Week 4: META DATA GALORE!

Witten:

Bibliographic systems:
1. Finding: locate item with known info.
2. Collocation: finding other things related to this item, such as other books the author has written.
3. Choice: A list of other available options arranged graphically (other editions) or topically (similar subjects).

Bibliographic entities
1. Documents: analog or digital form
2. Works: inhabitants of bibliographic universe: can have different forms, mediums and editions
3. Editions: multiple publications, revisions. Electronic form is usually a version, release or revision not an edition
4. Authors: Can have different names, numbers of authors, versions of name, can be a group or entity like the LOC. The LOC provides controlled vocabulary and standard names to clear up any problems.
5. Titles: straight forward attribution of the work
6. Subject: key-phrase extraction or key-phrase assignment. LOC uses a controlled vocabulary (LCSH) to standardize subject assignment.
7. Subject classification: organizing books on the shelf by subject. LC call number system does this automatically, as does Dewey. This allows the user to physically browse the shelves and gain access to the full content to choose materials.

Bibliographic Metadata
1. MARC: Machine Readable Catalog: using numerical tags, organizes info
2. Dublin Core: same concept, but simplified without all the numerical tags.
3. BibTex: prefered by scientific and technical authors who use a lot of mathematical structures.
4. Refer: basis of EndNote

Metadata for images, etc
1. Tagged Image File Format: TIFF. Used for images. Tags describe elements of the image, such as size, colors, etc.
2. MPEG-7: multimedia content description interface. Tags describe the data in the file.

Extracting Metadata
1. Reading the document helps one understand it.
2. Markup languages give clues as to the content without reading the full document: XML, etc.
3. Extracting information: generic entity extraction can pull information out using clues in the text
4. Bibliographic references: provide information in the form of citations. A citation index, such as a 'works cited' page organizes these.

Setting the Stage
This article covers the basics of different types of metadata systems already covered. However, what it does cover is how metadata, the structure of metadata, and the organization of metadata are important to extend to museums and archives, especially as those institutions move to digital resources. The use of metadata is second nature to libraries because they've been doing it for generations now. The analog metadata can easily be transcribed into digital systems when the items are digitized. However, archives and museums have resisted using metadata and instead use 'finding guides' to locate their items. This precludes amateur users from independently finding items, and it precludes digitization. This is a situation that needs to be rectified in order for these institutions to move into the digital age.

Border Crossings
This article looks back on the past 10 years of the efforts of the D-Lib DCMI management team. It talks about how necessary it is to create a universal and international system of metadata management. As information and metadata become more digitized and accessible over the Internet, the more important it is for the systems to be able to speak to each other. An overarching goal of libraries has always been for them to be able to share information with each other and make materials as accessible to patrons every where as possible. The Internet provides the infrastructure to make that happen, but in order to work, all the different systems must be able to communicate. This article focuses on the metadata aspect of that. I found especially applicable the comparison to the rail changes between Mongolia and China. Two given libraries ought not have the animosity of centuries that those nations do, so they certainly shouldn't have the level of complexity of communication that they do.

Puppy picture!
A tired puppy is a good puppy.