David Tebbutt, originally published on IWR

Richard Wallis is an inventive cove. He works for software company Talis
which is, itself, deeply into the semantic web. It describes its team as
“revolutionaries taking part in the mass liberation of data.” It’s best known
historically for its library applications and services.

Wallis has spent a few months working on an open source project which enables
anyone to slip a few lines of source code into their web pages in order to
massively widen their access to additional information.

The project is called Juice, which stands for Javascript User Interface
Componentised Extensions. Not bad, as acronyms go. Before you run away,
terrified, let me tell you that it is fairly straightforward, providing you’re
prepared to spend an hour or two watching a screencast and going through some
worked examples.

The end result is that if you were a librarian, say, you would be able to
display more information about books and libraries right there in your own OPAC
or resource pages. A Google map showing which libraries are carrying a
particular book? A link to reviews or other book information? Anything is
possible, providing your web page already contains the information needed to
pick up the details from elsewhere.

How the heck is this done with just three lines of code? Well, that’s the
easiest part of the exercise because two of them point to Juice Javascript files
and the third points to one you create.

The first two represent 5727 lines of code that you don’t have to write.
These are the elements that make it all work and make it safe for you to run.
The third contains the information you need to personalise Juice to your web
page layout. Fortunately, you can copy chunks of code from the Juice website and
simply tweak the web addresses used to match your own URLs. And, for each
function you want, you can add a pointer to a pre-existing file that you can
pick up from a library of these functions. Indeed, you can get other files like
CSS which determine how the browser presents the information on your screen.

You will also need a definition file to access external services. This tells
Juice where on your web page to find the information needed to perform the
search. It might be an ISBN or a library postcode, for example. If you are using
a standard application, you will probably be able to pick up a suitable
definition file from the Juice library. If you are using a bespoke application,
it’s best to grab an existing definition and tweak it to match the content of
your web page. Again, not very difficult, especially if you’re comfortable
picking your way through HTML to find the values.

Because the project is entirely open source, Wallis expects others to add
useful files which users can just pick up and pop into their own websites. A
tame programmer or web developer might save you a little time but this stuff is
fairly straightforward, providing you resist the urge to panic or think “this is
too techie for me”.

Juice is a glorious example of the open source movement in which people
develop stuff for themselves and then share it for the greater benefit of all.
Libraries, local authorities and businesses can enhance their web presence at
the cost of just a few hours’ work. Talis benefits from plenty of plugs on the
Juice website, but nothing too onerous and much of it is because it has provided
most of the content to date and, of course, most of the examples are drawn from
Wallis’ own experience. As more organisations see the benefits of allowing users
to extend their applications’ web pages in this way, the balance will shift away
from Talis. But nothing will ever take away the fact that the Juice project
originated with one of its staff.



Leave a Reply