Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Reimagining Serials handout: BIBFRAME Exercise

•
•
NASIG
NASIG

Starting in late 2014, the UCLA Continuing Resources Study Group began studying semantic web concepts in general and the BIBFRAME vocabulary specifically. Our goal was to understand how things would work for serials and integrating resources in the new linked data environment, to contribute to the discussion and development of linked data models for bibliographic resources, and (if possible!) to try something new. Using just a few free tools, we were able to convert records from MARC, create linked data graphs, and take the first steps in setting up a demonstration triple store. You can too! See the poster http://www.slideshare.net/NASIG/great-idea-reimagining-serials-small-steps-toward-a-linked-data-future Kevin Balster ERM/Continuing Resources Metadata Librarian, UCLA Kevin Balster is the ERM/Continuing Resources Metadata Librarian at UCLA. He received his MLIS from UCLA in 2011, and has been cataloging serials and maintaining e-resources since 2012.

1 of 4
BIBFRAME Exercise 2/18/2015 Review: 1. How to read a triple Reminder: A linked data triple consists of subject, predicate [or property], object. The subject is represented by a URI or blank node, the predicate is represented by a URI, and the object may be a literal value [a string of characters]; a URI; or a blank node. The following triples could be read: • The resource (OCLC 899586104) is a “Work” [has type “Work”] • The resource is “Text” [has type “Text”] • The resource has the authorized access point, “Journal of Parkinson’s disease and Alzheimer’s disease” • The resource is in “English” [has language “English”] 2. BIBFRAME.org Vocabulary a. Reminder: A vocabulary defines classes and properties in a given domain. In the case of BIBFRAME, the core classes are creative work, instance, authority, and annotation. b. Call up BIBFRAME. Brief orientation to Model, Category, and List views: http://bibframe.org/vocab/
Exercise Part 1: Export MARC record from OCLC [Alternative: save MARC record from Voyager] 1. In OCLC Connexion, use the top menu to choose Tools – Options; then find the “Export” tab 2. Click on “Record Characteristics” to see your options. For this exercise, you will be using MARC21; but you also have the option of exporting the record in XML. The character set should be UTF-8 unicode. Click “ok” to accept the options. 3. If you have never exported a record before, you should set up a file folder for exported records. Click on “create.” Accept “file” as the export destination and click “ok.” 4. Use the window next to “look in” to navigate to the drive and the folder you wish to use. If you would like to create a sub-folder, click the new-folder icon and name the folder. 5. Make sure that the destination path is highlighted; then click “apply” and “close” at the bottom of the options window 6. To export an OCLC record, retrieve the record and press F5. **VERY IMPORTANT** • If you are exporting records for other purposes, for example, Voyager cataloging, you MUST change the path back to the default when you finish your BIBFRAME work • If you are going to access MARCEdit via VDI, it is best to avoid saving the file to your desktop as the Desktop location in VDI is not your personal desktop. It is probably easiest to re-find the file on your H: drive or some other shared drive. • By default, OCLC records have an extension of “.dat” ; For convenience, you should change this extension to “.mrc” • The file name stays the same whether you export one record or 50 records. OCLC just keeps on adding records to your file. If you are used to working with MARCEdit, you can easily split all the records into separate files later. But if you do not want to learn this MARCEdit function, then re-name your file after exporting it, e.g., to the title or OCLC number. 1: location for file 2: name file
Part 2: Convert MARC record for Bruin standard to RDF/XML 1. Start MARCEdit. From menu, choose MARCNext; then choose “BibFrame Testbed.” 2. Fill in the form Source File: Click on the file icon. Click button/menu; change from “MARC File” to “All Files.” Navigate to the folder where you saved the file and highlight bruinstandard.mrc >>TIP: Today you will be using a record exported from OCLC. By default, OCLC records have an extension of “.dat”—but for convenience, you can change this to “.mrc” >>TIP2: You could also use a Voyager record. Just retrieve a record in the Cataloging client and save it to the folder of your choice. Change the extension from the Voyager default of “.bib” to “.mrc” Save File: Click on the green-arrow file icon. Name a file name, e.g., Bruinstandard.xml File Type: Choose MARC Base URI: Ignore Data Serialization: Choose RDFXML …Then click “Process” Result:
Part 3. View the results 1. Open the file in Notepad++ a. Find the folder where you saved the XML file. b. Highlight the .xml file. Right-click; select “edit with Notepad++” c. Copy the results >>TIP: Ctrl-a highlights all the text; Ctrl-c copies it 2. In a web browser, open the W3C Validation Service page: http://www.w3.org/RDF/Validator/. 1) Click on the tab for “check by direct input” 2) Delete the default text; paste the file that you copied from Notepad++ 3) Choose “triples and graph” or “graph” 4) Click on “parse RDF” 5) Scroll to see results 1 3 2 4

More Related Content

Reimagining Serials handout: BIBFRAME Exercise

  • 1. BIBFRAME Exercise 2/18/2015 Review: 1. How to read a triple Reminder: A linked data triple consists of subject, predicate [or property], object. The subject is represented by a URI or blank node, the predicate is represented by a URI, and the object may be a literal value [a string of characters]; a URI; or a blank node. The following triples could be read: • The resource (OCLC 899586104) is a “Work” [has type “Work”] • The resource is “Text” [has type “Text”] • The resource has the authorized access point, “Journal of Parkinson’s disease and Alzheimer’s disease” • The resource is in “English” [has language “English”] 2. BIBFRAME.org Vocabulary a. Reminder: A vocabulary defines classes and properties in a given domain. In the case of BIBFRAME, the core classes are creative work, instance, authority, and annotation. b. Call up BIBFRAME. Brief orientation to Model, Category, and List views: http://bibframe.org/vocab/
  • 2. Exercise Part 1: Export MARC record from OCLC [Alternative: save MARC record from Voyager] 1. In OCLC Connexion, use the top menu to choose Tools – Options; then find the “Export” tab 2. Click on “Record Characteristics” to see your options. For this exercise, you will be using MARC21; but you also have the option of exporting the record in XML. The character set should be UTF-8 unicode. Click “ok” to accept the options. 3. If you have never exported a record before, you should set up a file folder for exported records. Click on “create.” Accept “file” as the export destination and click “ok.” 4. Use the window next to “look in” to navigate to the drive and the folder you wish to use. If you would like to create a sub-folder, click the new-folder icon and name the folder. 5. Make sure that the destination path is highlighted; then click “apply” and “close” at the bottom of the options window 6. To export an OCLC record, retrieve the record and press F5. **VERY IMPORTANT** • If you are exporting records for other purposes, for example, Voyager cataloging, you MUST change the path back to the default when you finish your BIBFRAME work • If you are going to access MARCEdit via VDI, it is best to avoid saving the file to your desktop as the Desktop location in VDI is not your personal desktop. It is probably easiest to re-find the file on your H: drive or some other shared drive. • By default, OCLC records have an extension of “.dat” ; For convenience, you should change this extension to “.mrc” • The file name stays the same whether you export one record or 50 records. OCLC just keeps on adding records to your file. If you are used to working with MARCEdit, you can easily split all the records into separate files later. But if you do not want to learn this MARCEdit function, then re-name your file after exporting it, e.g., to the title or OCLC number. 1: location for file 2: name file
  • 3. Part 2: Convert MARC record for Bruin standard to RDF/XML 1. Start MARCEdit. From menu, choose MARCNext; then choose “BibFrame Testbed.” 2. Fill in the form Source File: Click on the file icon. Click button/menu; change from “MARC File” to “All Files.” Navigate to the folder where you saved the file and highlight bruinstandard.mrc >>TIP: Today you will be using a record exported from OCLC. By default, OCLC records have an extension of “.dat”—but for convenience, you can change this to “.mrc” >>TIP2: You could also use a Voyager record. Just retrieve a record in the Cataloging client and save it to the folder of your choice. Change the extension from the Voyager default of “.bib” to “.mrc” Save File: Click on the green-arrow file icon. Name a file name, e.g., Bruinstandard.xml File Type: Choose MARC Base URI: Ignore Data Serialization: Choose RDFXML …Then click “Process” Result:
  • 4. Part 3. View the results 1. Open the file in Notepad++ a. Find the folder where you saved the XML file. b. Highlight the .xml file. Right-click; select “edit with Notepad++” c. Copy the results >>TIP: Ctrl-a highlights all the text; Ctrl-c copies it 2. In a web browser, open the W3C Validation Service page: http://www.w3.org/RDF/Validator/. 1) Click on the tab for “check by direct input” 2) Delete the default text; paste the file that you copied from Notepad++ 3) Choose “triples and graph” or “graph” 4) Click on “parse RDF” 5) Scroll to see results 1 3 2 4