Gartner - Metadata for digital libraries: state of the art and future directions
This piece is one of the most well-written assigned readings I have come across. Finally, in plain English, a librarian has articulated why metadata schemes employing XML may be able to interoperate so well:
"XML has the crucial feature that a marked-up file can embed others encoded in different XML schemas directly within it (if, of course, it follows a schema that is designed with this function in its specification). This is made possible by a feature known as XML namespaces."
I was also very grateful for the author pointing out one technical problem which may arise involving namespace definitions. This is where XML schemas within the METS framework incorporate subsidiary schemes whose namespace definitions conflict with those in METS.
So very often librarians who are not IT-literate become dependent upon things being interoperable immediately and automatically, that we lack the sophistication to troubleshoot.
This piece also finally identified the mystery organization behind METS and MODS, namely, the MARC Standards Office.
Muddiest Point: Who are the other organizations involved in promulgating standards? Is there enmity or competition between them, or is there a sincere commitment to openness?
Gilliland - Setting the Stage
This article identified important functions of metadata, such as certifying the authenticity and degree of completeness of the content, and providing some information that might have been provided in a traditional, in-person reference or research setting.
Muddiest Point: Because metadata performs such important functions, should there be a code of ethics for people entering the metadata? If the object is something extremely important, like a volume from the Vatican, might there be the opportunity for an unethical monk to misrepresent the object and scurry it away into another hidden category?
Weibel - Border Crossings
I absolutely loved Weibel's analogy of changing train gears on the tracks from Beijing to Siberia, with how metadata schemes encounter interoperability challenges. As someone who has taken that train, I can fully relate!
Friday, September 21, 2012
Thursday, September 20, 2012
Week 4 - Reading Notes
Lesk - Chapters 2.1, 2.2, 2.7, 3
From Chapter 3.7, I especially liked learning about the Gallica Collection available at the Bibliothèque Nationale Française, because I speak and read French. After discovering this, I went online to Gallica itself, and was amazed to see the editor's proof of Les Fleurs du Mal, a very important book of French poetry from the 19th century . . . with the editor's and author's own handwritten proofs on the margins of the scanned version! Amazing. From a scholarship standpoint, a researcher no longer has to travel to Paris to read this vital copy; s/he can read it from the comfort of their own laptop. For an example, see http://gallica.bnf.fr/ark:/12148/btv1b86108314/f42.image
Arms - Chapter 9
On Panel 9.3, I was surprised to learn that EAD is in fact a sophisticated type of DTD. I have been using EAD for many years and have used it to mark up more than 31 linear feet of archival material. One day, I hope to see the actual code for EAD, because it is open-source and I hope to help edit its code.
Lynch - Identifiers and Their Role in Networked Information Applications
Muddiest Point: What are the politics behind competing standards? It seems that one group, like the Internet Engineering Task Force, will first try to develop a persistent identifier, like URN, only to be usurped by another group like OCLC who is trying to push PURL. Then, much later, NISO will push SICI, while AAP and CNRI will now push for DOI.
Who is right and who is wrong? Who is stepping on the other's toes? Why are there so many competing handles?
Paskin - Digital Object Identifier System
This article greatly clarified for me the relationship between DOI, URLs, and the digital object itself. I believe Figure 1 brought it home for me by showing that where the URL for the digital object changed, the DOI itself did not have to alter. The independence of a DOI from an object's URL guarantees its persistence over time.
From Chapter 3.7, I especially liked learning about the Gallica Collection available at the Bibliothèque Nationale Française, because I speak and read French. After discovering this, I went online to Gallica itself, and was amazed to see the editor's proof of Les Fleurs du Mal, a very important book of French poetry from the 19th century . . . with the editor's and author's own handwritten proofs on the margins of the scanned version! Amazing. From a scholarship standpoint, a researcher no longer has to travel to Paris to read this vital copy; s/he can read it from the comfort of their own laptop. For an example, see http://gallica.bnf.fr/ark:/12148/btv1b86108314/f42.image
Arms - Chapter 9
On Panel 9.3, I was surprised to learn that EAD is in fact a sophisticated type of DTD. I have been using EAD for many years and have used it to mark up more than 31 linear feet of archival material. One day, I hope to see the actual code for EAD, because it is open-source and I hope to help edit its code.
Lynch - Identifiers and Their Role in Networked Information Applications
Muddiest Point: What are the politics behind competing standards? It seems that one group, like the Internet Engineering Task Force, will first try to develop a persistent identifier, like URN, only to be usurped by another group like OCLC who is trying to push PURL. Then, much later, NISO will push SICI, while AAP and CNRI will now push for DOI.
Who is right and who is wrong? Who is stepping on the other's toes? Why are there so many competing handles?
Paskin - Digital Object Identifier System
This article greatly clarified for me the relationship between DOI, URLs, and the digital object itself. I believe Figure 1 brought it home for me by showing that where the URL for the digital object changed, the DOI itself did not have to alter. The independence of a DOI from an object's URL guarantees its persistence over time.
Wednesday, September 19, 2012
Week 3 - Reading Notes
Witten, Bainbrdige & Boddie: Greenstone, Open-Source Digital Library Software
The New Zealand Digital Library project's Greenstone software has features which appeal to me. First, it has interfaces in many different languages, including Chinese. Second, it has user activity logs which record every query made to its collection. This could greatly assist in building a taxonomy for a site, to choose labels and to create an index or FAQ. Third, I was impressed by the fact that the U.N. freely distributed Greenstone via CD-ROM to developing countries.
Smith et al.: DSpace - An Open Source Dynamic Digital Repository
It seems that DSpace has some major advantages over Greenstone. First, DSpace employs more standards in its protocols, such as the DublinCore metadata standard and OAI-PMH, and thus seems more interoperable. Second, DSpace is able to accommodate a complicated workflow system in an organization, with different players performing different functions (submitter, reviewer, metadata editor, etc.)
Muddiest Point: Does DSpace employ user activity logs, in order to capture user queries to refine its taxonomy and labels?
Biswas and Paul - An Evaluative Study - Special Reference to DSpace and Greenstone Digital Library
I really enjoyed the fact that this article pointed out some of the major features of DSpace, like its Lucene search engine, Handle system, and its employment of OAI-PMH. However, I was dismayed by the numerous misspellings which occurred in this article. I appreciated the fact that the authors pointed out one particular weakness of DSpace at that time: It lacks/lacked employment of the Metadata Encoding and Transmission Standard, METS.
The New Zealand Digital Library project's Greenstone software has features which appeal to me. First, it has interfaces in many different languages, including Chinese. Second, it has user activity logs which record every query made to its collection. This could greatly assist in building a taxonomy for a site, to choose labels and to create an index or FAQ. Third, I was impressed by the fact that the U.N. freely distributed Greenstone via CD-ROM to developing countries.
Smith et al.: DSpace - An Open Source Dynamic Digital Repository
It seems that DSpace has some major advantages over Greenstone. First, DSpace employs more standards in its protocols, such as the DublinCore metadata standard and OAI-PMH, and thus seems more interoperable. Second, DSpace is able to accommodate a complicated workflow system in an organization, with different players performing different functions (submitter, reviewer, metadata editor, etc.)
Muddiest Point: Does DSpace employ user activity logs, in order to capture user queries to refine its taxonomy and labels?
Biswas and Paul - An Evaluative Study - Special Reference to DSpace and Greenstone Digital Library
I really enjoyed the fact that this article pointed out some of the major features of DSpace, like its Lucene search engine, Handle system, and its employment of OAI-PMH. However, I was dismayed by the numerous misspellings which occurred in this article. I appreciated the fact that the authors pointed out one particular weakness of DSpace at that time: It lacks/lacked employment of the Metadata Encoding and Transmission Standard, METS.
Week 2 - Reading Notes
Suleman and Fox: A Framework for Building Open Digital Libraries
In this reading, I really enjoyed the authors' clear articulation of principles to guide an Open Digital Library Design:
In this reading, I really enjoyed the authors' clear articulation of principles to guide an Open Digital Library Design:
- All Digital Library services should be encapsulated within components that are extensions of Open Archives.
- All access to the Digital Library services should be through their extended OAI interfaces.
- The semantics of the OAI protocol should be extended or overloaded as allowed by the OAI protocol, but without contradicting the essential meaning.
- All Digital Library services should get access to other data sources using the extended OAI protocol.
- Digital Libraries should be constructed as networks of extended Open Archives.
However, my question is: Is OAI the acknowledged industry standard for interoperability between Digital Libraries? What is the difference between OAI-PMH and Dublin Core? Do those standards describe different things?
Arms, Blanchi, and Overy: An Architecture for Information in Digital Libraries
This piece helped give me a clear understanding of the outline of information architecture in a digital library.I especially appreciated its definition of data types, structural metadata, and meta-objects. I liked the rules which were made at the outset:
- All data is given an explicit data type.
- All metadata is encoded explicitly.
- Handles are given to individual items of intellectual property.
- Meta-objects are used to aggregate digital objects.
- Handles are used to identify items in meta-objects.
Because this article gave clear definitions to its outline, it provoked no questions from me.
Payette, Blanchi, Lagoze and Overly: Interoperability for Digital Objects and Repositories: The Cornell/CNRI Experiments
This article had a great definition for interoperability:
[I]nteroperability si defined as the ability of digital library components or services to be functionally and logically interchangeable by virtue of their having been implemented in accordance with a set of well-defined, publicly known interfaces. . . . When repositories and digital objects are created in this manner, the overall effect can be a federation of repositories that aggregate content with very different attributes, but that can be treated in the same manner due to their shared interface definitions.
My question is: What are the best interoperability protocol standards for digital libraries today? Where can I go to find such information, and information on how those standards are evolving? This article was written in 1999; where can I find new stuff?
I appreciated the ARMS' chapter which made an overview of the Web and its history, but I have read it many times now, so I don't have that many questions.
Subscribe to:
Posts (Atom)