Ricerca in OJAX attraverso la Firefox searchbar

From SPARC ML, excerpt from Judith Wusteman on OJAX search features:

“The latest version of the OJAX metasearch service is now tightly integrated with the Firefox 2 / IE 7 search feature. Repositories powered by OJAX can now be searched directly from the browser search bar, with no need to download or configure extensions or plugins.

OpenSearch Discovery – OJAX now supports OpenSearch Discovery. When a user visits any OJAX repository search page, their browser will automatically detect that the repository can be searched via OpenSearch and will offer to add the repository to the set of search plugins installed in their browser.

Once added, OJAX can be selected in the browser search bar and repository searches can be performed at any time without having to navigate back to the repository search page.

Search Suggestions (Auto-Completion) – Firefox 2.0 has extended the OpenSearch protocol to allow auto-completion in the browser search bar (which Firefox calls “Search Suggestions”). OJAX is one of the first search engines to implement the server side of this extended OpenSearch protocol, after Google, Yahoo! and Answers.com. It is probably the first open source search engine to do so.

Because OJAX 0.6 uses the increasingly popular OpenSearch 1.1 search descriptions, the new features described above require the latest versions of the two major browsers: Firefox 2.0 or Internet Explorer 7.0.”

Read full article of J. Wusteman and Pádraig O’hIceadha Using Ajax to Empower Dynamic Searching. Information Technology and Libraries (ITAL) Vol. 25, No. 2, June 2006, pp 57-64 and my blog post (italian language) OJAX: AJAX per il metasearching.

Programmable Web: il web 2.0 alla prova della creatività

Programmable Web è un sito dedicato ad alcune tra le più intriganti e complesse funzionalità introdotte dal web 2.0, API e mashup in testa:

“ProgrammableWeb is where you can keep-up with the latest mashups, what’s new and interesting with Web 2.0 APIs, and the Web as Platform in general. The core of the site is the blog and the 3 dashboards: Home, Mashups and APIs. All dashboards are updated daily.” (dall’About del sito)

Significativo il grafico della mashup timeline in home page, che mostra l’incremento subito dai mashup negli ultimi sei mesi e dà la misura concretamente di quanto l’imperativo reillyano (da O’Reilly, creatore del brand web 2.0) del content-remix sia quanto mai attuale e praticato in rete.

I mashup, che fruiscono spesso delle API per poter combinare risultati da applicazioni diverse in modalità inedite, sono realizzati sulle più diverse piattaforme web, benché si noti una certa ricorrenza di siti come Flickr, Yahoo o Amazon, che, per la natura aperta delle loro interfacce di interrogazione, si offrono come banchi di prova privilegiati per sperimentare web service sempre più interattivi e dinamici.

Non mancano in Programmable Web esperimenti rivolti al mondo delle biblioteche; in particolare sono da segnalare le API della Library of Congress relative ai protocolli di ricerca http sui quali da tempo questa lavora: SRU/SRW.

Assolutamente necessaria, infine, e utilissima, la scheda tecnica per ogni applicativo proposto, con i dati e le informazioni relative ai prodotti e ai loro autori, suddivisi nelle macro-categorie Overview, Functionality, Security, Support, Signup e Licensing.

Google lancia la “Library Catalog Search” in Google Book

Qualche giorno fa sul blog di Google è stata lanciata la notizia di una nuova funzionalità per Google Book: la Library catalog search, che permetterà, interrogando il motore di ricerca nella sua versione dedicata ai libri, di ricomprendere nei risultati anche gli item dei cataloghi delle biblioteche. Ecco un estratto dal post di Bruno Fonseca, Google Brazil:

“Today, we’re launching the Library Catalog Search feature in Google Book Search, designed to help casual readers and bookworms everywhere find gems in the libraries around the world. Queries on Google Book Search will automatically include results from library catalogs when appropriate.

Each result includes a “Find Libraries” link to help readers find libraries that hold the book — ideally a library nearby, or if need be, a library far away. For example, after reading Martin Gardner’s book Fads and Fallacies, I wanted to follow up on Immanuel Velikovsky’s books about scientific explanations for biblical miracles. Clicking on the “Find Libraries” link for Velikovsky’s Worlds in Collision, I found that a copy was available in the University of Sao Paulo library.

This is true of many types of books in countries all over the world — my colleagues in Google’s Zurich office tell me about being able to find Harry Potter and the Chambers of Secrets for their nephews and nieces. In many cases, it’s even possible to click through to the local library and reserve the book.

For this feature, we have worked with more than 15 library union catalogs that have information about libraries from more than 30 countries, as well as with our colleagues working on Google Scholar (which includes a similar feature just for scholarly books).”

I cataloghi di biblioteche interrogati risultano essere sottoscrittori di “Find in a library” di OCLC (la spiegazione della feature, nell’Help di Google Book, è: When you click on “Find it in a Library”, we send you to the OCLC Worldcat where you can enter your zip code and find a local library that has the book. These links appear for books that are digitized as a part of the Library Project), ma, mentre alcuni di essi offrono accesso diretto alle proprie interfacce (per es: la Biblioteca Nazionale Australiana o le Biblioteche Danesi), per altri il comando di ricerca degli item porta direttamente all’interno della schermata di OCLC. Interessante la possibilità di selezionare il radio button perché Google Books cerchi solo all’interno dei cataloghi online.

ATTENZIONE: ad oggi per usufruire di Library catalog search occorre interrogare Google Book nella sua versione inglese: per gli utenti italiani è necessario pertanto impostare l’inglese come lingua del browser.

Danish search interface à la Web 2.0

Per la Mashing up the library competition istituita questa estate da Talis, riportiamo la segnalazione di SUMMA, l’OPAC con cui la Statsbibliotheket danese si è iscritta al concorso. Questo il comunicato di Jens Hofman Hansen del Summa Team:

“[Summa] searches across a given number of data sources and presents the output for the user in a single search result. Summa also adds extra value to search results by pulling information from external web service based information sources, such as author bibliographies or images collections.

Summa’s philosophy is simple: development must be based on user needs and user behaviour. One important usability observation is that library users know little or nothing of the valuable resources and collections that many libraries maintain outside the library database. To address this issue, Summa mashes up information from a wide array of external web resources.

For instance, Summa includes book covers and editorial reviews from Amazon.com in search results. Also, music sound bites are collected from the Danish online music site netmusik.dk and are integrated where possible. Summa even includes information about the library’s academic experts in the search result allowing for swift contact.

New and useful features can easily be added: If, for instance, a “Did you mean” web service is released somewhere on the web, and if it is better than the one we built, we can easily integrate it to replace our own, thanks to the modular design of Summa.

In the near future, our intention is to turn Summa into a Open Source project allowing any library to freely use and benefit from it. The code is not quite ready for that yet, but we are working on it. Read more about the features at
http://www.statsbiblioteket.dk/summa/about.jsp and find all the technical details in our white paper at http://www.statsbiblioteket.dk/publ/summaenglish.pdf.”

L’About offerto sul sito esplicita in maniera chiara le funzionalità di SUMMA ed è utile per acquisire familiarità con lo strumento. Le feature che più mi hanno colpito, oltre a quelle menzionate sopra, sono senz’altro la Fully automatic classification per i cluster e l’uso di Ajax, ormai sempre più diffuso nelle applicazioni bibliotecarie come nei motori più classici, per l’auto-completamento dei termini di ricerca.

Ecco la lista completa di tutto ciò che è possibile ricercare e di come è possibile farlo:

• Integrated search and sort by relevance
• Extra information that adds value and meaning
• Availability shown in the search results
• Did you mean…?
• Suggestions
• Find an expert
• A picture is worth a thousand words
• Resource basket
• Multiple languages
• Other users who borrowed this, also borrowed…
• Editorial reviews from Amazon
• Display of closely-related works
• Clusters – fully automatic classification
• More information

Buone vacanze!

The Geek Librarian va in vacanza per due settimane; rientrerà alla base all’incirca alla fine di agosto per nuove ed emozionanti avventure…

Buon agosto a tutti e a presto!

FRBR secondo OCLC e Library Thing

Una delle funzionalità più interessanti sviluppate da OCLC è xISBN, “a library web service that supplies International Standard Book Numbers (ISBNs) associated with individual intellectual works represented in the OCLC WorldCat database. Give it an ISBN, and it returns a list of associated ISBNs.

Developed by the OCLC Office of Research, this experimental, proof of concept project:

• explores the value of WorldCat for intellectual work-related data
• tests the effectiveness OCLC Research’s FRBR Work-Set Algorithm, a freely available algorithm developed by OCLC Research to implement the FRBR model (a model for grouping bibliographic records)
• helps OCLC and other agencies experiment with web services-based query optimization
• provides basic functionality on which the OCLC Research library lookup bookmarklet is based.”

Library Thing, sito di social tagging dedicato al mondo dei libri, propone qualcosa di molto simile, e anche, se possibile, di più emotivamente evoluto: thingISBN, “LibraryThing’s answer to xISBN. Under the hood, xISBN is a test of FRBR, a highly-developed, well thought-out way for librarians to model bibliographic relationships. By contrast, thingISBN is based on LibraryThing’s everyone a librarian idea of bibliographic modeling. Users combine works as they see fit. If they make a mistake, other users can separate them. It’s a less nuanced and more chaotic way of doing things, but can yield some useful results.”

Per provare thingISBN, clicca qui (per cambiare ISBN, basta sostituirlo nell’URL), mentre qui trovi una comparazione tra xISBN e thingISBN.