TALIA TUTORIAL APPLICATION TOUR

This is an overview of the various features of the Talia Tutorial Application. This application is a perfect copy of the latest "stable" demo application for the Muruca project.

You can install this application by following the instructions on this page.

The application is meant to show some examples on how a digital library can be displayed to the user, -whether human or machine- and how it can be managed.

Talia Tutorial has two different frontends to show the data the first, sources dedicated to show the semantic data, its relations and some examples of use of those data. boxView, the second frontend, is a more "common" representation of the same data, using a more pleasant and dynamic presentation.

There is also a backend application, with user authorization and data management options. It is pretty simple, but allows an administrator to insert, delete and modify the data on the RDF store. There also an import option for files and RDF ontologies and data via file.

Finally, Talia Tutorial offers services for protocols in the semantic world, like OAI-PMH.

The "sources" frontend

We call objects of interest in a digital libraries "sources". These correspond, more or less, to "subjects" in our RDF triples.

Sources list

This front-end was designed as an interface to see a list of those subjects and have the possibility to see expanded data on one subject of your choice.

The sources are orginized in a pretty standard paginated table, the only point of interest is that, if a source has files, it will have tiny icons in the "data" column. Clicking on one of them shows a preview.

It is also possible to filter results using the Filter. Interesting thing about the filter is that it is creating using ontology information from the RDF store. If Sesame is used (see the tutorial), it will use inference to add more categories to the filter.

Source page

In the image we can see "http://localhost:3000/page/Dante_Alighieri" as the chosen source, and the list of "source properties" we know about it. Source properties are of course, objects (called "values" in the table) linked to our subject via predicates ("properties") at one step distance. You can see in the example that are also considered properties predicates to another source that use our own source as object. These are called "inverse" properties. For example, if "Dante Alighieri" wrote "La Divina Commedia", it is also true that "La Divina Commedia" had "Dante Alighieri" as author.

Properties that are not "literals" (strings, numbers) represent other sources. The properties table is human readable, so those kind of values are expanded a little so that instead of a URI for the source you see a name and a link to the relative page.

Some properties, like "description", "date" and , as we will see in a moment, files can be given more space, as you can see in the image.

Source files

Looking at "http://localhost:3000/page/Byzantine_art" we can see an example of how files are treated. A source is related to its own file via a talia:hasFile predicate, not shown in the properties table because it is handled internally and a file, while being a source in itself, is never shown on its own.

"Images" are shown as a small gallery in their own section. Each image can be enhanced by clicking on it and opening a flash image viewer application. You will need to have the latest flash plugin installed in your browser to see it.

The "layers" tab in the image viewer is part of the advanced Annotations feature, that we will be discussing later.

"Data records" lists non-image files linked to the current source. Depending on their type, some of those can have their own representation after this section.

In our example, ByzantineArt?-wikipedia.html, is an extract from, easily enough, Wikipedia. It is an html file, which can be easily included in our own page. And it is, under "Text". There are some interesting things that can be done with html and text files, as we will see when discussing Annotations.

Widgets

Widgets are small components that can be included in our application to add functionalities. We are all familiar with the ever-present google map. Talia Tutorial has some of these -we could actually have gotten a little overzelous with their use, I have to admit- as examples of showing interesting interpretations of our data.

Most widget are in the home page, because they work best with aggregate data for more than one source. The source image viewer from the "Source page" that we saw earlier is the only exception.

Filter

Already seen in the homepage description, it used ontology information to create a tree of "types" for the sources we can filter with.

Timeline

Some of the demo data has dates and historical periods, the timeline is a cool little widget that allows to see our sources based on their date informations if any.

Map

Here it is google map... I said it was ever-present, didn't I? Well, a lot of the demo sources have geographical information, so showing thei spacial distribution was a pretty obligatory step.

Images preview

To be honest this is here mostly because it's cool-looking. It is also a very good way to get a quick glance of the Demo data from a purely "picturial" point of view.

Annotations

Coming soon...

The "box view" frontend

Coming soon...

The administration interface

Coming soon...

The services

OAI-PMH

This is the short version for the mouthful-ish "Open Archives Initiative Protocol for Metadata Harvesting". Liberally copy-pasting -err... I mean "citing"- from Wikipedia, it is a protocol used to harvest (or collect) the metadata descriptions of the records in an archive so that services can be built using metadata from many archives.

The Talia Tutorial Application acts as a client for OAI-PMH at "http://localhost:3000/oai" (assuming you are seeing the local application from the tutorial). Pointing your browser at the address you should see an xml response like this:

<?xml version="1.0" encoding="UTF-8"?>
<OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd">
  <responseDate>2010-11-26T15:28:24Z</responseDate>
  <request>http://localhost:3000/oai</request>
  <error code="badVerb">Value of the verb argument is not a legal OAI-PMH verb, the verb argument is missing, or the verb argument is repeated.</error>
</OAI-PMH>

Ok, this is an error, but it is the _correct_ error, as we didn't provide a proper GET request. If you want to play with the service responses try these requests:

• http://localhost:3000/oai?verb=Identify
• http://localhost:3000/oai?verb=ListMetadataFormats
• http://localhost:3000/oai?verb=ListRecords&metadataPrefix=oai_dc

If you have a public Internet address for your installation of the Tutorial application, you can explore/validate the service using this online validator: http://re.cs.uct.ac.za/. You can always use the Demo site, http://islamicart.muruca.org to try this, using http://islamicart.muruca.org/oai as the OAI baseURL.

Linked Open Data

Coming soon...