js and I cannot speak anything more to this other than to repeat the claim that these combined libraries create a solution that is more secure than the embedded content than the iFrames of widgets past.

But notice the ‘Coming Soon’ statement in the top left hand corner? CardStack is still in beta with SDKs still being developed for iOS and Android systems.

Despite this, when I first stumbled upon Chris Tse’s presentations about Cardstack, I was really excited by his vision. But the crushing reality of the situation settled mere moments later.

Yes, a new system of cards to allow movement between siloed systems that could work in mobile or desktop environments, that would be wonderful - but wasn’t it all too late?

And what does it mean if we all collectively decide, that it is all just too late?

One of the challenges of promoting a new sort of container is that you can really show it off until you have some content to contain. You need a proof of concept.

When I checked in later to see what Chris was doing as I was drafting this presentation, I learned that he was now the CTO of a new platform - that he confirmed for me is built using Cardstack as a component.

This new platform has its origin from the art world’s Rhizome’s Seven on Seven conference. The Seven on Seven conference pairs seven visual artists with seven technologists for a 24 hour hackjam and in 2014, artist Kevin McCoy was paired up with technologist Anil Dash.

McCoy and Dash were both interested in improving the situation of digital artists whose work can be easily be copied. With copying, the provenance of a digital work can be lost and as well as the understanding of what was original and what has been modified.

They talked and worked on a proof of concept of a new service that would allow visual artists to register ownership of their digital work and transfer that ownership using blockchain technology - that's the technology behind bitcoin.

Over a year later, this idea has grown into a real platform that is private beta and is set to be released to the public this month.

I think two things are particularly awesome about this project. First the platform also allows for artists to decide for themselves whether the license for their work in the creative commons or requires a royalty and whether derivatives of their work is allowed.

The other thing I love about this project is its name. If you look at the top left hand of this screen you will find that the name of the platform is spelled m-o n-e-g-r-a-p-h. The platform is called MONOGRAPH.

And as we now know, you build a monegraph with cards.

We need to remember that the library and the bibliography are connected by the card.

As we continue to invest in crucial new endeavors in the digital realm, I think it's essential that librarians find new ways to surface our resources and allow them to be shared socially and to find the means by which scholars can save and sorting and re-use these resources that they find from our collections.

We are part of a generative process. Cards of single ideas that are arranged and stacked build theses, which in turn, build papers, books which, in turn, form bibliographies which fill libraries.

I would like libraries to find a way to back to Gessner’s Bibliotheca Universalis, a place where the library and the scholar were both connected.

After all...