Wednesday, October 26, 2011

5 library related tasks I hope Siri can do

While many people were slightly disappointed with iPhone 4S, the one feature that was considered cool and perhaps the main selling point is Siri , the intelligent assistant that among other tasks reads your email, manages your schedule, calls/texts your contacts etc.

Still, will people find it strange to speak commands out-loud in public? Is Siri smart enough most of the time or will it be frustratingly wrong most of the time?

It does make sense to use voice input controls for smartphones because typing is hard on the move particularly given the smaller keyboard, though honestly at this stage no-one knows if Siri or rather the broader paradigm of voice input controls (google had similar for a while already) will take off, though with Apple putting their weight behind it, there is a fighting chance of this happening.

If voice input controls do take off here are some possible library related functions that I would like to see. I don't have a iPhone 4S (yet) but some of the things below are already possible mostly I think, while others are more of a wish.

1. Ask Siri to text or email the library

In the ad above, the user says "Text my wife I am going to be late". As many libraries provide text a librarian service, users can now say "Text the library (or librarian?), " and then wait for a reply that will be read by Siri as well.  For libraries that don't offer SMS service, Siri can be used to email the library.  

I am not sure how many people have the library's sms or email address in their contacts though so that might be a stumbling block. I heard about one trick (sorry can't remember exactly where or from whom but might be at ALA Annual 2011) that might help. Basically the librarian conducts tours for freshman and while they are waiting for others to come and for the tour to start, she gets the waiting students to access the library Facebook on their smartphones and to like them. I suppose the same thing can be done for sms, emails. If the librarian has iPhone 4S or similar, she can even demo this trick :)

Practically speaking, countless surveys have shown that librarians are far down on the list in terms of people to contact when they have a question, so Siri probably won't change much if users don't want to contact you anyway. For those who do, this might lower the barrier of access somewhat?

2. Ask Siri to check if the library is open  

In the ad above the woman asks "Is it chilly in San Francisco this weekend?" And the answer comes back "Not too cold" with info on screen.

Would be nice if you could ask "Is library X open tomorrow at 9 pm?" and it answers?

Siri offers to do a web search so maybe currently it works already to some extent by doing a search and pulling up the right webpage on opening hours.

Still assuming Apple eventually opens the API so you can interface with apps, you might be able to interface with apps like Library:Australia , Ottawa (Canada) Libraries , UPLA (Hongkong), which have detailed information about all libraries in the region so Siri could ask questions like"Which library branch closest to me is open tomorrow at 8 am?" and answer directly.

3. Ask Siri to check if a book is available  

In the above video, based on the original Siri before it was sold to Apple, the guy asks "Where can I find Lunch?", later a lady asks "where can I find green tea latte?" and it suggests a few places based on location.

What happens if one asks "Where can I find the book - Steves Jobs?" would Siri just search and list Amazon results etc? How about "Where can I borrow the book ....?".  Again, wouldn't it be nice if Siri could interface with your library app, do a search then display the item.

With so many libraries having apps out for catalogues, whether it is their own, one that they get from their ILS (e.g BookMyne from SirsiDynixTRACpac etc) or maybe even the Worldcat app or the even broader Local books  from Librarything, I wonder if Siri could use information from there to answer and tell you which libraries near you has it and maybe even offer to place a hold.

4. Ask Siri if my books are overdue

Similar to #1, but Siri would somehow know you are asking library related questions based on your account such as "Is my book overdue?" or "How many books can I borrow from the library" and pull out the results from there. Again conceivably a general web search by Siri could find the answer, but if your app could be directly plugged in (some Q&A knowledge base system), it would give far more direct results, particularly if it could login with your loan record details to determine when a book is overdue (and offer to renew), or you could check if discussion rooms or pcs are full and book a slot etc.

Or even the ultimate holy grail, to tell you where the book you want is in the library or (the restroom is) and direct you there. "Take the staircase down...."

5. Ask Siri to search for papers and books for you on a certain topic

Siri already pulls information from Wolfram Alpha . Wolfram Alpha is not a search engine but is billed as a answer-engine that tries to answer questions directly as opposed to giving you a webpage that may have the ask. It can calculate answers from structured data such as computing the age of a famous personality now, or "What is the largest country in the world?" or even answers to maths equations.

Coupled with Siri it means in theory, Siri can also answer quick reference questions, but libraries have long stopped answering quick reference questions (at least the easy ones) due to google.

The 2 videos above show what Siri can do with Wolfram Alpha. The 1st is based on the older Siri where it manages to figure out some crazy family relationship tree using Wolfram Alpha.

The 2nd shows how to use Siri to figure out when the next elicipse is (based on your location). The 2nd video shows you that to use Wolfram alpha directly, you have to first say "Wolfram Alpha" (not Siri) , I suppose when other apps get intergreted, you have the same thing and you got say "Library app" first.

There is also one video by Steve Wozniak that raves about what Siri can do together with Wolfram Alpha (though I think he was saying the current Siri for whatever reason is less capable sometimes?)

But not every question can be answered directly, in particular doing searches for research. I wonder though if Siri can work with our library catalogues and databases to help make searching easier.

Maybe "Siri, Search Pubmed for articles on Cancer, written after 2009 , by ....." and it would pull up all the articles. Then it would read abstracts and you could download it for later viewing.

Or perhaps a scenario where you say "Search for me ebooks on disruptive innovation" and it would pull out ebooks on loan by your library, which you download and then let it read to you?? Talk about real "natural language" queries!

Not sure how difficult it would be to get something like this working, though the query is basically an advanced search (using facets?). Also unsure how many people do research on their phones, though this could be a case again where innovations on the smartphone later migrate to the desktops (as in the case where features in Apple's iOS is influencing their desktop OS). Google already offers voice input search, though I wonder how many people are actually using it.

A advanced searcher would probably laugh at this, and swiftly conduct more sophisticated searches particularly on a desktop, but one wonders if this might be helpful for more novice users who don't do advanced searches.


Some other uses that occur to me , including asking Siri to set location based reminders so you are reminded to return overdue books when you are in the campus, set up schedules for meeting with librarians etc.


I don't have iPhone 4S (perhaps those who have it can try out some library related questions and report how it does) so I could be totally offbase on what I wrote but it doesn't matter as this goes beyond Apple or Google related products, as I was just speculating on how library related tasks change with voice input.

Similarly I would guess that there is plenty of literature on using voice input for searching so all I wrote probably comes off as naive and I have no clue how hard or easy it is to interface with Siri or Android phones to accomplish some of my more ambitious wishes.

In any case, Jakob Nielsen has a old but still valid article on when Voice input would be most valuable , and he proposes that perhaps using it as an "additional component to a multi-modal dialogue might be most useful". 

I guess a Star Trek level type of interface with AI built-in would be fantastic but might be some ways off, though Siri is a very good first step that allows one to dream on what might be possible.

Saturday, October 15, 2011

What would Steve Jobs say? Reinventing the library catalogue

Steve Jobs first pitched for the iPhone at MacWorld 9 Jan 2007. Here is the original transcript and below is the video of the master at work.

There Steve claimed to be reinventing the phone. How would he reinvent the library catalogue?

"This is a day I've been looking forward to for almost 140 years. Every once in a while, a revolutionary product comes along that changes everything. And Libraries has been -- well, first of all, one's very fortunate if you get to work on just one of these in your career. Libraries been very fortunate. It's been able to introduce a few of these into the world. 1876, introduced the Card Catalogue. It didn't just change Libraries. It changed the whole print industry. In 1960, we introduced the first OPAC, and it didn't just change the way we all search for books, it changed the entire online world. Well, today, we're introducing three revolutionary products of this class. The first one is a next generation library catalogue search with faceted browsing and web 2.0 features. The second is a revolutionary searching device that searches articles from databases. And the third is a breakthrough mobile internet device. So, three things: a next generation library catalogue; a revolutionary searching device that searches articles; and a breakthrough mobile user interface.  These are not three separate devices, this is one device, and we are calling it icatalogue. Today, libraries are going to reinvent the catalogue!"

"So, before we get into it, let me talk about a category of things. The most advanced catalogues are called next generation catalogues, so they say. And they typically combine a catalogue search plus some social capability, plus they say it's web 2.0. It's sort of the baby web 2.0, into one device, and they all have these advanced search options in them. And the problem is that they're not so smart and they're not so easy to use, and so if you kind of make a Business School 101 graph of the smart axis and the easy-to-use axis, classic catalogues are right there, they're not so smart, and they're not so easy to use. But next generation catalogues are definitely a little smarter, but they actually are harder to use. They're really complicated. Just for the basic stuff like finding articles people have a hard time figuring out how to use them. Well, we don't want to do either one of these things. What we want to do is make a leapfrog product that is way smarter than any catalogue has ever been, and super-easy to use. This is what the icatalogue is OK?"

Okay so it didn't fit very well but then again I don't have a reality distortion field :) 

BTW the hypothetical "icatalogue" is in fact the web scale discovery product/unified index product, but improved further perhaps with more social features or should it be gamification (maybe along the lines of the social features for bibliocommons or Lemontree project)?

R.I.P Steve.

Bonus : Downfall Parody of savelibraries campaign in UK by Phil Bradley.

For those who are unaware this is a internet meme, where people post fake subtitles of a scene from the German movie Downfall, where Hitler discovers some unpleasant fact and starts ranting. Of course since this is a parody the fact discovered can be something utterly trivial (such as being banned from World of Warcraft to something bigger like Obama becoming president).

I have not found many library related downfall parodies (maybe 3) but this seems the best. With so many library issues from the serials crisis, budget cuts, library closings, ebook pricing , one suspects that more could be created...  Anyone game ? :)

Sunday, October 2, 2011

Doing a library session online - some options considered

We live in a world where it seems we have countless ways to communicate with our users. Recently, I have added a new one to my toolkit - WebEx a popular web conferencing software. At our institution we also have other tools/services options such as Camatasia Relay, Ink2Go, Adobe Breeze,Live sessions can be recorded as  "webcasts", while the free tools like Skype, Hangouts in Google+ gives us even more options.

With the diverse tools available I find myself  confused- what is the best tool to use? For example, if I am thinking of conducting a remote library orientation or perhaps a meeting. Should I screencast/webcast/videocast/slidecast/cast ?

Beyond the exact definition of each term more importantly when should each tool be used? This is my attempt to clarify all these tools in my mind.

One to One vs One to many

To me, one fundamental way to differentiate online tools is to consider whether the tool is meant for one to one sessions or one to many. In this blog post, I will discuss tools for one to many online sessions. The focus here will be on librarians sharing slides, videos, documents , browsers, displays of *their* screen to large groups online and narrating with their voices and sometimes showing videos of their faces from webcams.

Of course if a tool can be used for broadcasting to groups it can be used one-on-one. 

But arguably, one on one sessions often differs from group sessions usage. In particular one-on-one might use functions like pushing of pages (librarian sends user to specific pages) and cobrowsing (automatic synchronization of pages) or even remote desktop/access/assistant tools (user cede control and give full access to remote librarian which is great for troubleshooting).

Cobrowsing is a related but different technology to screensharing.

"Co-browsing differentiates itself from screen or application sharing because it is not achieved by sending a high rate of screenshots of the presenter screen to one or more participants but by distributing in a synchronous fashion the URL(s) being visited by the presenter to all participants' browsers."

My understanding is that libraries in the past have done cobrowsing/remote assistance usually in the context of one-to-one sessions via chat services like Questionpoint, but demand for this isn't particularly high compared to straight textual chat for various reasons.

To add further confusion there is a difference between "screen sharing" and "application sharing", where the later is "more specifically the ability not just to show specific applications to other connected participants but the combined ability of screen sharing with remote control.", though the distinction usually isn't made.

But for most use cases I am considering the focus is usually for the presenter to share the view of his screen to a group of audience rather then to look at the audience's screen and seldom is the audience allowed to control his screen either. 

While some advanced services do allow both screensharing, application sharing & cobrowsing type facilities, with higher bandwidths and technical issues with cobrowsing, screensharing is the main method of presentation.

In what follows I will focus on librarians doing screensharing ie sharing their screens or presentations to others rather than cobrowsing/remote assistant type functions.

Prerecorded sessions

The main decision you have to make is whether you want to have something prerecorded so your users can watch it anytime they want. This would be asynchronous type of session. The advantage is that you can have "digital extensions" of yourself that is always ready to help 24/7. The disadvantage is that it is all prerecorded so any element of interactivity is gone.

Some presenters are more comfortable with prerecorded sessions compared to "live" sessions because they can ensure that whatever they do is "perfect", recording over and over again until they are happy. On the other hand the pursuit for perfection can be very time consuming and often can also lead to a very boring presentation that lacks a certain spark a real live presentation includes.

Hybrid ideas include doing a "live presentation" which is also recorded for future viewing. So for example WebEx a web conferencing software generally meant for live seasons has build-in recording functions.

Another interesting idea for presenters who are afraid of making errors in real-time is to prerecord certain sections, play them in the "live" online season, then react in real-time to questions sent online. Of course this whole thing will then be recorded. Again Web-ex does allow this.

Below are some tools that are I believe purely meant for prerecording with little or no live casting capabilities.

Tools you can do to prerecord include anything from full-blown powerful tools like Camtasia/Adobe Captivate to free/freemium services like Camstudio, Jing or webbased ones like
Screencast-O-Matic, ScreenCastle, Screenr (comparison on Edtech blog). Though many of these webbased tools have limited session lengths for recording.

They all have similar elements, you can record whatever is on your screen (which is usually a program or a browser screen), you can add a voiceover, usually you can add text captions etc. The more advanced ones allow better editing functions, zooming, auto-upload to Youtube, maybe even inserting webcam pics on top (picture in picture)

The above software/services records whatever is on your screen, so you could record powerpoint slides, videos, browser screens, programs or a mix of whatever you want. But there are tools that are presentations tools that are for one type of presentation only (e.g Powerpoints).

It is of course possible to use pure pointpoint plus audio recording for video over (using built-in powerpoint "Record narration function") but you can also use software like Breeze powered by Adobe Presenter (now replaced by full webconferencing Acrobat Connect Pro). These are basically software designed around a certain type of presentation typically powerpoint slides rather then recording everything on your desktop.

Compared to desktop screenrecording this can be more "professional" because your screen/desktop is often cluttered with a lot of additional irrelevant items.

Livecasting tools

For tools that allow you to do live online sessions there are many options , and they differ in terms of capabilities including
  • Audio support (Some of the basic versions only allow the main presenter to speak)
  • Support of webcams overlaying main content (slides/videos/documents etc)
  • Screensharing  (presenter only or anyone, control of desktop etc)
  • Other interactive options e.g Whiteboards, annotation sharing options etc.
  • Supports viewing on mobile
The most flexible options are I believe top tier enterprise level web conferencing software like Webex, Blackboard Collaborate (formerly Elluminate), Adobe Connect etc. I only started using Webex (subscribed by my University) and it is a very powerful software bursting with features. You can share desktop, a specific program, or documents, there is built-in support for multiple webcams, you can easily change presenters, chat options including Q&A functions for sessions where there are many attendees and chat is likely to get too noisy, polls, interactive whiteboards, ability to divide into smaller "breakout" groups the list goes on and on.

Essentially you can use this for anything from webinars with hundreds of attendees to one on one sessions (though that might be overkill).

MSN options are Microsoft Office Live Meeting (formerly NetMeeting)
I haven't looked at many free or near free options, many free tools I tried in the past such as dimdim and yugma have gone paid and some free tools like Zoho meeting are limited to 1 on 1 for free versions. seems to be one viable option on par with Webex. The free one is adsupported but does up to 10 people only. Mikogo is a free webconferencing software those is selfhosted.

Another one to consider is Google+ hangout. A recent update, Google+ hangout with extras allows you not only to hangout via videos but allow you to share your screen. It is limited to 10 people but you can broadcast your hangout live to many more. But it's pretty limited and is not a full blown web conferencing software.

Google hangout with extra here I am sharing a google doc

If you are not looking for the ability to do full blown desktop sharing, there are tools like zipcast from slideshare allow you to have online meetings over slides posted on slideshare (voice only + slides on Slideshare - no other screensharing) , tools that allow you to watch youtube videos together (synchtube etc).

The key thing about such tools is that they should be as easy to use as possible. Ideally all you need to do is to give out a url for users to click and it starts to run on their browser with as little additional installation as possible. 

Webex is very powerful but is relatively slow and complicated for users to join the session. Google+ hangout is on the other extreme is relatively easy to setup (though you may need to download a googletalk plugin).  


I have only considered one use case, librarian presenting a library orientation or information literacy session online to a group of users. For other use cases such as teams doing collaboration online there are other collaborative tools which basically involve collaborative tools (office suites, wikis etc) adding videos and voice. Another area I haven't explored is the idea of packaging recorded live sessions or prerecorded sessions as video podcasts (does that make sense?)

Kiosk-like systems stationed at places were librarians are not around for users to contact us would be yet another use case which would require a different mix of functionality.  Would users want to see the librarian's face? Hear their voice? Or would text based chat be enough? Should it be a full 2-way video call so the librarian can see and hear the user or just one-way? My guess is for such devices to work it has to be almost seamless, as easy to use as a light switch (idea from David Lee King's presentation at NLS5), loadup time must be almost zero, otherwise no-one would use it.

Would be interested in librarians experiences with Kiosk-like systems, do users use it? Or is it too troublesome compared to the ease of opening one's mouth to ask a real life librarian standing in front of you?

Sources: An excellent blog on such tools is 

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...