Notes on:

A Guide to Enhancing Access to Art and Material Culture Information

by Elisa Lanzi

http://www.getty.edu/research/conducting_research/vocabularies/introvocabs/

Importance of documenting art and material culture:
Contextual Information – pictures cannot speak for themselves, give factual information about the maker, context etc.

TEN POINTS of Focus

1. Opening new worlds to new audiences – broadening interests
2. Telling a story – giving narrative and meaning
3. Being multidisciplinary – reaching out to a wider audience
4. Being international – reach even further and gaining new perspectives
5. Focusing on an “act of creation” – where does the piece come from
6. Resource for education and entertainment – visual material as a learning tool vs. commodity
7. Having a point of view – subjectivity as a forum for establishing context
8. Recording by a variety of media – pooling information from different sources for richer context
9. Scientific analysis or quantification – only if you really have to…
10. Art information is complex – it helps to understand the language!

Four major approaches to documentation of art and material culture have evolved over time: archival, library, museum, and visual resources.

The archival approach involves the arrangement and description of records, personal papers, and manuscripts. Access, in-hand, groups of objects

The library approach involves the cataloging and classification of books and other published textual materials. This tradition is also known as bibliographic cataloging and classification. In-hand, individual objects, subject access, data sharing.

The museum approach involves the documentation of museum objects (e.g., works of art, artifacts, and specimens). Complex, unique objects, progressive in giving internet access to artifacts.

The Visual Resources approach involves the cataloging, classification, and indexing of images. Giving access to visual information, adopting the museum model of image as object, non-standardized classification systems.

STANDARDS

“represent professional consensus on best practice. The process that produces standards typically brings together knowledgeable practitioners to codify a reasonable body of practice based on a wide range of experiences.”

There are four types of data standards designed to improve access to your information.

1. Data Structure Standards define the categories into which information is to be divided.

ex. CDWA http://www.getty.edu/research/conducting_research/standards/cdwa/

2. Data Communication Standards – the language used by a data structure system.

Ex. MARC

3. Data Content Standards are guidelines that govern the order, syntax, and form in which the data values are entered into the structure categories.

4. Data Value Standards control the vocabulary in the data structure fields.

All of these are important to preserve the integrity of information, provide access by establishing quality, consistency and compatibility.

MARC – Machine Readeable Cataloguing

http://www.loc.gov/marc/umb/

AACR2 – Replaced with RDA (Research Description and Access)
- focuses on relationships between objects
- creates the same MARC records as AACR2 but presents them in a slightly different way.

AARC2

- title pages are an important source of information/if one is not available the colophon may be a source of information
- important to determine a single form for name, subject ets to cover all possible varioations for the same entry (name changes or various spelling)
- 8 areas of description: IFLA (International Federation of Library Associations)
International Standard Book Description – created to standardize cataloguing methods around the globe
- if no title exists, the chosen title is put in [made-up title]
- punctuation – cataloguing conventions, very important
… – omisssions
/ – means “statement of responsibility” or what is stated on the title page (transcription)
1.1B – Title proper, transcribe as appears on the title
subfields $a – title proper (transcribe as it appears, word order, spelling – not capitalization – onkly first word and proper names
$b – subtitle
$c – ramaineder of the title page/statement of responsibility

MARC – access points

010 – Library of Congress control number. replace dashes with 0
020 – ISBN, no dashes, only if it exists in the original source
040 – Who is responsible for the cataloguing
subfield a. who did the cataloguing (ex. Library of Congress – DLC)
b. language
c. who created the marc record
d. add ryerson’s symbol CaOTR (Canada Ontatio Toronto Ryerson)

245 – title statement (NR) non repeatable
indicator 1 – title added entry
- 0, means there is no 100 filed in the record
- 1, there is an author
indicator 2 – nonfiling characters
- number of nonfiling characters, as in “The ” 4 non filing characters

246 – a way to add access points to various titles, if the title on the spine is different or various spellings are used… see list of indicators
indicator 1
- 1 will be distrayed, indexed for search
- 3 will not display
indicator 2
- type of title
- #, 1, 4, 8

250 – Edition Statement (NR)
- $a edition statement

Main Entries – 100, 110
100 – Author, main entry
if more than three authors, title becomes the main entry
110 – Corporate body

300 – Physical Description (R)

Added Entries – 700, 710, 711
when multiple authors, co-authors, editors
710 – corporate authors

600 – Subject
910 – name of cataloguer

Catholic New Times – digitizing and cataloguing images. Science & Technology Folder.

Scanning Images- scan b&w images in colour and convert in CS3. Different monitors, papers, output devices will generate different results.

Four types of scanners – drum, flatbed, film & transparency or a (specialized) digital camera. Some are hybrid scanners, like Imacon. CCD & CMOS technology is very precise now and therefore can generate very high resolution images.

Networks, System Architecture, Storage – beyond the scope of collection manager, professional should be hired (and develop this strategy witht the commitee). Backup is VERY important and very expensive!

Online and Offline Storage – cd/dvds are completely useless for archiving data because they will brake, scratch, degrade etc… use external hard-drives (5 years), magnetic tapes (5-7 years). Make two copies of everything, every x number of years.

Wired – The Long Tail By Chris Anderson

http://www.wired.com/wired/archive/12.10/tail.html

Apple & Adobe (Bridge) are both using matadata schemas, but no controlled vocabulary.

Quality Control – needs to be worked into the process of digitization. Metadata also needs to be controlled. Spelling is too important; if you can’t find it, you may as well not have it. 

Intellectual Property – security and access policies need to be in place to protect the work in your collection. A record of the rights needs to be part of the metadata, for each item.

 Trina:

amazon mechanical turk

Accessioning system – numbering things as they come into the collection, versus a classification system (dui decimal system.)

OCLC – Changing Metadata Creation Processes Program

RLG Programs Descriptive Metadata Practices Survey Results

Introduction to Image Access

MARC

Expert Searching Techniques 

Trina:

Use creative words when keyword searching, use other fields beyond keyword, know the jargon of the discipline you’re searching, use the help feature.

The computer doesn’t know the difference between a venetian blind and a blind venetian

Boolean operator – AND, OR, ( ) …

http://www.csa.com/help/Search_Tools/boolean_operators.html

Controlled vocabulary – http://www.controlledvocabulary.com

David:

How we see colour – http://www.accessexcellence.org/AE/AEC/CC/vision_background.html

Gamut – actual range of colour we can create in a colour space (CMYK, RGB)

Colour space is device dependent, therefore each screen will project the colour numbers differently. A colour profile acts as a translator.

Use Adobe 1998b because it has a wider gamut than sRGB.

Bit Depth/Dynamic Range – use 16 bit whenever possible for digital capture, you will get a lot more detail in the shadow areas.

All digital cameras capture an analogue image, the camera them converts the analogue to digital using a converter.

Resolution = density of elements. Dependent on the viewing device or distance from image. The sample rate is a more accurate in determining image quality. (Scan at maximum resolution.)

72 dpi is not a standard screen resolution, it varies between devices. Each printer has its own optimal resolution,  300 dpi is not a standard.

Compression – 2 classes

Lossless compression- retains actual information from time of creation. Ideal especially for original scans and master images. (eg. TIFF)

Lossy compression – takes advantage of the inherent deterioration of the device it is viewed on. Sufficient for derivative files but not ideal as it uses average values and resampling. (eg. JPEG) Compression artifacts really show up in print.

File Formats – are problematic because they have not been around for long enough. Will you be able to read a TIFF in 50 years? Format obsolescence is a concern, as is material degradation. The only way to deal with this is to periodically transfer the data to newer formats. The hardware to access outdated formats is equally important!

RAW file format does not exist – only a proprietary file format  that needs a converter to be able to read it.

http://www.getty.edu/research/conducting_research/standards/introimages/index.html

‘Media Decay’ – I like this term, ever played Tokobot!?! The digital image, made up of nothing more than numbers is only as real as the computer that can interpret it. And I though all those floppy discs from high school were going to preserve my masterpieces forever. Metadata… if you say it is what it is… I guess I have to believe you. The computer can’t tell the difference. How much longer until AI? If a picture speaks a thousand words, how many will suffice to retrieve it by a thousand different people?

Interoperability, I should use this word more often…

Standards, to each his own.

Rights management, intellectual property – Physical originals can pose a greater problem than digital originals as the issue of rights management is likely addressed at the time of creation.

Metadata Crosswalks – a way to link systems of metadata from multiple collections using different metadata schemas.

digital image – changes in voltage, raster – made up of lines
sampling is a way of converting analog info to digital info. The bigger the sample, the more accurate the image.

dave is funny