Month: October 2014

One of our readings this week was “Classification and its Structures” (Sperberg-McQueen), which explained the potential uses for a classification scheme and described different types of classification schemes. Nominal classifications were the first to be introduced, and they appeared to be the simplest: the different classes in the scheme are “discrete classes which, taken together, subdivide the set of things being classified.” Nominal classifications are easy enough if the different classes remain distinct; however, problems can emerge when the data grows over time to such an extent that the classification scheme demands additional classes, many of which have the potential for shared characteristics.

The article described this situation as a classification scheme that allowed for “variable depth.” Sperberg-McQueen’s example of the Dewey system subdividing the class of English drama (itself already a subdivision) by period made me wonder if anyone had tried to apply a similar system to music. iTunes sorts everything into about twenty generic classes, but it is pretty clear that a label like “Alternative” is so broad that it has the potential to encompass many very divergent acts. After a little research, I found the project Every Noise at Once, which displays impressive specificity in its attempt to represent every music genre conceivable.

 

 

 

 

 

 

 

According to the website, Every Noise at Once is “an ongoing attempt at an algorithmically-generated, readability-adjusted scatter-plot of the musical genre-space, based on data tracked and analyzed for 1284 genres by The Echo Nest.” In addition to that brief description of the project, the home page also includes ins tructions for browsing the website, and although the “calibration is fuzzy,” there does appear to be some sort of order to the visual representation of the data. Although the site might not function very well as a music discovery tool, the website incorporates so many subdivisions that finding relevant acts becomes a surprisingly efficient process. Understandably, there is some overlap between genres when searching by artist, and the website does not attempt to explain the distinctions that are being made, but it is still a very systematic look at a data set that had a number of possible organizational schemes.

 

While doing this week’s readings, the article introducing metadata immediately caught my eye. I had always heard references of this abstract concept known as metadata, but I never truly understood what it was or how it could possibly be relevant in my own life. In Anne J. Gilliland’s article “Setting the Stage” from Introduction to Meta, Gilliland describes the “big picture” definition of metadata as “the sum total of what one can say about any information object—[anything that can be addressed and manipulated as a discrete entity by a human being or an information system]—at any level of aggregation.” At this point, I still was not completely positive what the purpose of metadata was. As I read further, Gilliand explains, “Cultural heritage information professionals such as museum registrars, library catalogers, and archival processors often apply the term metadata to the value-added information that they create to arrange, describe, track, and otherwise enhance access to information objects and the physical collections related to those objects.” Finally, this I could understand, since I do exactly this everyday.

A fun-fact that you may not know about me is that I run a “food-blog Instagram” account. Basically, twice a day I upload deliciously sinful food pictures onto an Instagram account, which I use solely for this purpose. In each post, I geo-tag the location of the restaurant where the food is from, tag the restaurant’s own account, use specific hashtags to describe the food, and tag the user who deserves photo credit. Every step of this process categorizes each specific picture into its own unique domain on Instagram. As seen in the image above, the picture is geo-tagged at Novel Café in Westwood and the restaurant itself is hash-tagged (#novelcafe). In every picture I also include a “hashtag overload,” of other food-blog accounts’ personal hashtags. This allows other accounts access to my photos and have photos appear in their own Instagram domain. All of these tags on Instagram contain photos compiled into an archive from all of the different users who utilized the same tags, which allows easy access to specific images by searching for the desired tag. These tags allow users to archive, arrange, describe, and track the progress and information provided in each image.

My food Instagram provides me with a glimpse of the importance of metadata in everyday life. According to Gilliland, “Medata establishes and documents the context of the content; identifies and exploits the structural relationships that exist within and between information objects; [and] provides a range of intellectual access points for an increasingly diverse range of users.” As seen, metadata is essential in supplying a resource to create structure in our digital world.

Works Cited:

Anne Gilliland, “Setting the Stage,” from Murtha Baca, ed., Introduction to Metadata (Los Angeles: Getty, 2008)

The first thing that came to mind while reading this week’s reading covering selecting, sorting, and classifying is the Dewey Decimal Classification. In Classification and its Structures, C.M. Sperberg-McQueen actually mentions this system as an example of a classification scheme that operates within n-dimensional space. The DDC “assigns class numbers in the 800s to literary works. Within the 800s, it assigns numbers in the 820s to English literature, in the 830s to German literature, the 840s to French, etc. Within the 820s, the number 821 denotes English poetry, 822 to English drama, 823 English fiction, and so on. Further digits after the third make even finer distinctions” (Sperberg-McQueen). Creating a “tree-like hierarchy”, the DDC designated the widest, most important label first (i.e. country of origin), then refines it farther and farther into a classified system.

Researching more about the DDC, I wondered why and how Melvil Dewey decided “American” would be assigned number one, “English” number two, etc. in 1876. In a 2012 post to the Association for the Library Service to Children blog, guest contributor Tali Balas Kaplan emphasizes the outdated quality of the DDC, arguing, “Successful systems have clear logic and the different pieces are connected in ways that make sense to people who’re using the system. Students may be able to navigate the numbers if you spend enough time teaching Dewey and find pieces of it, such as the 636.7 books or the 745.5 books. But the logic, the sense of it, will escape them because it’s based on criteria that are unknown or irrelevant to them” (Kaplan). Answering my own question, the DDC has also received criticism for its bias towards Anglo-American voice in its hierarchical designations.¹

This particular disadvantage of the DDC relates closely with Julia Gaffield’s article Haiti’s Declaration of Independence: Digging for Lost Documents in the Archives of the Atlantic World. The process of filtering information, “by identifying the properties relevant for such judgments of similarity and dissimilarity can make explicit a particular view concerning the nature of the objects being classified” (Sperberg-McQueen). In the beginning of her research, her associates warned Gaffield of the very few resources available on the early period of Haiti’s independence. Nonetheless, Gaffield devised a new research strategy, in which she traveled to six other countries besides Haiti, including Jamaica, Great Britain, France, the United States, Netherlands and Denmark in order to “emphasize the interconnectedness of the Atlantic World” (Gaffield). In doing so, Gaffield discovered the long lost original copy of Haiti’s Declaration of Independence at the National Archives of the United Kingdom.

Further, Gaffield’s curation of her research process reveals “the interconnectedness of empires, colonies, and countries in the early modern period. Historians are beginning to conduct their research in new ways, traveling to archives in multiple countries and researching in several languages” (Gaffield). When confronted with a mass body of material like a library of a country’s history, one must decide how and why they curate information. The implications of both the DDC and Gaffield’s work show that their responsibility is enormous. The impact and voice of this information within culture is up to the way in which it is classified.

1. Fandino, Marta (2008). “UDC or DDC: a note about the suitable choice for the National Library of Liechtenstein”. Extensions and Corrections to the UDC. Retrieved 12 October 2014.

Works Cited

Balas Kaplan, Tali. “Done with Dewey.” Web log post. ALSC Blog. Association for Library Service to Children, 17 Apr. 2012. Web. 12 Oct. 2014.

Gaffield, Julia. “Haiti’s Declaration of Independence: Digging for Lost Documents in the Archives of the Atlantic World.” The Appendix 2, no. 1 (October 2014)

Sperberg-McQueen, “Classification and its Structures,” in Schreibman et al., ed., Companion to Digital Humanities (Malden, Mass: Blackwell, 2004)

 

In his article, “Classifications and its Structures”, C. M. Sperberg-McQueen gives readers an idea of the complex nature of classifications and the nearly infinite possibilities that we may encounter when trying to classify certain pieces of information. The section on “One-Dimensional Classifications” describes nominal classifications, which “consist simply of a set of categories: male and female; French, German, English, and other; noun, verb, article, adjective, adverb, etc.” (Sperberg-McQueen). This notion of a nominal classification of gender reminded me of the binary gender system I learned about when I took Gender Studies 10 last spring. Most of society assumes that there are only two finite choices in terms of gender, male or female; however, in actuality there are over 50 gender options. It has taken time and a lot of effort on the part of feminists and others who fight for gender equality, but Facebook now recognizes 51 different categories when asking users to identify their gender. I found an article entitled, “What Each of Facebook’s 51 New Gender Options Means” by Debby Herbenick PhD and Aleta Baldwin on thedailybeast.com (http://http://www.thedailybeast.com/articles/2014/02/15/the-complete-glossary-of-facebook-s-51-gender-options.html), relating to this foreign concept of many different gender identities.  Before I took Gender Studies 10, I had always identified myself as female and had never encountered any sort of problem when asked to pick between the common gender classifications. Since I did not have a problem with them, I naively disregarded the people who do not fit within the binary gender identities of male and female. Now, I see that the politically correct way for me to classify myself is “cis female” or “a female who identifies as a woman/ has feminine gender identity”. Even though the classification of gender is still nominal, as there is a discrete number of choices, it is much more complex than the majority of us realize. If you identify yourself as “gender nonconforming”, but must pick between the two categories of male or female when filling out paperwork at the DMV, you may face an internal conflict and feel marginalized. By recognizing a wider array of gender identities, Facebook is helping to eliminate the binary gender system and raise awareness about the complexity of gender classification. Sperbeg-McQueen recognizes that most classifications are even more complicated than “one-dimensional classifications” and fall under the category of “N-dimensional”. If something like gender, which seems so straightforward, is actually much more complex, I cannot even imagine the effort and time that goes into correctly classifying multivariable data sets.

Metadata’s core principles have been molded alongside the recent advancements in technology over the years. As Anne Gilliland states in “Setting the Stage”, the increase in the amount of information being shared and created digitally has drastically expanded the importance and widespread use of metadata. What used to be a term only mentioned and understood by cataloging professionals has transformed into a concept known and practiced by all. Metadata is crucial for individuals to effectively locate and access needed information as well as for companies and organizations to archive important information. Attention to detail with regards to the creation of metadata has also become vital in this day and age, tied to this expansion of digitally accessible information.

This digital adaptation that we have seen over the years has also caused controversy. The rights surrounding access to certain metadata have been heavily argued and in a specific case concerning Yahoo, government threats have been discovered (Link to Article). This article covers Yahoo bringing light to a battle between itself and the US government in 2008 regarding the constitutionality of the government’s request of Yahoo’s data. The government threatened Yahoo with heavy fines if they failed to comply with their “surveillance efforts”. Yahoo released the documents associated with their legal dispute to the general public in an attempt to show the extensive effort they went through to keep certain personal data safe that was tied with user information. Yahoo ultimately ended up losing the legal fight and was forced to cooperate with the National Security Agency, granting them access to their information.

This controversial incident expands upon Gilliland’s article as it brings security into the conversation regarding metadata’s increased role over the years. It questions how much regulation should be involved with certain metadata that may include personal information. This is a very delicate issue with technology expanding, as the amount of personal information that is stored and managed online is only going to expand. With the recent security breaches around the web, it is safe to say that no personal information put online is absolutely secure. With this in mind, Gilliland’s question of “how much [metadata] is too much” can be looked at from a different perspective. It raises both questions of how much regulation should be allowed as well as how much personal information should be stored online to begin with. Metadata has been great for the purpose of recording information and making it available to the general public. However, with the always-present possibility of personal information being digitally transferred between organizations as well as recent security breaches, it will be interesting to see if metadata is managed differently in the coming years.

 

Sources:

“Yahoo was threatened with heavy fines by US government over metadata”. Washington Post. September  11, 2014. http://www.washingtonpost.com/business/technology/us-threatened-massive-fine-to-force-yahoo-to-release-data/2014/09/11/38a7f69e-39e8-11e4-9c9f-ebb47272e40e_story.html

Anne Gilliland, “Setting the Stage,” from Murtha Baca, ed., Introduction to Metadata (Los Angeles: Getty, 2008): http://www.getty.edu/research/publications/electronic_publications/intrometadata/setting.html

Image: http://www.ccmostwanted.com/wp-content/uploads/2013/08/Cyber-Security.jpg

In “Classification and its Structures,” C. M. Sperberg-McQueen argues, “classification schemes are felt more useful if the classes are organized around properties relevant to the purpose of the classification.” This is true because any kind of organization system implies knowledge of the subject. Without this insight, the classification becomes arbitrary. For example, one might classify paintings by acidity for an analysis of Renaissance art & culture. What would be gained from this work? Acidity is irrelevant to the classification subject and project scope. Instead, scholars must first decide which information is relevant to the topic and then decide on a classification system that is relevant to the information. The implications of these decisions are numerous and important to the overall project.

One of the most interesting classification systems that I am familiar with is for typefaces. According to Ellen Lupton’s Thinking With Type, three main classes were introduced in the 19^th century: humanist, transitional, and modern. These classes correspond roughly to the Renaissance, Baroque, and Enlightenment periods in art and literature. Although these letterforms may seem similar to the untrained eye, small changes in transitions, angles, and serifs represented huge shifts in print culture. In this case, the classification is useful because it serves both purposes outlined by M. Sperberg-McQueen: “by grouping together objects which share properties, it brings like objects together into a class… [and] by separating objects with unlike properties into separate classes, it distinguishes between things which are different in ways relevant to the purpose of the classification.” Historians and typography critics have since created more serif-based classifications (e.g. slab-serif, humanist sans-serif, and geometric sans-serif) and “proposed more finely grained schemes that attempt to better capture the diversity of letterforms” (Lupton).

Finally, classification also reminded me of another aspect of typography: hierarchy. Just as a classification system exposes similarities and differences between objects, “a typographic hierarchy expresses the organization of content, emphasizing some elements and subordinating others” (Lupton). Hierarchies help readers navigate through text similar to how classifications make data analysis and visualizations easier for scholars. Although there are infinitely many ways to accomplish either task, their implications are much the same. Sometimes one difference is emphasized at the expense of another or some information is omitted from the system because it does not easily fall in to any of the groups. This cannot be avoided; instead, designers and scholars must optimize their organization systems based on the scope, data, and end user’s needs.

https://www.flickr.com/groups/camerabag/

This is a link to the “What’s in your camera bag?” group on Flickr. Members can join and add a photograph of their camera bag with its contents laid out around it: the cameras, lenses, and other photographic equipment they own, their favorite film types, their cellphone, laptop, tablet, notebooks, and assorted personal effects. It’s the photographer’s equivalent of “What’s in your bag/purse?”

Like members of other groups on Flickr, they are basically making their own collection of information objects. In addition to content, they create metadata in the form of titles, descriptions, and tags. Each person writes as much or as little metadata as they like, which can lead to difficulties in searching for specific information. For example, the group currently has 2,382 photographs, yet only 252 have the “camera+bag” tag attached to it. As a result of the inconsistency in tagging, searching for photos like this outside of the group’s photostream is completely ineffective. The titles that people write are also inconsistent because, for instance, some might prefer the title “What’s in MY camera bag” over “What’s in YOUR camera bag?” Obviously, there are problems with the quality of user-generated metadata on the Internet, i.e., folksonomies, because there are no standards and no controlled vocabularies.

In “Setting the Stage,” Anne Gilliand addresses the bewildering variety of metadata standards by categorizing them according to the functions of metadata. There is no one standard that works well for everybody, so Gilliand argues that by understanding the various types of standards, as well as the attributes and characteristics of metadata, and the phases of its life cycle, information professionals will be able to choose the most appropriate standard or combination of standards for the needs of their institutions, collections, and users. Creating and maintaining metadata is extremely labor intensive and costly, so information professionals have to choose which standard(s) will cover their needs both now and in the future. The example of inconsistent tagging and titling in photos of camera bags and their contents is admittedly trivial and does not require high quality metadata, nor is it likely that casual Internet users will spend much effort improving their metadata, but it does illustrate a few of the problems that can be caused by low quality metadata. The importance of high quality metadata is much higher for museums, archives, libraries, etc., especially nowadays when the Internet has created a larger, more diverse audience for collections and repositories. Luckily, digital tools allow for the accommodation of multiple audiences that include scholars as well as less highly-trained students, teachers, and the general public.

In the past I have worked with metadata in regards to artifacts and documents. Usually, I find these helpful in tracing the history of a piece as well as determining copyright. It is for this reason that I was reminded of several art scandals where background information and metadata can be crucial to understanding the piece. For instance, a few years ago, Steve Martin bought a “newly discovered” Campendonk painting only to discover later that it had been the work of the infamous Beltracchi, an art forger. While not digital, this is a perfect example of where cataloging and metadata can contribute to authenticity. Gilliland states that metadata plays a role in “preserv[ing] cultural heritage information.” I agree that metadata should accompany every document or piece that is exchanged not only in the digital world but also in the physical world. At some point, all information will be available in digital form and to not lose the original work, it is necessary to keep track of metatdata. In the case of the Capendonk painting, pieces such as these should have an incredibly detailed metadata to accompany them. The painting was proven a fake by the type of paint used; information such as this can be applied to other art pieces.  Artwork metadata at its best would consist of the painter, style, dimensions, description, location history, purchase history, chemical analysis, and names of individuals who can attest to the authenticity of the painting.

The Daily Beast, http://www.thedailybeast.com/articles/2011/06/01/steve-martins-campendonk-painting-art-forgery-scandal.html

Another example I am fond of is the Amber Room; a famous treasure of Russia that was dismantled by the Germans during WWII and remains lost. In this situation, metadata serves to give information about an object that no longer exists. If an artifact is destroyed, researchers must rely on images and stored information rather than the original.  Digital metadata is therefore also a method of preservation or documents or artifacts that no longer physically exist. Also, metadata serves to inform the public of basic information rather than having every researcher investigate copyright, take dimensions, determine authenticity, etc.

Smithsonian Magazine: Amber Room, http://www.smithsonianmag.com/history/a-brief-history-of-the-amber-room-160940121/?no-ist

Gillilan has also stated the need for standards in regards to metadata. I agree with the need to standardize the rules of metadata so as to have consistency in reports. I am also curious however that at what point does the metadata overcome the object? Focusing on the upkeep of the metadata, In my opinion, cannot take the place of upkeep of the object. Originals and non-digitized documents are still the responsibility of libraries and while digitization and metadata may make the information more accessible, neglecting the original information is unacceptable.

Anne Gilliland, “Setting the Stage,” from Murtha Baca, ed., Introduction to Metadata (Los Angeles: Getty, 2008)

MT @norasnoodlings: @SmithsonianFolk tackle moldy collections @SmithsonianArch labs #manicuremonday in nitrile gloves http://t.co/QVuiT2SILW

— Smithsonian Archives (@SmithsonianArch) July 21, 2014

After delving into the material assigned for this week’s readings, I couldn’t help but find myself reminiscing about the past summer, which I spent living in Washington D.C., working as a (regrettably) unpaid intern at the Smithsonian Institution’s Center for Folklife and Cultural Heritage, abbreviated as the CFCH (that’s me, on the left in the blue dress).

During my internship I worked largely within the Center’s archives, processing different art and correspondence collections, and attempting to discern what might be worth including in recent digitization efforts funded by a federally subsidized, and therefore unsurprisingly modest budget. Along with other interns (namely my friend on the right, Heather), we sorted through hundreds of original works of art from records produced at the Center during the 20th century (you can browse many of the recordings here: http://www.folkways.si.edu/folkways-recordings/smithsonian).  The works of album artwork that we processed were  historically produced as a part of an  effort to release more unconventional or unorthodox music during the 1950s and 60s, when government censorship and the era of McCarthyism surged across the nation. Some of the most famous folk musicians of the 20th century such as Woody Guthrie and Pete Seeger recorded with Folkways and left original works of art and recordings in the hands of the Smithsonian’s CFCH. Even at an institution as prestigious as the Smithsonian, these materials were not accessible to the public online. Individuals are still required to make an appointment to view the collection, and many of the materials have yet to be cataloged at all.

During our efforts, we grappled with many of the same questions brought to light by Chon Noriega in “Preservation Matters,” such as what the function of our collections were on both a public and an institutional level, and what the responsibility of the Center was in preserving these historic materials. Essentially, our boss wanted us to pick out particularly good examples of well-preserved or exceptional works of art to be digitized and eventually put online for public access. However in doing this, we were adding weight to Gamboa’s theory that it was impossible to give an accurate account of a  history because of the very fact that these materials were part of an archive. Heather and I were selecting what history would be publicly shown, and which pieces of art would remain hidden away or potentially even discarded (such as those with hazardous mold or other damages). I think this is one of the most universally important things to keep in mind when using archives and collections. True “History,” is unknowable because archives are an imperfect and meager snapshot of what happened, how it happened and who was involved.

For more information on the archives and collections I worked with check out these links:

http://www.folklife.si.edu/

http://www.folklife.si.edu/archive