JISC LMS meeting in Glasgow

Huw and I (Dan) went to Glasgow to the JISC LMS meeting.

So, if the joys of two nights spent catching sleep on the Caledonian Sleeper are the downside, what were the benefits?

Many actually. In three important ways.

First, we heard a lot of ideas. We plan to implement a fair few of them in our project. For example, another JISC-funded project, ConnectedWorks, has led to us investigating a browser plugin called iCite. When someone mentioned the idea of adding a "find in my library" button to amazon searches, we realised we had just the tool for the job. So we're going to add a web-assistant browse widget to the project. There were many other examples, some we will have time to do, others will have to wait for future projects.

Secondly, we learnt about other projects in the programme, including projects we could work with. The Wolfie project at UEA is a standout there, as is us using some of LSE's earlier work. Generally, lots of good networking went on.

Finally, we were helped to develop our arguments and pitches for our projects, to help us describe them to each other, and the various people who might be able to help us in their execution.

There was also some excellent discussion around UX, and how we might be better at pooling the outputs of our user research (to avoid duplicating effort) and be better critical consumers of UX research (we all know how much poor-quality UX research is there: how do we wind the diamonds amongst the glass?).

One Minute Pitch

One of the requests which came out of the JISC LMS meeting was that we prepare a one minute pitch for our projects. (Commonly known as "elevator pitches". The idiosyncratic lift at the venue was very much the ghost at the feast on the first day of the meeting, having a tendency to steal the limelight by appearing on our floor unbidden and aimlessly chomping away with its doors throughout proceedings But, anyway,...).

Often, the biggest obstacle to overcome when explaining widgets is to overcome the frivolity associated with the concept: the silly name; their ubiquitous use on social networking sites for producing pointless distractions, and so on. We encourage you to set such preconceptions aside, at least temporarily.

There are two reasons for using widgets for library services.

Number One: widgets make things happen in institutions with complex organisation and a tendency for long meetings and endless debate. That's one of their main advantages. Development is in three parts. The information provider provides an API (as they see fit). The consumer provides a widget container (again, as they consider appropriate). Neither of these two parts require extensive consultation. The final part, the widget itself, -- which is the only part which ties the two parts together, -- can be written in a day.

Widgets are small (we're talking person-hours or -days) and easily rewritten. This allows you to take risks in creating them. If some users have indicated that a widget might be useful, you can spend half a day creating one: much less time than it would take to perform a formal survey.

Think of how many projects there have been at your institution concerning integration of various computer systems. Think, nationally, how many projects there have been to integrate just catalogue and circulation information between VLEs with LMSes. Consider their budgets. We have such functionality already, in production, against our enterprise sites, in use by academics and students, as a result of a small project. And that's just one widget.

Number Two: as a user, widgets allow you to do things "from where you are". There's no "that's in another system" issues. A user can, effectively, design their own screens to fit the way they work. No need to login again (auth permitting), or remember the URL or navigating to the page you want from the front page of a particular site. A screen of widgets might contain some parts derived from the LMS, some from the VLE, some from teacher-provided data (such as reading lists), and so on.

Many of these containers (for example, in a VLE) have rich knowledge of the current user, too. That means that you can present "zero-click" search results, for example for exam papers: no need even to search for them because the results displayed by default are usually what the user is after. Both of these mean that they spend more time interacting with the real content, less time navigating.

Both of these things excite us, and present an opportunity to improve the way we do things and make what we offer to students more compelling.

So we're producing example APIs and Widgets as might be used at other institutions, along with guidance of how to do so. We're putting our money where our mouth is by developing at least half a dozen widgets we anticipate being useful to students and academics (some of them have been repeatedly explicitly requested, some others are widget-ising of existing useful functionality), and writing up how you can go about it at other institutions.

So, we'd particularly recommend this approach to institutions which have struggled, technically or socially, with integration in the past, and where the students are faced with a confusing array of different systems to use, as widgets can help with both of these.

Despite their silly name.

Back to school, off to Glasgow

After a typically fragmented August (if you're not on holiday yourself then everyone you need to speak to is) it's back to school for the CULWidgets team, starting with a trip to Glasgow for the JISC LMS Programme Meeting. As we draw slowly towards the end of the project it's a great opportunity to focus on what we've achieved and what we still need to achieve as well as a chance to see what everyone else is doing!

One thing we have made progress with over the Summer is getting the student data we need to give our widget services relevance. We have one minor technical hitch to get over, and then we should be able to get a test version of our Exam Papers Widget up and running.

This really is a key strand of the project for us. It involves institutional collaboration - the University Library, CARET, DSpace@Cambridge (institutional repository) and the Management Information Services Division (who run the student registry) are all integral to the project. It involves a new kind of material - the Library has traditionally only dealt with exam papers in paper form. It involves the provision of relevant services in relevant places. And it involves direct engagement with the student and academic community. Finally, we hope it will be a blueprint for the course-based delivery of teaching and learning materials across the University.

Reading List Hackathon - a Summary

On Thursday and Friday of last week, I attended a Reading List Hackathon run by DevCSI, at the Møller Centre, Cambridge. The event was designed to encourage the hacking of library information systems and reading list software.

We looked at two major pieces of reading list software:

• Telstar

and

• List8D

with talks from Mendeley (who provide citation software), Talis (who provide library software and reading list software, amongst other things) and Emerald (a publisher who are creating peer-reviewed reading lists as a free service).

After these talks, we got together into teams, and started hacking, creating, amongst other things:

• a way of pushing Emerald RSS feeds through Yahoo pipes, allowing filtered searching
• some extensions to List8D
• Six degrees of Harry Potter, which used Talis API data to generate a graph of degrees of separation
• Compare the Citation: a teaching tool to educate undergraduates about citation styles
• an extension to a library catalogue, showing availability of items on a reading list
• an extension to List8D, allowing users to display citations in different styles
• using the Talis APIs in List8D
• Llikes, a tool to reorder a reading list based on social metrics
  

I'll write a couple of blog posts detailing what people said in their talks, and more about what was created during the event.

But basically, I had a really good time; I discovered lots of new things; some really cool and interesting things were created, and I'd love to go to a similar event again.

Reading List Hackathon

One of the things we're interested in doing is innovating ways to disseminate and manage Reading Lists.

Our latest widget, the Reference List widget I mentioned last week, covers a related but slightly different use case to Reading Lists. Reading Lists are much more broad, and could include published books, journals, local (e.g. departmental) reading material etc; whereas the Reference List widget caters for lists of citations in a very well defined format.

Right now one of our team members, Verity, is at a DevCSI workshop exploring the interoperability between reading list software and other software systems - we'll let you know of any useful things she picks up at the Hackathon.

A new widget and a new gallery!

Over the past few weeks we've started a "production line" of widgets where widgets start off life as ideas, become prototypes, then emerge as demos and gradually develop into more finished products. Watch this space for more news about some of the other widgets we have in "incubation" in the production line.

The first new widget to emerge from this production line as a "beta" release is a Reference List widget - this displays a list of journal citations and allows the user to filter them by various criteria like keyword and author. This could be used as a handy way for a course lecturer to disseminate the References relevant to various lectures within a course. It is powered by public JSON feed from a Google Spreadsheet which makes it easy for non-developers to maintain the source data. I'll blog a bit more about how this widget works in a later post.

In the meantime, I've set up a quick and dirty Gallery Page to show off our widgets as they emerge from the production line - you can see the Reference List widget in action on there, along with the two other widgets that are in a ready-to-demo state (see earlier posts for more about them). The widgets and the Gallery Page itself have some rough-edges which we'll tidy-up in the coming weeks but for now you get a flavour of what we've been up to. We also hope to supplement this demo page with a more "directory-like" page, a bit like an "app store", which will tell you more about each widget, and allow you to add the widgets to your own portals (e.g. Facebook, iGoogle).

Widgets, mobile and atomized services

Interesting thoughts from Lorcan Dempsey:

http://orweblog.oclc.org/archives/002115.html

Talking about mobile devices, but one point in particular is very relevant to widget design

"Atomization: get to relevance quickly. Mobile encourages designers to think of atomic services rather than complicated workflows or rich multilayered experiences. And to think about services that are immediately relevant and convenient. Room or equipment booking or bus time tables may become more visible, for example."

The simplest widget just takes an existing web page and puts it in a box - but this really isn't a very effective use of the medium.

When we're discussing ideas, a question we ask ourselves is "what makes you want to turn this into a widget?" Just because you can doesn't mean you ought to.

A few possible reasons:

• You want to make services/information available outside traditional interfaces i.e. in social networking sites
  
• You want to isolate a particular, commonly used service and provide a streamlined way to access it
• You want to bundle a few related services/bits of information together around a particular set of workflows

I suppose the "widgetisation" of an interface is to do with how and where you want access to happen. But the restrictions (i.e. space) on widget development make you think very hard about use.

In a nutshell, it seems to me that widget development is about providing relevant services in a context. You could probably say this about all development, but there's something about widgets which concentrates the mind!

Another Hello

Hello my name is Anne and I am a CARET-based developer who will be working on Library Widgets with Raymond.

I am getting up to speed with this project and took part in a Widgets Brainstorming session last week. I have also been benefiting from reading everyone's postings on this Blog.

I have been tasked with writing PHP mediator scripts to automatically run a series of searches using both the Aquabrowser and Voyager APIs. The results from these searches will be merged together to allow Search Widgets to do their stuff with the data.

I'm looking forward to meeting and working with our colleagues at the UL.

transitions

Dan has now returned to CARET from the University Library, and will be returning to the helm of this project over the next few weeks. So there will be less from me, and more from him!

Google Code Site

Software Developers - have you checked out our Google Code site yet? This is where we're keeping documentation and code, and where we'll track any issues like bugs and feature requests.

In particular, you can browse our source code - all covered by the Apache License 2.0. We will be augmenting the code with full documentation of how to host, configure and install your own version of each widget in due course, but for now there is a snapshot of the code that currently powers the original Library Widget, and most of the code used in the prototype of the Exam Papers Widget.

If you simply want to use or try our widgets as-is, there's no need to delve around in the source code - find out more about how to use the Library Widget and the mobile version, CamLib.

Cambridge Library APIs

I've blogged a few times about how our widgets and interfaces are underpinned by web service APIs to our data sources. We've now made the decision to publish all of our public APIs - documentation can be found at http://www.lib.cam.ac.uk/api/

This means that any developer can access and build functionality around Cambridge Library data. It's going to be interesting to see what they come up with!

CamLib mobile interface and Aquabrowser

One of our aims in this project was to integrate our widgets with the new (forthcoming) Aquabrowser discovery layer, which will be launched later this summer. Cambridge library resources come in from a huge variety of data sources, including eight (yes, eight) Voyager databases. Searching across these collections presents a huge challenge for our users. Look at the previous search options:

• UL & Dependents
• Departments and Faculties A-E
• Departments and Faculties F-M
• Departments and Faculties O-Z
• Colleges A-N
• Colleges P-W
• Affiliated Institutions
• Manuscripts

So it was difficult to even search across all colleges, let alone the whole of Cambridge.

Aquabrowser takes feeds from all these data sources and rearranges and tidies up the data (through a mixture of straight deduplication and FRBR) to make it all more manageable - and more importantly provides a single point from which to search all of our collections.

We've now managed to integrate Camlib, our mobile interface, with the Aquabrowser APIs to provide a much improved search service. Our new search options are:

• Search All
• UL & Dependents (incl. Manuscripts)
• Departments and Faculties
• Colleges
• Affiliated Institutions
• Electronic Resources

The big difference is that you can now search across the whole of Cambridge. But you can also search across the different types of collection more sensibly. As an extra we've separated out electronic resources as a distinct section for those only interested in the online.

The search results are very differently presented - one result for each title, followed by a list of the libraries who have a copy.

In short, if you are looking for all copies of a particular title in Cambridge, instead of having to make eight separate searches and plough through multiple results in each, you make one search with all the relevant information returned under one result.

We're hoping that a by-product of this kind of searching will be increasing integration of library services in Cambridge - if people see us and our collections as an integrated whole, then we might move faster to meet their expectations.

And we've even got some funky book covers from Syndetics to spice up our results list ...

Proprietary web services and proxies

A couple of weeks ago I blogged about writing local web services to extract information and embed functionality. However, many data sources come with their own range of ready-made proprietary web services. The big advantage of using proprietary web services is that you don't have to write your own. The disadvantage is that (usually) you can't alter them. Unless you have a very good relationship with your system supplier, you're stuck with what you're given.

Most data sources will allow you to write web services which extract data from the system, but the majority won't allow you to write web services which alter data in the system. If you want to use functionality in your widgets rather than simply displaying information (i.e. renew books, rather than just say what you have on loan) then you may find yourself having to use proprietary web services.

Basically, if the existing proprietary web services do exactly what you want, use them. If not, you might want to consider writing your own to extract data. But as far as altering data in the system goes, you may well be stuck with the proprietary ones.

[In practice, we find ourselves using a mixture of local and proprietary web services for data retrieval, and proprietary web services for data alteration]

Somewhere in your system's documentation you should find a list of existing web services and some notes on how to use them. This should include the service address, input parameters and return values. Which translates as how to find the service, how to ask it to do something, and how it will respond. Pretty much the same as using locally written web services.

But there is one potential complication with proprietary web services. If you make a request to a web service from a browser and then try to read the response, the browser will first establish what the response is and where it comes from. If the response is in XML and the domain of the web service is different from the domain of the page calling it, the browser gets worried about security and the whole enterprise collapses.

There are two ways round this. If the proprietary web service has the capability to return its response in json then you don't have worry about where it's coming from. If not, you will have to write a proxy.

A proxy is an intermediate web service which can sit alongside your local web services, in your own domain. You access it just like any local web service, and your browser will treat the xml it returns as if it had come from a local web service.

When you send a request to a proxy it passes it on to the proprietary web service, reads the response and then passes it back to you. Because the proxy is not a browser it doesn't care where the proprietary web service is, and because the proxy is in your domain, your browser is cool with it too.

An additional advantage of going through a proxy is that if your system's proprietary web services return responses which are over-long, oddly structured or confusing, then you can talior that response on the way back so it gives you what you actually need.

We use one proxy web service to connect to all of our library management system's proprietary web services. So we pass the service name, and the request to the proxy. The proxy calls the appropriate service with the request. Then it processes it and returns it to the browser.

Opening Hours and Library Pics

We've now integrated library information (address, opening hours) into various parts of CamLib

Opening hours are pretty important. When a user finds a book online, the next thing they want to know is usually "when can I get this book" - though in Cambridge they may also be asking "can I borrow this book" or even "can I use this library". All of this information is available in CamLib, but not in a particularly joined-up way.

The next challenge is not just to display this kind of information but to use it in the provision of relevant services - allowing users to filter searches to only bring back items in libraries they can use, or items in libraries they can use where they can borrow, or items in libraries which are open and which they can use.

We've also put some pictures of libraries up. Not all libraries have pics yet - we're adding them as they get sent in. The pics make the interface look nicer, but also have a practical purpose. The bigger sites in Cambridge have large numbers of buildings, often crowded together. A map reference only gets you into the general vicinity.

We asked libraries to include photos of the building as a whole and the entrance to the library so that when users are there or thereabouts they can use them to find the particular library they need.

Developing for mobile devices

Good article in UKOLN's Ariadne on developing for mobile devices http://www.ariadne.ac.uk/issue63/massam-et-al/#16 Lots of useful links and tips, and some nice quotes from students

Basically reaffirms that students will use mobile devices for the web when it's cheap enough, quick enough, and when screens are big enough to let them see what's going on ... and popular wisdom suggests that we are reaching this point.

Students also seemed to be saying that they would use the mobile internet to check (and possibly discover) rather than to consume (or possibly create). My hunch is that users will want to be able to duck into mobile interfaces to perform a distinct task - checking a reference, renewing a book, finding a library - and then duck out again.

The minimal startup time for mobile devices compared to laptops/desktops means that I often check something on a mobile even when I have access to a "proper" computer. For fact checking, the advantages of a proper keyboard/big screen don't outweigh that interminable wait for the whole thing to get going.

I use a laptop/desktop if I need to write anything longer than a tweet, or read anything longer than a (short) article - which I may have discovered on my mobile.

Providing library content on mobile devices is another matter, particularly as most of our electronic content is accessed via third party interfaces.

Integrating our own service

We've talked a little about collaboration and integration of services between institutions - but sometimes the need for integration is closer to home. At the UL we have two sources of library data - our Voyager library management system, and a local database holding specific information about each library in Cambridge. This ranges from opening hours to location, staff details to system information. The two databases are linked by common identifiers so they can "talk to each other".

This local database is used to drive interfaces such as the Cambridge Libraries Directory and Libraries Map. It's also the foundation of a number of services we provide for Cambridge librarians - web based reporting from Voyager, course booking, record enrichment, staff management and stockchecking being just a few. Essentially, we have used development based around this local database to fill in the gaps where Voyager either has no slot for the information we want to store, or doesn't provide the service we need.

Widgets provide the perfect opportunity to start tying this data together more closely. We're already doing this to a certain extent in the CamLib mobile interface - Voyager knows which library a book is in, our local database knows where that library is, allowing us to link out to a map.

But there are all sorts of other possibilities - from simple ones like including library opening times in searches, to more complex functionality like "show me the nearest copy of this book", or "only show me books which are held in libraries which are open now". Another example of how bringing data together to get an idea of context allows you to provide relevant services for your users.

Collaboration with DSpace@Cambridge

This is a quick hello from the CARET-based developers on the project, Raymond and Bert!

So far in our portfolio of widgets we have Huw's "Library widget", we have CamLib, a mobile version of that widget currently in beta release - and a prototype version of an Exam Papers widget written by Harriet, a previous Arcadia fellow.

As Huw explained in an earlier post, there are two main types of widget - those that retrieve data, and those that alter data. All of our widgets so far are primarily about the retrieval of information - the next new widget will be one focussed on alteration: a DSpace Upload Widget. This will make it easier for users to contribute content to our institutional repository: DSpace@Cambridge.

To facilitate this, we'll be collaborating with the DSpace@Cambridge team, who are based at the UL. Bert and I met with the DSpace team last week to kick this off.

Writing Web Services for widgets

In a previous post I outlined the importance of web services to widget development, and pointed to two kinds of web service - proprietary ones which you can't alter, and locally written ones which you can. Today, I'm going to concentrate on locally written web services, and talk about some of the issues involved in writing your own web services. We're going to be making our code available towards the end of the project, so you should be able to pick up some complete examples soon. In the meantime, here are some of the basic ingredients of a home-baked web service.

1. Getting started

• Programming language: we use Perl (and occasionally PHP).
• Data source: the database or dataset you want to interact with (i.e. Library Management System).
• Query language: the language you will use to retrieve or alter the data in this system. Usually SQL.

2. Finding a home

• Your web services will need to live somewhere on your web server. In theory they can go anywhere, but it's best to have a separate directory for web services in general, and then distinct sub-directories for each system or dataset you want to interact with.
  
• Putting all your web services into one overall directory has the added benefit of making it easier for your system administrator (or you) to keep on top of the permissions and settings needed for web service files. Basically, if you're writing web services in Perl, the directory needs to be CGI enabled, and the files need to be readable and executable by anybody (usual octal permissions 755).

3. Do a bit of thinking

• If you're doing a lot of interaction with a dataset you'll probably have a range of questions/alterations you want to throw at it. Most "complete" transactions with datasets are made up of multiple smaller transactions (i.e. to book a train ticket, you might need to find the current time, find the next train, and book the ticket.)
• Instead of trying to write one huge web service to do everything, break each activity into its smallest distinct parts. This will leave you with lots of small, simple web services, each of which does only one thing.
• Now you can chain these simple services together into more complex transactions - but you can also reuse them in other transactions. This saves you asking the same question (i.e. "what's the time?") again and again in multiple web services.
  

4. What input do you need?

• Web services are triggered by requests. These requests (like web services themselves) can be of two types - questions for services that retrieve data or instructions for services that alter data.
• Requests are made up of the location of the web service, plus as many parameters as are necessary to either frame the question or perform the instruction. These parameters make up the input for the web service.
• The smallest amount of parameters needed is none. Think about a question like "what's the time?". If you had a web service which just gave back the current time when triggered, no further parameters would be needed.
• Most web services take at least one parameter. Taking the "what's the time?" service, if you introduced a parameter of "city" the web service could alter the query depending on the value of the parameter. So the service could ask "what's the time - in London?", "what's the time - in New York?", depending on the value of the parameter.
• The rule of thumb with parameters is that you should use as few of them as possible in order to ask the question/perform the alteration. You can always add more if required. Again, you should be thinking of small, simple services, each related to a specific task.

5. Checking input

• There are two things you need to check your parameters for - existence and content.
• For each parameter, you need to decide if it is compulsory or optional. Compulsory parameters are necessary for the web service to perform its function. For example, a service to book a train will need the train time in order to make the booking.
• But you might make the "city" parameter of "what's the time?" optional - if it exists, return the time in that city - if not, just return the current time in the UK.
• All parameters, compulsory or optional, need to be checked for content. Either the value will have to be in a limited range (i.e. if the parameter value isn't "Y" or "N" then it is an invalid value). Or it might need to be a type of value (i.e. a number, or a string). Even in the case of parameters which allow "free text" values, you will still have to check for "dangerous" characters. Web services (and all programs executable by the web server) are vulnerable to security attacks, and you will need to make sure that incoming parameters don't contain anything dodgy.
• A good rule is to decide what values or kind of values you will allow in parameter values and build from there, rather than trying to disallow all the possibly dangerous characters (you may miss one). So a rule like "the value must be a mixture of letters, numbers, commas and whitespace" is better (or more secure) than "the value must not contain semicolons).
  

6. Building the query

• Having gathered and checked parameters (if any) the web service now has to perform its function. To do this, it needs to talk to the dataset - and the usual language for doing this is SQL.
• Web services which retrieve data will use SQL to ask the dataset a question - such as "What's the time in New York?"
• Web services which alter data will use SQL to form an instruction. For instance "Book a seat on the 9.15 from Cambridge to London"
• If your web services use parameters for input, these will (usually) be interpolated into the SQL. So, for the train booking example above, "9.15" would have been the value of the parameter "time", "Cambridge" the value of the parameter "from" and "London" the value of the parameter "to". Parameters are a means of altering the question or instruction which gets sent to the dataset
  

7. Returning the response

• All web services should send a response back to whoever made the request.
  
• For web services which retrieve data, this will be the answer to the question - i.e. for question "What is the time in New York" the response might be "The time in New York is 10.30".
• For web services which alter data, the response will tell you what the web service has done. So for the instruction "Book me a seat on the 9.15 from Cambridge to London" the response might be "Seat booked" or even the more verbose "Seat no. XX booked on the 9.15 from Cambridge to London".
  
• The response should be tailored to how it's going to be used. Sometimes, you'll just want to know if the request was processed ok, sometimes you'll need much more detailed information.
• Your web services should always send a response, even if the request fails. A response of "your request has failed" is far more useful than nothing at all!

8. Chaining web services together

• A common way of using web services is to chain them together - which means using the response from one web service as part of the input for the next.
• Taking the (very simplified) example of booking a seat on the next train, the chain might go something like this:
• Request to "what's the time?" returns value TIME
• Request to "what's the next train?" with parameter TIME returns next train after time as value TRAIN
• Request to "book seat" with parameter TRAIN returns value SEAT BOOKED

Et in Arcadia ego

Avid followers of this blog will have spotted references to the Arcadia programme, a three-year programme funded by a generous grant from the Arcadia Fund to Cambridge University Library. The grant will enable us to explore the role of academic libraries in a digital age, create new programmes and services, particularly for undergraduates -- and also to improve the external environment of the library.

If this sounds like your cup of tea, you might also be interested in the Arcadia project blog, written by many of those involved in the project, as Arcadia Fellows or supporters. This includes a surprisingly large number of the CULwidgets team!

Project Plan Post 7 of 7: Budget

Here's our project budget! (and this is the last of the "project plan" posts, so, dear reader, you can relax as we return to our normally scheduled blogging about progress and things we've learned.)

We have some travel and other expenses included, to enable us to meet up with other jiscLMS projects and to share what we learn on the project, and to pay for food and incentives for users who are involved with our user testing.