Dublin Core

DUBLIN CORE ASSIGNMENT

Element #1

1.Coverage

2. It seems that Dublin Core recommends that this attribute be used to connect the object of the record with a place or time (or range of either). The spatial or temporal period should be relevant to the overarching themes of the record, and does not need to be expressed in numbers (for example, one could use a time period like “classical era” for music). In my opinion, Dublin Core means for this attribute to be used to define larger themes or areas that encompass what is being cataloged, and to provide more flexibility than a date or location field. For example, if one was cataloging a cat, coverage could be used to define the life span of the animal, rather than the birth date or adoption date. Similarly, if one was cataloging an artifact found at a dig, coverage could be used to identify the country of origin, rather than where it is sitting currently (though the element could be used in that manner). In short, I think that coverage is meant to provide broader context to the object.

3. Coverage could be used in two ways to describe an event. It could be used to describe the total time period of the event, from start to finish, including any aftermath. This should not to be used as a creation date; rather, it should describe the overall time period of the event. Coverage could also be used to describe the location of the event pinpointing the exact location of an event if there is one (either by name or by coordinates), or by defining a geographic area (such as Asia, or San Francisco). Another use of coverage could be to use it to define the scope of the event. For example, in a natural disaster, coverage could be used to describe the area affected (by defining a “blast radius” or other delimitation).

4. I would recommend that coverage be included in this metadata assignment. It encompasses extremely important information that, in the case of the geographic coverage, does not have another home in any of the other metadata elements. The type of information stored in coverage is information that a patron my search by, or use to hone down a results list.

5. I propose that coverage be used to describe the location of the disasters being cataloged. To keep the vocabulary consistent, Dublin Core recommends using a controlled vocabulary from which location terms can be drawn. I suggest that we use their recommended vocabulary, the Thesaurus of Geographic Names, to make the data in this element consistent. Granularity of data coverage will be at the cataloger’s discretion, as will be entering multiple locations. However, the preferred term for the location must be used, and entered as seen in the thesaurus. Therefore, the entry format will be free text. The hierarchy of terms will not be represented in one element entry – if more than one entry on the location’s hierarchy is needed to fully explain the geography of the event, the element should be duplicated and the second (or third, etc) location should be added as a separate element.

6. Coverage should be used to describe the geographic location or locations of the event being cataloged. The level of specificity needed is left to the cataloger’s discretion, but it should provide a reasonable level of detail in relation to the event. Vocabulary for this element will be drawn from the Thesaurus of Geographic Names, and only the preferred term for locations should be used. The entry mode will be free text, but the locations should be entered as they appear in TGN. The element can be repeated should more than one term from TGN be needed to fully describe the geographic location.

Element #2

1. Language

2. Dublin Core recommends that the language element be used to express the language of the item being cataloged, whether it is a written, spoken, or signaled. There are numerous sub-tags that identify primary and secondary languages, region, and many other features of a language. It is also highly recommended that a controlled vocabulary be used to identify both the language and its features.

3. For cataloging events, the language tag could be used to describe the language used, particularly if it is a concert, lecture, or other performance-based event. In the context of the metadata record assignment, the language tag could identify the languages of the people affected by the disasters that we are cataloging, or the predominant language of the area.

4. I would not recommend that the language element be used for the metadata assignment. The disasters that we are cataloging are not performance-based, so defining the language of the event becomes very difficult. Many of the disasters take place in low-income areas (which are traditionally more diverse), or else over giant areas of continents. Each poses difficulties defining all the languages used during the disaster. Though the dominant language could be used, it doesn’t fully represent the population effected. Finally, the information contained in the field is arguably unimportant from a cataloging and retrieval point of view – very few people will be searching for the disaster based on its predominant language.

5. n/a

6. Language should not be used as an element in this metadata assignment. Disasters do not inherently have language, and to try and catalog all the languages of the people affected is virtually impossible. Additionally, cataloging the languages associated with the event would add little to the record, as the completeness of the information would be dubious, and very few people would search according to the language of the disaster.

Personal Information Management

The Lerner Vinyl Collection

Currently, my husband and I have close to 600 album sets, many with multiple discs, in three formats: 45rpm, 33⅓rpm, and 78rpm. The collection is spread across two pieces of furniture that have two rows of four cubbyholes each. There are three main genres: classical, world music, and everything else. “Everything else” is spread across the top cubbyholes, only includes 33⅓rpm records, and is alphabetized first by artist last name or band, and then by album name. World music begins the bottom row, and also only includes 33⅓rpm records. Again, they are first alphabetized by artist and then album. Next are the classical records, which are arranged alphabetically by composer. Those records with more than one composer (many classical records are compilations, or have two related pieces by different composers) are taken on a case-by-case basis and put under the composer whose piece on the record we listen to more. 78rpm records (all classical) followed by 45rpm records are the final components; they are arranged in the same alphabetical sequence as the other records.

Why is the Collection Organized This Way?

While I appreciate personal organization, when categorizing music one can get caught in minutia and soon the system can only be interpreted by the creator. Because vinyl is a physical manifestation of music in that the actual vinyl cannot be searched electronically, the system needed to be simple enough for friends to use, but detailed enough so that we didn’t need to hunt every time we wanted to listen to something new. Part of the joy of the collection is watching other people pick what to listen to, and serendipitous discovery is an important aspect of that process. Due to this, I made the organization specific enough to guide whoever is looking to the general area, but broad enough to allow the searcher to discover other items of interest. Simple genre categories allow the searcher to eliminate certain types of music, while separation of format allows one to immediately discount certain records. Beyond these broad separations, alphabetical ordering seemed most logical for our needs. As I know my own collection fairly well, this system allows me to pinpoint a record’s location surprisingly quickly. Upon reflection, I realize that I created this system more for our friends’ benefit than for my own – in this instance of my personal information management, I was far more concerned with how other people would access the materials than I was with my own access needs. Perhaps this speaks to the social nature of the materials in question. Either way, my audience and their discovery is by far the most significant reason my collection is organized in this fashion.

Are There Certain Aspects that are Successful for the Material?

In particular, I think that subdividing the collection into three loose genres is extremely useful for the materials. As I stated earlier, I did not want to micro-divide the collection, but I think that the three genres really lend themselves to three different listening experiences. No one, in my experience, has had difficulty choosing between those three options at any one particular moment. If a goal for my organization scheme was to be able to roughly find records, but also allow people to discover things they might not have found otherwise, this subdivision points them in the right direction without leading them straight to one record. The divisions also reflect the three main areas that we collect: classical, world music, and a little bit of everything else.

Improvements?

Thinking back to the Weinberger piece, “The New Order of Order,” the collection is rooted in the first order of order, the physical location of actual objects. While this works quite well for most of the collection, it is the cause of much consternation in the classical sections because classical records often have more than one composer, artist, or piece on a recording. There is more than one logical place for many of the items. Finding particular pieces of music is more difficult here than in other sections of the collection because not only does one need to remember the name of the piece, but one also needs to remember the composer and whether or not it was on a record with another piece by a different composer, which composer it was filed under, and who is performing the piece. There are so many variables in each aspect that finding any common thread by which to organize has been unsuccessful to date. Perhaps I need to enact Weinberger’s second order of order, and create cross references so that all pieces and composers are fairly represented and accessible.

One Final Observation:

I find it interesting that this scheme incorporates almost all of the PIM behaviors discussed by Barreau in “The Persistence of Behavior and Form in the Organization of Personal Information”: filing, piling, and spring cleaning. Obviously, the records are all filed in particular locations; however, I am often lackadaisical about actually filing records where they belong immediately after listening. As a result, there are piles of records all over the living room, each one the evidence of a particular listening session. Finally, about twice a year, we spring clean, re-ordering everything, weeding, and marking items for sale as we go. Though I think that Barreau described these behaviors as being more intentional than described here (besides the spring cleaning), I would guess that most people employ all three behaviors in different areas of their personal information management.

Works Referenced:

Barreau, D. (2008) The Persistence of Behavior in the Organization of Personal Information. Journal of the American Society for Information Science and Technology. 59(2) p 307-317.

Weinberger. (2007). Everything is Miscellaneous. New York: Times Books. Chapter 1: The New Order of Order.

Electronic Resources

Blog post #4
Please note that I was sick when this was due, and a new due date for me was set at Sunday, 7 March 2010.

1. LISA

2. I used the topic “controlled vocabulary.” Doing a search using this complete term yielded 464 published works. The results seemed to cover the scope of the topic (including social tagging, metadata, and terminology mapping) as well as lexicographical variations such as “controlled vocabularies.”

3. Thesaurus

a. The thesaurus for LISA was extremely easy to find and use. There was a link at thebottom of the search page that took me straight to a page that allowed me to search the thesaurus. Results could be viewed in their place alphabetically in the thesaurus, or else as a hierarchy or rotated index. One can check boxes next to the terms and use those to either narrow or broaden their search. Additionally, terms can be exploded to include all narrower terms within an entry. Finally, related terms are included in each entry. The only real drawback I could find to the interface was that the thesaurus couldn’t be browsed, if, for example, a user didn’t have a specific search term in mind. There was also no help feature for using the thesaurus that I could find, but this was not a real drawback for me, as the interface was so intuitive.

b. Controlled Vocabulary was in the thesaurus when I searched for the term. It had an alternate term (UF), vocabulary control, as well as three related terms. There were no narrower or broader terms; however, after digging in the thesaurus for a bit, I did find terms that used those signifiers, so the thesaurus does employ them.

c. The thesaurus is extremely easy to use. All one need do is navigate through the thesaurus, marking terms that should be included in the new search. Additionally, how the terms should be used in this new search needs to be chosen (either an AND, OR, or EXPLODE command). A fairly large downside to this method is that each new term cannot be given its own command. Therefore, I cannot search for [[controlled vocabulary AND taxonomies] OR thesauri] directly from the thesaurus. However, searches like this can be done through the advanced search pane, though it seems somewhat of an annoyance to the user to have to first search for the preferred terms and then have to navigate to another pane to do the actual search.

d. Other functionalies offered through the thesaurus search pane include selecting a thesaurus for a different database in the bundle, and links to the previous and next terms in the list. As stated earlier, a nice function to add would be delimiting search commands for individual terms rather than for the marked terms as a whole.

e. For the most part, the LISA thesaurus abides by the recommendations set forth by the Craven guide. In particular, the LISA thesaurus standardizes the forms of the entries (pluralizing when appropriate), following the rules for making multi-word terms the preferred terms, indicating syndetic relationships in a consistent manner, and though I did not encounter any homographs in the terms that I looked through, the adherence to other standards would lead me to believe that they appropriately label homographs. The only real deviation from Craven’s guidelines that I could find was the lack of thesaurus guide. There was no introduction, no scope statement, and no real indication of what the thesaurus was about or how it was structured. Though scope notes were included where appropriate for terms, they did not aid in determining scope, structure, or intent. There are of course, assumptions that can be made – it is a thesaurus for library materials, so the scope should be the library and information science discipline; however, I think that Crane’s guide makes clear that it is important for this type of information to be stated explicitly: for the LISA thesaurus, it is not.

f. LISA’s thesaurus is an extremely helpful tool for database users at all levels of expertise. I think it more effective for those users at the novice or intermediate level, due to the fact that searching directly from the thesaurus is limited to one search command (AND, OR, or EXPLODE). Due to this, compound searches are more complicated to execute, and seem to demand more familiarity with the system. The thesaurus is intuitive enough for a novice user to gain some value, and more importantly, the thesaurus term search feature allows a user who is unfamiliar with the thesaurus vocabulary to search (without having to browse through hundreds of terms) for terms until they hit on one that can serve as a starting point. For intermediate users, features such as the explode command allow search ranges to be expanded. In total, the LISA thesaurus is an effective, if incomplete tool that allows basic and intermediate users access to a resource that adds value to their search. Expert users may have a harder time due to the limitations of the “search directly from the thesaurus” features, but nonetheless, they can still use it for its intended function: to provide a controlled vocabulary with which to search the database more effectively.

4. After my explorations, I would certainly modify my search; use the thesaurus in the future; and tell both friends and patrons to use the thesaurus in their searches.

5. The new search results, after having used the thesaurus to broaden my search, brought back 1366 hits, on subjects relating to controlled vocabularies. Topics included ontologies, thesauri, social tagging, taxonomies, and terminology mapping. Overall, the hits were related to what I wanted to retrieve, so the new search was a success.

6. Craven states in “Thesaurus Construction,” that “a thesaurus is a tool for vocabulary control. By guiding indexers and searchers about which terms to use, it can help to improve the quality of retrieval.” I think that this assignment effectively taught me that databases that I use weekly can be aided by controlled vocabulary, which is an odd realization because I use MeSH constantly at my position at the Health Sciences Library. There seems to be a divide when it comes to the scientific community and the rest of the academic community in terms of how databases are used. I have taught people all about MeSH, but I have never explored thesauri in other databases. That trend seems to be reflected in the academic community’s electronic resources – MeSH is more commonly known and accepted across the medical discipline, but to my knowledge there is no unified or commonly known vocabulary for other fields. Of course, they do exist, but I could not name any besides those in scientific fields, and therein lies the disparity. Perhaps this is because hard science and medical fields are grounded in facts and figures (rather than interpretive disciplines) that can be pinned down with great accuracy by language, and thusly are the easiest to quantify in a controlled vocabulary. Maybe as electronic resources become even more integrated into all academic communities, other controlled vocabularies will rise to the surface and become fully integrated into their respective communities.

Bed Bath & Beyond v Linens 'n Things

Navigation through the Bed Bath & Beyond (BBB) site was fairly intuitive. On the left side of the page there was a navigation bar that listed 33 main subject headings. As the user rolls the mouse over each one, subheadings appear. Some of the subheadings had a double arrow indicating that they too had their own subheadings that appear when the mouse rolls over that term, and so on. In this manner, the hierarchy of the vocabulary is made explicitly clear to the user, in addition to the preferred vocabulary used to describe the inventory. When a shopper actually clicks on a term, they are taken to a new page that lists items that fall match or are related to that term. Linens ‘N Things (LNT) has a similar set-up in that the hierarchy of the preferred vocabulary is made explicit to the user. On the LNT home page, there is also a navigation pane on the left side of the page. It lists the 11 departments, and clicking on the name of the department brings the user to a new page that lists subheadings within that department. If the shopper desires something closer to the BBB model, across the top of every page are 10 main subject headings that operate in the same manner as BBB. When the shopper rolls over a term with the mouse, a list of subheadings appears; if they roll over a subheading with its own subheadings, then those appear as well. Each site also had a catchall search bar at the top of every page in which the user can enter search terms of their choosing, key words, or item numbers. The main difference between navigating the two sites is that the LNT site always gives the user access to the controlled vocabulary, while the BBB site requires the user to return to the main page to access the list of terms if they enter their own search terms rather than navigate the hierarchy. When a user enters a search term of their own, both sites allow the user to refine that search on the left side navigation bar, by, for example, color or brand, in the style of facet searching.

To discuss controlled vocabulary (CV) issues in this post, two contextual examples will be used: bath items (specifically towels) and drapery. While both LNT and BBB both have a large and similar inventory, each uses a slightly different approach in their CV. While I have chosen examples that are slightly exaggerated in their difference to prove my points, these differences are trends across the sites, though there are of course, exceptions.

BBB has a far more extensive vocabulary, which is clear just by looking at the main subject headings of each site: BBB has 33 compared to LNT’s 10. However, just because there are more main subject headings does not mean that BBB’s CV has more depth or less interaction between broad terms (BT) and narrow terms (NT). To determine that, we are going to look more closely at a specific set of subheadings: towels. The BBB directory tree that leads to towels looks like this:

Bath
-------Towels
--------------Solid Bath Towels
--------------Print Bath Towels
--------------Decorative Bath Towels
--------------Monogrammed Bath Towels
--------------Beach Towels

LNT, in contrast, leads the shopper to towels in this manner:

Bath
-------Bath Towels and Rugs

From the outset, BBB prefers the user to refine their search via the hierarchy as much as possible before viewing items for purchase. While this is an efficient method for guiding shoppers to their product, it might not allow for as much serendipitous discovery as with the LNT method. LNT guides the user via the CV hierarchy to a much broader category (bath towels and rugs). Once the shopper is redirected to the page with all the bath towel and rug products, they are given different options such as color by which to narrow the search. A drawback to this semi-faceted method is that more than one option cannot be chosen, so the user could not, for example, find blue Hotel Collection towels, only blue towels or Hotel Collection towels. BBB does not allow for users to narrow their search when they have arrived to a product page via the CV hierarchy. By this example, it becomes clear that BBB has a CV with a far more nuanced BT/NT relationship structure. BBB uses more terms and more categories across the board, allowing the shopper a surprising amount of product specificity by navigating the CV hierarchy. However, even though there are more terms to navigate in the BBB model, both vocabularies have a similar hierarchical depth. Each main heading usually has terms with at least one, and in some cases two subsequent levels. The main difference is that the BBB structure has more specificity.

The issue of term choice has been touched on briefly in the context of the number of terms used. To compare the actual vocabulary used, we will look at window treatments. BBB sets up their window treatment vocabulary as follows:

Window Treatments
-------Panels
--------------Contemporary
--------------Traditional
--------------Casual
--------------Formal
--------------See all window panels
-------Sheers
-------Extended Length
-------Energy Saving/Room Darkening
-------Kids + Teens
-------Kitchen Tiers
-------Valences
-------Specialties
-------Blinds/Shades
-------Hardware/Accessories
-------Tiebacks
-------Clearance

There is a great amount of specificity and to an extent, industry language included in these terms. Personally, I did not know what a kitchen tier was before I looked at the pictures. LNT uses this structure:

Window
-------Window Panels
-------Drapes
-------Window Blinds
-------Window Valences
-------Window Tiers
-------Blackout Liners
-------Clearance
-------Window Hardware

LNT uses nine terms to describe window treatment products, while BBB uses 18. More importantly, though, is the difference in terminology. Is a window tier different from a kitchen tier? With BBB, drapes are included in window panels, while at LNT, they have their own subheading. Though they are essentially describing the same products, BBB again shows a level of specificity that I believe is geared more toward someone who knows what they want, rather than a browser. Personally, I find the LNT terms easier to use because I know nothing about window treatments, and would rather browse my options than go straight to a specific product. Herein, perhaps, lies what I think is the biggest difference between the two websites. LNT is better for browsing products due to its broader categories and semi-faceted sub-searches, while BBB is geared more towards the connoisseur who knows exactly what they want and can navigate through the specific terms to land very close to the actual product without entering any search terms.

In-Class Ruminations (Blog Post #2)

The article that I have found to be most interesting in the class thus far was the Buckland piece “What is a ‘document’?” I had not read the article previous to this class, and though I know it seems naïve, it really opened my eyes to ideas that had been ruminating in my head, but I had been unable to articulate because I don’t yet have the foundations and vocabulary to do so. Questioning something as simple as the definition of the unit of study – a document – not only helps to identify the scope of the profession, but also identifies areas of uncertainty, dissent, and cultural differentiation of the field. These things are all important to me as a student as I formulate my own opinion and definition of the field. So often in the excitement of learning new technologies and theories, we forget (or at least I do!) to study the founders of a field; Buckland’s piece gave me just enough insight into the earlier theorists of documentation to make me realize how much I didn’t know. Additionally, I find it fascinating to see how each opinion of what constitutes a document interacts with the other – how earlier opinions inform later ones, and in what ways they differ.

Three opinions that Buckland presents in the article were particularly interesting to me: Briet, Otlet, and Ranganathan. I especially enjoyed Briet, with her unequivocal definition of a document in 1951: “any physical or symbolic sign, preserved or recorded, intended to represent, to reconstruct, or to demonstrate a physical or conceptual phenomenon.” I love that she defines, with simple yeses and nos, what is or is not a document. It takes audacity to lay out a black and white definition of the scope of a field, and while I don’t necessarily think that it is all that simple, I do think that her definition of what constitutes a document allows leniency for new technologies and formats – she is more forward thinking than her colleagues of what the future might hold. Otlet provided a foundation for Briet’s opinions by allowing, in 1934, that documents can also be objects provided that they inform the observer. This has obvious ramifications on Briet’s thinking, as allowing an object to be a document, rather than just written papers is fundamental to her idea that anything that is “evidence in support of a fact” is a document. Alternatively, Ranganathan is an interesting juxtaposition to Briet and Otlet, as he rejects their ideas of objects as documents and contends in 1961 that a document is a “record on a more or less flat surface,” and an “‘embodied micro thought’ on paper ‘or other material, fit for physical handling, transport across space, and preservation through time.’” I wonder if there is a cultural difference as to the way documentation is handled in India versus Europe. Whatever the reason, I find it fascinating that with a foundation of Briet and Otlet, (if in fact he read them, which I assume he did, if he was writing in the field), Ranganathan took a completely conservative view on the subject, relegating documents to physical pieces of paper. Would Ranganathan consider a picture of a clay pot with metadata attached a document, rather than the clay pot itself?

Grocery Store Post

For this exercise, I visited the Carrboro Harris Teeter, in the Carr Mill Plaza on Greensboro Street.

Toothpicks

I found the toothpicks nestled between plastic cups and bowls, straws, and napkins. The common thread was party planning – each of the items in direct proximity to the toothpicks could be used when entertaining. Other items that could be placed nearby are Popsicle sticks, cocktail straws/umbrellas, and disposable cutlery. Perhaps items that would be used with toothpicks, such as olives, could also be close by. Initially, I thought that the toothpicks would be in the baking aisle – I had to look around and finally read every sign until I found the correct aisle.

Pasta Sauce

Items near the pasta sauce were white sauce, clam sauce, pizza sauce, vodka sauce, canned parmesan cheese, dried pasta, and can openers. It makes sense that these items are placed near traditional pasta sauce; obviously the other sauces are just variations on a theme, so they should be placed nearby, and pasta is what you would ostensibly put the pasta sauce on top of, so this is a smart placement as well. Parmesan cheese is a traditional topping for pasta sauce, so that product placement is convenient too, potentially provoking a customer into purchasing something that they did not intend to purchase. The only other things that I could think of to add to this display were items such as capers, fresh cheeses, and other toppings for pasta. In addition, it might be nice to place some pizza crusts near the sauces for those who might want to create their own pizzas.

Peppercinis

Fresh peppercinis were predictably surrounded by other fresh peppers. Varieties included habanero, jalapeño, chili, poblano, sweet peppers, cubanelle, and red, yellow, and green bell peppers. Additionally, and, in my opinion, oddly, the peppercinis were next to several varieties of cucumbers. As most of these products were also fresh peppers, it makes organizational sense to place them together. Regarding the cucumbers, I can discern no other relationship other than “vegetable.” As a pepper is almost a universal item in terms of cuisine, specific to no single dish or style, it becomes difficult to determine what it should be placed next to in the physical store. I think that it should remain with other varieties of peppers, and with other fresh vegetables.

The Store

The arrangement of the store was easy for me to comprehend due to the simple fact that it is my regular grocery store. Save for the toothpicks, which I found on my second try, I knew where to locate everything on the list. The basic problem with the organization harkens back to the Weinberger piece, “The New Order of Order.” The grocery store is grounded in the first order of order, whereby it is concerned with the placement of the actual physical objects (Weinberger, 2007). Following the same logic as in Weinberger’s piece, the grocery store can be improved by second order organization. While there are finding aids such as signage in the store, I am more interested in the level of access granted by cross referencing. The problem with the grocery store is that an item like shredded cheese can arguably live near the cheese, near the taco items, near the pasta items, and in the dairy case. Which one does the shopper investigate first? Where is the “correct” home for the shredded cheese? If some sort of system could be created where items could be cross referenced, either with representations that marked the physical location of the item, or with iterations of the item living in multiple spots in the store, it would provide a greater level of access to the consumer.

Is a Vegetable a Document?

In examining the case of whether or not a vegetable is a document, I am considering two of the documentarians presented in Buckland’s piece: Briet and Otlet. Briet sets forth four evaluative standards for determining an object’s document status:

1. There is materiality: Physical objects and physical signs only

2. There is intentionality: It is intended that the object be treated as evidence

3. The objects have to be processed: They have to be made into documents

4. There is a phenomenological position: the object is perceived to be a document (Buckland, 1997)

Let’s consider these criteria in relation to a yam in a grocery store. A yam has a physical presence – it is a tangible object in the material world. This yam is in a grocery store; it is not growing in the wild. It might be a bit of a stretch to consider the yam as evidence (more on this to follow) – it is most likely considered food – but there is intentionality. The yam has been taken from its natural habitat and put in an aisle to be displayed for the public eye. The yam has also been processed in order to be on display – it has gone through a factory to be cleaned, boxed, and shipped to a precise location. The last factor, the perception of the yam to be a document, is the deciding factor.

Briet asserts that, “a document is evidence in support of a fact” (Buckland, 1997). Does our yam support any facts? Otlet provides some insight into the matter by stating that, “objects themselves can be regarded as documents if you are informed by your observation of them” (Buckland, 1997). Synthesized, an object is a document when it provides information or insight to the observer in support of facts or information. It is here that I believe the most important distinction between object and document is made. Does the yam provide us with any more insight or fact than the mere presence of the actual object? I would argue that it depends. The yam itself does not provide us with much primary data or contextual information with which to gain insight about this yam. However, were some metadata provided with the yam, for example a sign that read, “organic beauregard yams, SheepHorse Farm, NC,” then there is contextual information with which we can draw insights, facts, and conclusions. We now know that SheepHorse Farm possesses the correct soil mixture with which to grow healthy yams, that they grow them organically, and that this is a locally grown item. Just as a clay pot in a museum means nothing and provides no insight without context, neither does our yam. I assert that this is the crucial element of an object’s transformation into a document: both the evidential intentionality glossed over in my previous paragraph (providing metadata), and the ability to use that metadata to draw conclusions about the object and its context. Thusly, if enough metadata is provided with the vegetable, it is, in my estimation, a document. However, the vegetable alone is merely a vegetable.

Works Referenced:

Buckland, M. (1997). What is a “document”? Journal of the American Society for Information Science, 48(9), 804-809.

Weinberger. (2007). Everything is Miscellaneous. New York: Times Books. Chapter 1: The New Order of Order.