Sunday, July 31, 2011

Flipboard & personalized reading magazines for academic research

In 2009, I wrote Aggregating sources for academic research in a web 2.0 world . In it, I wrote

"about the different sources one could add to stay on top of one's research area. These include RSS feeds from traditional databases (citation alerts, table of contents of favourite journals), library opac feeds of searches and new additions, book vendor sites (e.g Amazon) book sharing sites (e.g LibraryThing), social bookmarking sites both generic (e.g. Delicious) and research 2.0 sites (e.g. citeulike), Google alerts and more"

In particular I also mentioned 3 main types of RSS aggregators to put everything together basically "traditional" (e.g. Google reader), Startup pages (e.g. Netvibes) and Lifestream (e.g Friendfeed).

Since then, the last seem to have declined. I also wrote this

"A sub class of these aggregators allow you to "build" your own newspaper from RSS feeds, essentially these are just RSS feed readers but with more innovative layouts that mimick newspapers

Examples include FeedJournal, Feed Chronicle"

Two years have passed since I wrote this and since then the following trends have intensified

1. Smartphone usage and in particular tablet usage have exploded
2. Facebook and Twitter (the later was considered to be strictly early adopter in 2009) is now almost used routinely by organizations to stream information.
In particular seems I have underestimated this "Sub class" of readers that visualize results in newspapers like format. The rise of tablets & smartphones with the corresponding dominance of touch based UI has driven an interest in alternative visualizations of content obtained from RSS, Twitter, Facebook and social media streams. The poster child for this trend is the iPad app Flipboard. There's a very good reason why, it's the app that almost every iPad user has.

I only realized the power of Flipboard when I started using it to go through Twitter and Facebook shared content. It's something that is not easy to understand if you have not used it regularly.

The speed at which I processed material far exceeded that of going through Twitter or Facebook the traditional way for two reasons

1. I could go through several at one go, thanks to the magazine like layout
2. Precaching of content, speeds up reading

Precaching of content was important, since in a typical Twitter client (e.g Tweetdeck), there is a slight delay when I click on a link shared from Twitter. This seems small but I eventually discovered the cost of going back to to the client if the article is not interesting was in fact discouraging me from going through unless the article looks really looks interesting.

Reading my Twitter stream using Flipboard


In fact, I not only read faster, but I suspect I read more, since I could judge directly whether an article was worth reading rather than relying on the description given in the tweet which could be misleading.


Touching one of the stories automatically loads up the full news


One effect of reading Twitter this way was that it leads one to focus on content and I started to stop noticing who was the one who shared the content in the first place (a throwaway comment also made by panelists at the ALA Top techs roundup), which may or may not be a disadvantage.

For those without Flipboard, desktop services like Twittertimes, Paperli, Postano (formerly PostPost) etc give you a taste of how it looks like, though you really need a touch based interface to "flick" through pages to appreciate the power of this layout.





Postano (formerly PostPost)


Google also offers Fastflip, and Blog platforms like Blogger, WordPress have become offering themes that allow one to read content in a similar way.

Information Twist blog when viewed on iPad



The Flipboard way of sharing content while most common is not the only way, iPad and iPhone apps like Taptu, Pulse, Fludnews & desktop services like Dipity & Feedsquares visualize content in different ways.










Consuming Academic Content?

For me, the obvious question is this, how does this apply to consuming content for academic research?

While the goal of Flipboard and its competitors is geared towards creating a personal newspapers, it is just a short jump to creating a personal research journal, populated with articles you want to read no matter what journal or source it comes form.

This is simple enough to do I suppose. Simply load up RSS feeds (most flipboard like alternatives accept RSS feeds - even OPML files or indirectly via Google reader) and also Facebook, Twitter accounts of Library Journals, Library vendors even researchers who tend to share information in a particular area

You might have noted the Paperli example above is made out of RSS feeds of Table of contents of Journals. Paperli and Postano are flexible enough to handle many different types of streams but they are desktop based rather than tablet based (which means iPad for now)

So let's try this on Flipboard.



Flipboard on iPad is the most polished example of this class of tools and it's the natural one to try. It accepts a wide variety of content including streams from your Twitter account (Including Twitter lists, saved searches, people you follow, people following you etc), Facebook streams (Groups, Pages etc), Google reader (Your shared item, starred items, people you follow, Feeds & Folders) as well as curated content (from Blekko for example).


Some Flipboard options of streams you can use to populate  it



My flipboard drawing from "Favourite journals" folder in Google reader

Reading academic journal articles does have a couple of pitfalls as they are often behind paywalls.

Here are my proposed steps


1.  Get a google reader account and fill it up with RSS feeds from journals (I usually put them into one folder), you can use ticTOCS to find RSS feeds faster. I haven't really tried but beyond these, you could try curated results from Blekko or Twitter lists from Listorious.

Since I wrote in 2009, many academic libraries have gotten discovery services like Summon or Ebsco Discovery which allows you to easily create RSS feeds of searches that cover everything your library owns.

2.Treat all rss feeds like this , so the proxy stem is build-in to authenticate or use something like the proxy bookmarklet to handle sign-ons.

3. Load up Flipboard with folders of RSS feeds and other interesting feeds e.g Twitter streams of researchers, keywords of google alerts etc



How well does it work?





Above shows an example of a Google reader folder of RSS feeds from journals. First off while it works fine, it looks a bit boring , no pictures etc. I wonder why it doesn't show any pictures? Maybe RSS feeds of journals should add one or two diagrams :)

As Flipboard type tools are meant only for consumption  only they don't store anything long term, so you need some way to handle this.

While one can currently push information to services such as deliciousinstapaper , twitter etc sending directly to academic related networks like citeulike or even citation managers like Mendeley is not possible. I suppose one could "like" or tweet these articles then later go back to import them into your citation manager but that seems inefficient.

One workaround that might work is this. Say you see an interesting article on iPad via Flipbord.. click through to the vendor site on the paper. On that page, use a bookmarklet from citeulike to push it there which will push to Mendeley/Zotero (See Citeulike-Mendeley link-up).

While there are iPad citation managers like Papers and Mendeley,  I don't think they can directly grab the citation directly though they can store the pdf on the fly?


More intelligence/collaborative filtering ?


Flipboard unfortunately does not help you handle or prioritize items to read, as it lacks intelligence of any kind and simply lists everything you have. I wrote about my experiments using bayesian filtering to train the RSS reader to recognise which articles are interesting  perhaps something similar here would be desirable.

Twittertimes has something simple that take into account how often the piece of content was tweeted, shared on facebook etc (by your friends or overall)



Twitter Times scores content by how often they are shared on Twitter by your friends and friends of friends


Zite goes further and learns what you might like using machine learning techniques and is close to what I was talking about.

"It works by looking at the articles you click on and the characteristics of those articles. Is the article longer or shorter? Is it skewed toward one element of a topic or another? Is it a political blog? If so, does it have have a right- or left-wing slant?" Source

You can also explicitly tell it whether you like the article you read and which topics you would like to see more. It also learns implicitly "soft yes and no", based on whether you click on something or ignores it.


Some explicit feedback you can give articles in Zite

I suppose a system designed for academic content should also take into account different characteristics the number of cites etc, but given that most of the academic content you will be consuming will be new, this probably won't come into play.



Conclusion

What type of help can Librarians offer given such trends? Firstly, keep uptodate about such tools and teach researchers how to use flipboard or alternatives to handle research streams. Secondly, perhaps even curate such pages and share it with researchers.

In fact, while tools like Paperli  allow some form of curation,  tools and services like Scoopit and curate us are starting to show even more promise and should be watched closely.

blog comments powered by Disqus

Share this!

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
Related Posts Plugin for WordPress, Blogger...