1. Have there been any in-depth looks to see how 'tagging' hierarchies on websites like YouTube have been influenced by metadata standards? Even if not by strict adoption of established terms, at least in how they are organized and used to make searching simpler for users?
2. "Metadata can come from a variety of sources." Are there any good programs out there for automated capture/creation of meta data other than what gets embedded in a file when it is created?
3. I really liked how the author made explaining metadata very simple and easy to understand, but also that she made connections to other types of metadata people are not only familiar with, but might not have realized fit the definition of metadata. We use so many slang terms for it like 'tags' and 'filters' and 'keywords', I think when people come across the term they initially panic (particularly when there's a standard attached to it, to say nothing of Dublin Core being flexible to work with).
1. The author says there is an increasing need for system designers to incorporate metadata in order to move the system forward across time. So is that the more metadata the better? What’s the maximum length of metadata? Would it be possible that metadata becomes larger than the data it is describing?
2. If the metadata reaches a certain scale, would there be metadata about metadata? If yes, what’s the difference between the nature of metadata and that of the metadata of metadata?
3. What’s the life cycle of metadata? Would it go through the same cycle as the information it is describing?
1 - Metadata is one arena where I see the post-custodial archival method really enabling archives to do their work both well and efficiently. In a post-custodial scenario (such as the HRDI's partnerships with the Kigali Genocide Memorial and Witness) the record creators generate all metadata and the archive attaches it to the digital objects. For example, in my work environment, I take metadata and video files from the Kigali Genocide Memorial and update said files in Glifos on the Genocide Archive of Rwanda website without adding any new or additional content, aside from quality control measures (making sure titles/file identifiers are consistent, checking for spelling errors, etc) and then the files are good to go. This gives record creators more control over how their materials are accessed and contextualized and provides archives with a more efficient system of generating metadata. So I guess my question is, why aren't more archives using the post-custodial method? It's generally viewed as a "pie in the sky" ideal, but has worked quite well for the HRDI.
2 - How can we educate records creators, especially those working in digital media, about metadata to ensure both enduring access to their materials, and preserve the evidentiary value of their records? Will there eventually be more positions where archivists work in a consultant capacity, such as Witness or Rhizome, where creators are given training and tools to generate their own metadata, or is this just wishful thinking because I desperately want that to be my future job?
1. As in the “Introduction to Metadata” article, Gilliland-Swetland here discusses the different types of metadata. She presents five categories: administrative, descriptive, preservation, technical, and use. However, in the previous article, the author listed only four – descriptive, administrative, structural, and use. While both authors recognized that the categories were broad and not all-encompassing, the fact that only three categories overlap is interesting. Is this change in metadata types related to the difference in time between the articles? Have our terms changed because our understanding has changed? Or is this just a difference of opinion?
2. Later in the article, the author talks about how large the accumulation of metadata can be, and that there is and will be a need to manage it as it is incorporated into information objects. She claims that “metadata registries and schema recordkeeping systems . . . are likely to develop” (7). Have these registries developed as the author predicted? Is her concept of the future of metadata our reality today?
3. The author frequently states the importance of having consistent metadata. In one instance, she states that, “metadata can also make it possible to search across multiple collections or to create virtual collections from materials that are distributed across several repositories, but only if the descriptive metadata are the same or can be mapped across each site” (9). Does she mean here that all metadata should have the same vocabularies? Or that it should have a standard format? Is she predicting that metadata could be used in different communities and should therefore be translatable between them?
1. Gilliland states that “well-crafted metadata results in the best metadata management in the short and long-term.” It is just often difficult to anticipate all of the metadata that may later be needed, and retrofitting metadata onto a collection is a Herculean task, as observed by the posts about Google Books. Do common metadata standards help this task by providing useful templates?
2. As we progress in our informational needs, we seem to require deeper and deeper layers of metadata to try and organize increasingly large collections of data. As investigated in the article about legal document searching, the amount of information available continues to explode. Will our metadata schema be scalable in these increasing levels? Has this been tested much?
3. This leads to the question of automation of metadata. Often now when one creates a digital object, it creates a certain amount of metadata about itself (such as a photograph) as a means of authentication and uniqueness. However, automation of metadata can go terribly awry. Does it seem likely that some level of human intervention or supervision is going to continue to be needed in the foreseeable future?
1 The author mentions mapping between common data elements or metadata on the page 4. How does this works? How to define the coverage? And for what purposes we need to map the metadata?
2 On table 2, the author gives nature of metadata. One is lay metadata created by persons who are neither subject nor information specialists, often the original creator of the information object, another one is expert metadata created by either subject or information specialists, often not the original creator of the information object. Is there a gap of understanding information object between original workers and expert who edits metadata? And how about the gap between expert and the user?
3 On the "Other Little-Known Facts about Metadata" part, the author states that metadata can come from a variety of sources, such as human creation or automatically created by a computer, or inferred through a relationship to another resource such as a hyperlink. How this inferring progress works? Who to infer, human or computer generated?
In response to your #2, I believe there is definitely a gap of understand between the original creators and the metadata editors. At my internship, I have to add metadata to digital files that can be up to 7 years old. Often times, the creator made abbreviations in the file name that I can not recognize what it means. The abbreviation obviously is of some kind of importance though for the creator to have added it. Often times I have no idea what the photo was originally created for or the context to be able to add metadata about its creation. This happens all the time in archives too, in that the archivist cannot understand a file order which probably made sense to the creator. I think that the nature of metadata category should separate the metadata added by the creator. I would suspect this metadata to be the most accurate as the creator understands the meaning or purpose for which they created the object. I think that the original creator would be the most expert in creating some metadata.
1. I've read Heather's first question and I got the one almost the same. Two articles present different categories about metadata and only part of the categories are overlapped. Does the difference between the articles suggest information professionals in academic circle have not make an agreement on some of key points of metadata?
2. To state why metadata is important, the author mentioned the retention of context. In this section, the author said "In an archive, for example, by documenting the content, context, and structure of an archival record, metadata in the form of an archival finding aid is what helps to distinguish that record from decontextualized information." Does this feature of metadata also apply the objects on the web? I mean, for example, copying images from other sites and pasting them on another page could lose their original context information for now. How could metadata change the situation?
3. In the section of "Legal issues": Metadata allows repositories to track the many layers of rights and reproduction information that exist for information objects and their multiple versions. I'm wondering how it works. To protect legal issues, should metadata of every object essentially obtains information about its copy right?
1. On pg. 11, Gilliland-Swetland starts a sentence with “If thorough, consistent metadata has been created,…”. That is a big ‘if’. Considering all of the different people with varying levels of expertise, backgrounds, experience, etc., I am not sure it is even possible to create consistent metadata, even if a certain controlled vocabulary or standard is employed. What kind of measures do different institutions use to ensure that metadata (for example) is consistent in the same collection after one librarian leaves and another replaces her/him? Is this where knowledge management should play a larger role in institutions?
2. Gilliland-Swetland refers to the information life cycle in her article. For a single object’s life cycle, does metadata change at each stage? If so, to what extent and is it possible to link that metadata together in a comprehensive and cohesive way? Or is it just dependent on the institution and what kind of metadata they are creating, be it descriptive, administrative, etc.?
3. On pg. 9, under the section ‘Increased accessibility’, the author states that different metadata standards allow users to search on the item/collection level, both within and across information systems. It is the ‘across information systems’ statement that I find hard to believe. I believe that metadata standards (or at least consistency) are hugely important and that they are probably normally adhered to within a single organization. But to search for materials cross-institutionally can be exceedingly difficult and frustrating, because everyone thinks they’re an exception to the rule. Different organizations may use the same set of Dublin Core but their personal prejudices and thoughts on how things should be inevitably vary, and lead to different sets of metadata, maybe even for the same object. Is this something that will ever be avoidable, or is it just part of being a human being and not a robot, that things are different from organization to organization, and even person to person?
1. While the author identified many attributes and characteristics of Metadata one thing I find troubling is the lack of concern for privacy and security. It seems like a common trend in technology that richer metadata is being collected on digital objects we create but I haven’t seen a lot of discussion about the potential for abuse in these technologies. I am thinking specifically about things like digital cameras that utilize GPS to attribute coordinates to each image you take, leaving a virtual record of every place you have been and a record of what you were doing. As metadata starts to become richer should we also be thinking about ways to ensure its protection?
2. The author mentions that metadata can be used in planning for new systems by benchmarking technical data that is collected automatically by a computer. I am a little uncertain about what the author is talking about here?
3. With the creation of metadata from the source being so vital for accuracy in archives I am wondering how systems could incorporate features that encourage more intelligent file creation. This works quite will in programs like iPhoto which periodically ask the user to identify faces in order to sort pictures in more intelligent ways. I wonder if similar features could be built into programs like Microsoft Word or your email too.
1. There are many types of metadata, and the article talks about MARC, AMC, ISAD, EAD, and SAA as examples. With so many different types, is there any standardization across all of them?
2. Majority of the metadata are English based, I wonder if there would be issues when metadata is applied across different languages?
3. In searching around the web about metadata, keeping it accurate as you continue to use and import more content seems to be a common problem for any software and collection type. I found countless articles and youtube videos related to fixing Mircosoft's active directory metadata. This is pretty frightening, considering active directory is what many businesses use to allow employee access into the business. With so many errors to constantly fix, security becomes a major issue. On a personal note, I found numerous articles on how to clean and maintain your metadata for your iTunes collection. Perhaps we need one schema that includes everything, enabling easy communications from one system to another.
1) If metadata types are most often based on English, how does this affect resources written in a different language, or catologers, archivists, etc. cataloging from non-English speaking countries? How does this affect the globalization of metadata?
2) Can metadata ever be consistent over any lengthy period of time? So many people go into creating a record.
3) Do we need deeper layers of metadata than we used to? Do we want to know more or is there more to know?
1 - Language is a very interesting and complicated topic and I can only speak from my limited experience. From what I've seen, finding aids and other information for a collection, regardless of language, are typically written in English as an entry point to the collection, with metadata being written in several languages, depending on the original language of the material and the research focus of the library. For example, the Benson's finding aids are in English (aside from perhaps collection titles) but interfaces to access audiovisual materials usually contain transcripts or metadata in the original language. I would love to see the creation of more bilingual finding aids in the US, especially when materials are written in a non-English language, as it seems kind of silly and limiting to write about the materials in English when non-English speakers will have difficulty accessing materials written in their own language, whereas English speakers may find the material but be unable to use it.
1. Metadata is as “the sum total of what one can say about any information object at any level of aggregation”, is that means everything, which can be addressed and manipulated by a human or a system as a discrete entity, has their metadata? If so, could metadata ha its own metadata?
2. In this article, it mentioned the Dublin Core Metadata Element Set (DC), which identifies a small set of metadata elements that can be used to describe and search across a wide variety of information resources on the World Wide Web. How could this standard ensure different kinds of descriptive metadata are able to interoperate with each other?
3. In table 2, since metadata could be classified with the methods of metadata creation, it said that manual metadata created by humans. If so, I wonder if it is a sub-classification that the manual metadata could classify into “lay metadata” and “experts metadata” because both of them are created by humans.
1. What methodology did the author use to create the categories for metadata types/functions (pg. 5) and metadata characteristics/attributes (pg. 6-7)? Have any of these changed since this article's publication in 2000? Have any new metadata types/functions and metadata characteristics/attributes come into existence since this time? 2. In what ways do the different metadata types and metadata attributes relate to one another? Are there particular relationships between certain types/functions and specific attributes? Or perhaps, are all of the various attributes found within each type? 3. In regard to metadata, I'm curious as to the ways in which expanding information use and usership is being addressed by information professionals? What changes have been made in metadata structure and creation processes as a result of these efforts?
In the JISC article, Technical and Preservation are not classified as different types of metadata but instead fall under the administrative banner. In addition to that difference, Gilliland does not use Structural as a type of metadata. Is one group of types better than the other, in the sense that one provides a better overall template? Or does it not matter how the types are arranged?
On page 7, Gilliland notes that metadata management and metadata creation have become exceedingly complex when thrown into an information life cycle. With all the competing and complementary life cycles going on in the information world would it not be best to define a singular semantic style?
In the “Expanding use” section on page 10, Gilliland brings up an interesting point regarding how different users interact with information and the need for metadata to be accessible to a wide range of users. Because of this need, especially in today’s world and the world going forward, does metadata simply need a “unified field theory” or does it need to wear different masks depending on the individual accessing it?
1. The author mentions that acquisition records, exhibition catalogs, and use data are all examples of these, even though they are still largely created in paper form. Is it too subjective? The author should provide essential evident that support they are largely created in paper from.
2. In the article we read last week, it is mentioned that the current multimedia search methodology is still based on the text on the same page with the multimedia. However, individuals digitizing images might contains information in the header field. Should we use the header field as searching area?
3. “Metadata can come from a variety of courses” It can be supplied by a human, or created automatically by a computer. Should the accuracy of automatically generated metadata be a problem, since we had learn the quality of it in former class?
1. On page 3, Gilliland mentions that “when applied outside the repository, the term metadata acquires and even broader scope” and goes on to list what metadata might mean to certain communities. The list is somewhat long and varied and I’m sure doesn’t even cover a small piece of it. Given this, is it important to limit the definition of what metadata is and can be? Does having so many understandings of metadata cause confusion as to its purpose maybe even specifically within the information field?
2. Gilliland writes, in talking about metadata standards, that there is an exciting but potentially bewildering array of organizational and descriptive schemas from which information professionals can choose. I’m curious, though, with the emphasis placed on interoperability of these systems, if it’s possible to truly make a wrong decision in deciding which to use? Is it more a matter of which standard best suits your institution?
3. Towards the end, when discussing how metadata aids in expanding use of digital information objects, Gilliland writes about how information professionals need to figure out how to make materials more “intellectually accessible” to a wider range of users. She states that “the new communities of users may have significantly different needs to those of the traditional users for whom many existing information services have been designed” (pg. 10). I’m assuming traditional users would be scholars (which is interesting in itself), but how exactly could you adapt metadata to make materials more “intellectually accessible”?
2. Interoperability is definitely important, but I wonder (based on some allusions in the JISC article) whether these descriptive schemas are consistently interoperable in practice. Do shifts between metadata schemas ever create serious problems, or does the field have interoperability pretty well figured out? (I'm new to the field, myself, so I am not familiar with how well different schemas interact with each other.)
1. When discussing library metadata Gilliland distinguishes "structural and content standards" like MARC from "authority forms" like LCSH. Please explain what the differences are between the two.
2. Number 5 on the author's list of Little-Known Facts states "One information object's metadata can simultaneously be another information object's data." How would that work? What would the relationship be between the two?
3. I've noticed that interoperability is often mentioned in articles on metadata. With all of the metadata standards discussed in these articles like MARC, EAD, RKMS, Dublin Core, etc., do all of these provide for interoperability among them?
1) One important thing this article highlighted is the varied definitions, uses, and user needs pertaining to metadata in different disciplines. On one level, the variation makes sense, because naturally different disciplines and institutions will often have extremely different needs. But how much of this variation is necessary and how much of it hinders communication and interdisciplinarity rather than helping it?
2) Since this article was last revised 2000, metadata has become increasingly more digital, but the article acknowledges the continued existence and importance of analog metadata. Does digital metadata tend to look and behave similarly to analog? What are the differences, and how are they shaped by the different functionalities of each?
1. With sensors becoming a part of more and more devices that are more common in every day life, how will the growing collection of the metadata of our lives be stored and who will have access to it? With the services offered by large companies like Google who have access to significant amounts of personal data, I think this is a relevant question.
2. Is deep metadata utilized in the act of online searching? What I mean is, besides things like the date or location of the publication, there must be some collection of metadata on internet search results that users don't see. Is that invisible metadata used in regular internet searches?
3. The metadata of process (i.e. the metadata not about the object but about the process through which the object has been through) is potentially very large and only becomes larger the older an object is. Similar to archving in general, who decides what metadata is important about the metadata of process and should be recorded?
1. In the “Other Little Known Facts about Metadata” box, #5 says that “One information object’s metadata can simultaneously be another information object’s data.” What does this mean? How can the metadata be a separate object? The author doesn’t add any additional clarification to this point.
2. The author breaks down the categories of metadata, providing characteristics and examples of each. I think most of them make sense except for the status category. It is unclear to me how the listed characteristics of this kind of metadata relate. I think the examples given also seem to fall into other categories. For example, rights information and directory structure fall under source of metadata. User transaction logs should fall under the use category. What qualities of the object is status metadata trying to account for?
1. In ‘Introduction to metadata-Setting up the stage’ , the author puts forth that a large component of archival and museum activities has traditionally focused on context. When a broader term like context is described, is metadata included in it? Is metadata a description of the context rather than the context itself? 2. In the lifecycle of objects contained in a digital information system, should encryption also be a stage? Can a metadata also have important and sensitive information? Is protected or encrypted metadata used in archives and digital systems? 3. Under ‘Other little notes about metadata’, the author conveys that one information object’s metadata can simultaneously be another information object’s data. In what context is this applicable? Isn't data a big entity when compared to metadata and contains much more information? How to metadata and data become comparable?
1. Gilliland asserts that it is more important than ever that "users of digital content understand the critical roles of different types of metadata..." and refers to children learning about provenance and authority in school. I'm kind of old, but I'm wondering how widespread this is. Our public education system, at least, does not seem invested in teaching rigorous critical thinking skills generally; the inability of my rural relatives to, at a bare minimum, consult Snopes.com before forwarding me political chain letters filled with xenophobic hysteria leads me to agree with the author but doubt the real commitment of state education boards to teach information literacy. What can we, as information professionals - especially those of us not working in school libraries or the K-12 system - do to further this type of education, especially in the digital information age?
2. A very broad (meta, if you will) question about metadata goals: how much context should we strive to provide for objects and information? Total context is impossible, as we cannot know and incorporate the experience, values, and needs of every potential individual user (not only because of manpower and server space, but because of the nature of human cognitive processes and of information itself). However, both in these readings and in those for my metadata class, the impression is definitely given that metadata is expanding and becoming more and more sophisticated. How do we as information professionals think about the scope of "context" generally, and how do we apply this theoretical background to on-the-ground cataloging in our own specific institutions, serving specific users and communities?
3. I'm personally very interested in folksonomies, and the increase in potential because of the web. I'm an avid, longtime Flickr user, and I've created my own metadata involving tags, sets, collections, groups, and social media. I'd really enjoy a comparative discussion of the user-created metadata of various online communities that allow significant user-direction and sharing, such as Flickr. What are some further reading lists on this topic, as well?
1. Under the section "Conclusion and Outstanding Questions" it says “If thorough, consistent metadata has been created, it is possible to conceive of it being used in an almost infinite number of new ways to meet the needs of non-traditional users, for multi-versioning, and for data mining.” If metadata is always changing and being updated is it really consistent? What if someone wants to access metadata from years ago only to find it has been discarded or updated?
2. This article listed five types of metadata, administrative, descriptive, preservation, technical, and use. Are there any other kinds of metadata? I’m guessing there are depending on the institution, but are there any standardized types of metadata recognized universally?
3. This article mentioned many types of different metadata programs. Ex.) AMC, MARC, ISAD, EAD, RLIN AMC, etc. What is the lifecycle of metadata for each of these different types? Are all these different programs necessary and is it possible to convert metadata from one system to another automatically or does it have to be done by hand/keyboard?
1. On the first page, the author said that "In general all information objects...have three features — content, context, and structure — all of which can be reflected through metadata." Later,on page 5, she also introduced four types of metadata. So, Which type(s) could cover the three features of information?
2. What did the author mean when she mentioned "intellectual information" and "intellectual access to content"? Does she mean "digital format"? And, how does metadata make an influence on this type of information? It is difficult to understand and imagine without definition and examples.
3. My third question is about the attributes and characteristics of metadata in Table 2. I think they look much like "metadata about metadata", or high-level metadata. So, by following this method, would there be a hierarchy of metadata, in which metadata about information objects lays on the bottom and "metadata about metadata" on higher levels?
1. In this article the author describes several different metadata schemas. She describes metadata schemas that are in use in libraries, like MARC, and in archives, like EAD. Which of the four different metadata categories that are described in the JISC article do you think that these schemas fall into? Do you think that the library and archival schemas would fall into different categories due to the different requirements that they have? 2. In this article the author states that metadata can be created at every different level of the lifecycle of a digital information object. She states that different types of metadata can be created at different points of the lifecycle. At which point in the information lifecycle do you thin that the erroneous metadata from the Google Books database was created? 3. The author states that one of the reasons that metadata is important is because it facilitates the accessibility of information objects. By including metadata with an information object you are not only making easier to find that object but you make it possible to create virtual collections that include that object and other objects that use the same metadata standard that might not be physically located together. Do you think that this ability of metadata to create virtual collections through collaboration will lead to a single metadata standard that is used in all fields to describe different types of information objects? Do you think that this would be a good idea?
1. On page 4 the author states that the Dublin Core Metadata Set can be used by anyone to describe a wide variety of information resources and then states that “such metadata standards are necessary in order to ensure that different kinds of descriptive metadata are able to interoperate with each other.” How interoperable is Dublin Core if there are no standards to control how the information is put into the fields? Does the information become less interoperable if one person decides to enter names in direct order and another in inverted order? Does this matter? 2. On page 7 there is mention of metadata registries and schema record keeping. Is this a true solution to having different communities develop different schemas and will this actually help to make different schemas interoperable? What exactly would/do these registries store? 3. I was confused by the fifth little known fact about metadata on page 9; “one information object’s metadata can simultaneously be another information object’s data.” I can’t seem to think of an example of this and didn’t see one in the reading. Any ideas on how a situation like this could occur?
3. An example I'd imagine would work would be the use of circulation statistics for items in a library to aid in decisions about material deaccession. Information about item usage is recorded and stored as metadata for that particular item. When assessments are done to consider removing items from the collection, these pieces of metadata become key data used to make that decision. Maybe a situation like that is what the author was referencing?
Q1: The article mentioned that more highly structured an information object is, the more that structure can be exploited for searching, manipulation, and interrelating with other information objects. Is there a balance point for the information object`s structure? Is that possible that the exceedingly structured information object will make the information object hard to search and manipulate? Q2: In the Table 2- Attributes and Characteristics of Metadata, the author categorized the metadata into structured metadata and unstructured metadata and stated unstructured metadata do not conform to a predictable structure. Is there a conflict occurred with the definition of the metadata? Can the unstructured metadata still be called “metadata”? Q3: The author uses “object”(sometimes “information object”) instead of “information”throughout the article. Is the scope of “object”broader than the scope of “information”? Is there any special reasons why the author emphasized the “object”?
1. On pages 2 and 3, Gilliland brings up “highly-structured information objects.” Does that just refer to something like an article that is highly organized and uses lots of sections and headings etc? Or is it referring to another kind of structure?
2. What is a metadata registry, and who uses it?
3. What does it mean to “to negotiate between descriptive surrogates of information objects and digital versions of the objects themselves”? Is this something that users often have a problem doing?
1. The author mentions that as information artifacts move through the information life cycle, new layers of metadata can be created and stored either with the item or separately. Is there a benefit one way or another for either storage method?
2. I found it interesting to consider the importance metadata has in preserving the contexts of information artifacts. I'd not really considered how "information about information" could aid in the understanding of archival and museum collections in particular.
1. Gilliland says that there are three features: content, context, and structure that can be reflected through metadata. I can understand content is intrinsic and context is extrinsic, but I wonder why structure can be intrinsic or extrinsic. I think structure is intrinsic.
2. In table 1, it shows us some different types of metadata. I wonder whether there are some overlaps among different types. For example, I think the preservation and use are somewhat the same. I mean preservation is a kind of use.
3. In table 2, for the methods of metadata creation, which one is more popular, automatic or manual metadata creation in practice? Why?
1. Have there been any in-depth looks to see how 'tagging' hierarchies on websites like YouTube have been influenced by metadata standards? Even if not by strict adoption of established terms, at least in how they are organized and used to make searching simpler for users?
ReplyDelete2. "Metadata can come from a variety of sources." Are there any good programs out there for automated capture/creation of meta data other than what gets embedded in a file when it is created?
3. I really liked how the author made explaining metadata very simple and easy to understand, but also that she made connections to other types of metadata people are not only familiar with, but might not have realized fit the definition of metadata. We use so many slang terms for it like 'tags' and 'filters' and 'keywords', I think when people come across the term they initially panic (particularly when there's a standard attached to it, to say nothing of Dublin Core being flexible to work with).
1. The author says there is an increasing need for system designers to incorporate metadata in order to move the system forward across time. So is that the more metadata the better? What’s the maximum length of metadata? Would it be possible that metadata becomes larger than the data it is describing?
ReplyDelete2. If the metadata reaches a certain scale, would there be metadata about metadata? If yes, what’s the difference between the nature of metadata and that of the metadata of metadata?
3. What’s the life cycle of metadata? Would it go through the same cycle as the information it is describing?
1 - Metadata is one arena where I see the post-custodial archival method really enabling archives to do their work both well and efficiently. In a post-custodial scenario (such as the HRDI's partnerships with the Kigali Genocide Memorial and Witness) the record creators generate all metadata and the archive attaches it to the digital objects. For example, in my work environment, I take metadata and video files from the Kigali Genocide Memorial and update said files in Glifos on the Genocide Archive of Rwanda website without adding any new or additional content, aside from quality control measures (making sure titles/file identifiers are consistent, checking for spelling errors, etc) and then the files are good to go. This gives record creators more control over how their materials are accessed and contextualized and provides archives with a more efficient system of generating metadata. So I guess my question is, why aren't more archives using the post-custodial method? It's generally viewed as a "pie in the sky" ideal, but has worked quite well for the HRDI.
ReplyDelete2 - How can we educate records creators, especially those working in digital media, about metadata to ensure both enduring access to their materials, and preserve the evidentiary value of their records? Will there eventually be more positions where archivists work in a consultant capacity, such as Witness or Rhizome, where creators are given training and tools to generate their own metadata, or is this just wishful thinking because I desperately want that to be my future job?
1. As in the “Introduction to Metadata” article, Gilliland-Swetland here discusses the different types of metadata. She presents five categories: administrative, descriptive, preservation, technical, and use. However, in the previous article, the author listed only four – descriptive, administrative, structural, and use. While both authors recognized that the categories were broad and not all-encompassing, the fact that only three categories overlap is interesting. Is this change in metadata types related to the difference in time between the articles? Have our terms changed because our understanding has changed? Or is this just a difference of opinion?
ReplyDelete2. Later in the article, the author talks about how large the accumulation of metadata can be, and that there is and will be a need to manage it as it is incorporated into information objects. She claims that “metadata registries and schema recordkeeping systems . . . are likely to develop” (7). Have these registries developed as the author predicted? Is her concept of the future of metadata our reality today?
3. The author frequently states the importance of having consistent metadata. In one instance, she states that, “metadata can also make it possible to search across multiple collections or to create virtual collections from materials that are distributed across several repositories, but only if the descriptive metadata are the same or can be mapped across each site” (9). Does she mean here that all metadata should have the same vocabularies? Or that it should have a standard format? Is she predicting that metadata could be used in different communities and should therefore be translatable between them?
1. Gilliland states that “well-crafted metadata results in the best metadata management in the short and long-term.” It is just often difficult to anticipate all of the metadata that may later be needed, and retrofitting metadata onto a collection is a Herculean task, as observed by the posts about Google Books. Do common metadata standards help this task by providing useful templates?
ReplyDelete2. As we progress in our informational needs, we seem to require deeper and deeper layers of metadata to try and organize increasingly large collections of data. As investigated in the article about legal document searching, the amount of information available continues to explode. Will our metadata schema be scalable in these increasing levels? Has this been tested much?
3. This leads to the question of automation of metadata. Often now when one creates a digital object, it creates a certain amount of metadata about itself (such as a photograph) as a means of authentication and uniqueness. However, automation of metadata can go terribly awry. Does it seem likely that some level of human intervention or supervision is going to continue to be needed in the foreseeable future?
This comment has been removed by the author.
ReplyDelete1 The author mentions mapping between common data elements or metadata on the page 4. How does this works? How to define the coverage? And for what purposes we need to map the metadata?
ReplyDelete2 On table 2, the author gives nature of metadata. One is lay metadata created by persons who are neither subject nor information specialists, often the original creator of the information object, another one is expert metadata created by either subject or information specialists, often not the original creator of the information object. Is there a gap of understanding information object between original workers and expert who edits metadata? And how about the gap between expert and the user?
3 On the "Other Little-Known Facts about Metadata" part, the author states that metadata can come from a variety of sources, such as human creation or automatically created by a computer, or inferred through a relationship to another resource such as a hyperlink. How this inferring progress works? Who to infer, human or computer generated?
In response to your #2, I believe there is definitely a gap of understand between the original creators and the metadata editors. At my internship, I have to add metadata to digital files that can be up to 7 years old. Often times, the creator made abbreviations in the file name that I can not recognize what it means. The abbreviation obviously is of some kind of importance though for the creator to have added it. Often times I have no idea what the photo was originally created for or the context to be able to add metadata about its creation. This happens all the time in archives too, in that the archivist cannot understand a file order which probably made sense to the creator. I think that the nature of metadata category should separate the metadata added by the creator. I would suspect this metadata to be the most accurate as the creator understands the meaning or purpose for which they created the object. I think that the original creator would be the most expert in creating some metadata.
Delete1. I've read Heather's first question and I got the one almost the same. Two articles present different categories about metadata and only part of the categories are overlapped. Does the difference between the articles suggest information professionals in academic circle have not make an agreement on some of key points of metadata?
ReplyDelete2. To state why metadata is important, the author mentioned the retention of context. In this section, the author said "In an archive, for example, by documenting the content, context, and structure of an archival record, metadata in the form of an archival finding aid is what helps to
distinguish that record from decontextualized information." Does this feature of metadata also apply the objects on the web? I mean, for example, copying images from other sites and pasting them on another page could lose their original context information for now. How could metadata change the situation?
3. In the section of "Legal issues": Metadata allows repositories to track the many layers of rights and reproduction information that exist for information objects and their multiple versions. I'm wondering how it works. To protect legal issues, should metadata of every object essentially obtains information about its copy right?
1. On pg. 11, Gilliland-Swetland starts a sentence with “If thorough, consistent metadata has been created,…”. That is a big ‘if’. Considering all of the different people with varying levels of expertise, backgrounds, experience, etc., I am not sure it is even possible to create consistent metadata, even if a certain controlled vocabulary or standard is employed. What kind of measures do different institutions use to ensure that metadata (for example) is consistent in the same collection after one librarian leaves and another replaces her/him? Is this where knowledge management should play a larger role in institutions?
ReplyDelete2. Gilliland-Swetland refers to the information life cycle in her article. For a single object’s life cycle, does metadata change at each stage? If so, to what extent and is it possible to link that metadata together in a comprehensive and cohesive way? Or is it just dependent on the institution and what kind of metadata they are creating, be it descriptive, administrative, etc.?
3. On pg. 9, under the section ‘Increased accessibility’, the author states that different metadata standards allow users to search on the item/collection level, both within and across information systems. It is the ‘across information systems’ statement that I find hard to believe. I believe that metadata standards (or at least consistency) are hugely important and that they are probably normally adhered to within a single organization. But to search for materials cross-institutionally can be exceedingly difficult and frustrating, because everyone thinks they’re an exception to the rule. Different organizations may use the same set of Dublin Core but their personal prejudices and thoughts on how things should be inevitably vary, and lead to different sets of metadata, maybe even for the same object. Is this something that will ever be avoidable, or is it just part of being a human being and not a robot, that things are different from organization to organization, and even person to person?
1. While the author identified many attributes and characteristics of Metadata one thing I find troubling is the lack of concern for privacy and security. It seems like a common trend in technology that richer metadata is being collected on digital objects we create but I haven’t seen a lot of discussion about the potential for abuse in these technologies. I am thinking specifically about things like digital cameras that utilize GPS to attribute coordinates to each image you take, leaving a virtual record of every place you have been and a record of what you were doing. As metadata starts to become richer should we also be thinking about ways to ensure its protection?
ReplyDelete2. The author mentions that metadata can be used in planning for new systems by benchmarking technical data that is collected automatically by a computer. I am a little uncertain about what the author is talking about here?
3. With the creation of metadata from the source being so vital for accuracy in archives I am wondering how systems could incorporate features that encourage more intelligent file creation. This works quite will in programs like iPhoto which periodically ask the user to identify faces in order to sort pictures in more intelligent ways. I wonder if similar features could be built into programs like Microsoft Word or your email too.
This comment has been removed by the author.
ReplyDelete1. There are many types of metadata, and the article talks about MARC, AMC, ISAD, EAD, and SAA as examples. With so many different types, is there any standardization across all of them?
ReplyDelete2. Majority of the metadata are English based, I wonder if there would be issues when metadata is applied across different languages?
3. In searching around the web about metadata, keeping it accurate as you continue to use and import more content seems to be a common problem for any software and collection type. I found countless articles and youtube videos related to fixing Mircosoft's active directory metadata. This is pretty frightening, considering active directory is what many businesses use to allow employee access into the business. With so many errors to constantly fix, security becomes a major issue. On a personal note, I found numerous articles on how to clean and maintain your metadata for your iTunes collection. Perhaps we need one schema that includes everything, enabling easy communications from one system to another.
1) If metadata types are most often based on English, how does this affect resources written in a different language, or catologers, archivists, etc. cataloging from non-English speaking countries? How does this affect the globalization of metadata?
ReplyDelete2) Can metadata ever be consistent over any lengthy period of time? So many people go into creating a record.
3) Do we need deeper layers of metadata than we used to? Do we want to know more or is there more to know?
1 - Language is a very interesting and complicated topic and I can only speak from my limited experience. From what I've seen, finding aids and other information for a collection, regardless of language, are typically written in English as an entry point to the collection, with metadata being written in several languages, depending on the original language of the material and the research focus of the library. For example, the Benson's finding aids are in English (aside from perhaps collection titles) but interfaces to access audiovisual materials usually contain transcripts or metadata in the original language. I would love to see the creation of more bilingual finding aids in the US, especially when materials are written in a non-English language, as it seems kind of silly and limiting to write about the materials in English when non-English speakers will have difficulty accessing materials written in their own language, whereas English speakers may find the material but be unable to use it.
Delete1. Metadata is as “the sum total of what one can say about any information object at any level of aggregation”, is that means everything, which can be addressed and manipulated by a human or a system as a discrete entity, has their metadata? If so, could metadata ha its own metadata?
ReplyDelete2. In this article, it mentioned the Dublin Core Metadata Element Set (DC), which identifies a small set of metadata elements that can be used to describe and search across a wide variety of information resources on the World Wide Web. How could this standard ensure different kinds of descriptive metadata are able to interoperate with each other?
3. In table 2, since metadata could be classified with the methods of metadata creation, it said that manual metadata created by humans. If so, I wonder if it is a sub-classification that the manual metadata could classify into “lay metadata” and “experts metadata” because both of them are created by humans.
1. What methodology did the author use to create the categories for metadata types/functions (pg. 5) and metadata characteristics/attributes (pg. 6-7)? Have any of these changed since this article's publication in 2000? Have any new metadata types/functions and metadata characteristics/attributes come into existence since this time?
ReplyDelete2. In what ways do the different metadata types and metadata attributes relate to one another? Are there particular relationships between certain types/functions and specific attributes? Or perhaps, are all of the various attributes found within each type?
3. In regard to metadata, I'm curious as to the ways in which expanding information use and usership is being addressed by information professionals? What changes have been made in metadata structure and creation processes as a result of these efforts?
In the JISC article, Technical and Preservation are not classified as different types of metadata but instead fall under the administrative banner. In addition to that difference, Gilliland does not use Structural as a type of metadata. Is one group of types better than the other, in the sense that one provides a better overall template? Or does it not matter how the types are arranged?
ReplyDeleteOn page 7, Gilliland notes that metadata management and metadata creation have become exceedingly complex when thrown into an information life cycle. With all the competing and complementary life cycles going on in the information world would it not be best to define a singular semantic style?
In the “Expanding use” section on page 10, Gilliland brings up an interesting point regarding how different users interact with information and the need for metadata to be accessible to a wide range of users. Because of this need, especially in today’s world and the world going forward, does metadata simply need a “unified field theory” or does it need to wear different masks depending on the individual accessing it?
1. The author mentions that acquisition records, exhibition catalogs, and use data are all examples of these, even though they are still largely created in paper form. Is it too subjective? The author should provide essential evident that support they are largely created in paper from.
ReplyDelete2. In the article we read last week, it is mentioned that the current multimedia search methodology is still based on the text on the same page with the multimedia. However, individuals digitizing images might contains information in the header field. Should we use the header field as searching area?
3. “Metadata can come from a variety of courses” It can be supplied by a human, or created automatically by a computer. Should the accuracy of automatically generated metadata be a problem, since we had learn the quality of it in former class?
1. On page 3, Gilliland mentions that “when applied outside the repository, the term metadata acquires and even broader scope” and goes on to list what metadata might mean to certain communities. The list is somewhat long and varied and I’m sure doesn’t even cover a small piece of it. Given this, is it important to limit the definition of what metadata is and can be? Does having so many understandings of metadata cause confusion as to its purpose maybe even specifically within the information field?
ReplyDelete2. Gilliland writes, in talking about metadata standards, that there is an exciting but potentially bewildering array of organizational and descriptive schemas from which information professionals can choose. I’m curious, though, with the emphasis placed on interoperability of these systems, if it’s possible to truly make a wrong decision in deciding which to use? Is it more a matter of which standard best suits your institution?
3. Towards the end, when discussing how metadata aids in expanding use of digital information objects, Gilliland writes about how information professionals need to figure out how to make materials more “intellectually accessible” to a wider range of users. She states that “the new communities of users may have significantly different needs to those of the traditional users for whom many existing information services have been designed” (pg. 10). I’m assuming traditional users would be scholars (which is interesting in itself), but how exactly could you adapt metadata to make materials more “intellectually accessible”?
2. Interoperability is definitely important, but I wonder (based on some allusions in the JISC article) whether these descriptive schemas are consistently interoperable in practice. Do shifts between metadata schemas ever create serious problems, or does the field have interoperability pretty well figured out? (I'm new to the field, myself, so I am not familiar with how well different schemas interact with each other.)
Delete1. When discussing library metadata Gilliland distinguishes "structural and content standards" like MARC from "authority forms" like LCSH. Please explain what the differences are between the two.
ReplyDelete2. Number 5 on the author's list of Little-Known Facts states "One information object's metadata can simultaneously be another information object's data." How would that work? What would the relationship be between the two?
3. I've noticed that interoperability is often mentioned in articles on metadata. With all of the metadata standards discussed in these articles like MARC, EAD, RKMS, Dublin Core, etc., do all of these provide for interoperability among them?
1) One important thing this article highlighted is the varied definitions, uses, and user needs pertaining to metadata in different disciplines. On one level, the variation makes sense, because naturally different disciplines and institutions will often have extremely different needs. But how much of this variation is necessary and how much of it hinders communication and interdisciplinarity rather than helping it?
ReplyDelete2) Since this article was last revised 2000, metadata has become increasingly more digital, but the article acknowledges the continued existence and importance of analog metadata. Does digital metadata tend to look and behave similarly to analog? What are the differences, and how are they shaped by the different functionalities of each?
1. With sensors becoming a part of more and more devices that are more common in every day life, how will the growing collection of the metadata of our lives be stored and who will have access to it? With the services offered by large companies like Google who have access to significant amounts of personal data, I think this is a relevant question.
ReplyDelete2. Is deep metadata utilized in the act of online searching? What I mean is, besides things like the date or location of the publication, there must be some collection of metadata on internet search results that users don't see. Is that invisible metadata used in regular internet searches?
3. The metadata of process (i.e. the metadata not about the object but about the process through which the object has been through) is potentially very large and only becomes larger the older an object is. Similar to archving in general, who decides what metadata is important about the metadata of process and should be recorded?
1. In the “Other Little Known Facts about Metadata” box, #5 says that “One information object’s metadata can simultaneously be another information object’s data.” What does this mean? How can the metadata be a separate object? The author doesn’t add any additional clarification to this point.
ReplyDelete2. The author breaks down the categories of metadata, providing characteristics and examples of each. I think most of them make sense except for the status category. It is unclear to me how the listed characteristics of this kind of metadata relate. I think the examples given also seem to fall into other categories. For example, rights information and directory structure fall under source of metadata. User transaction logs should fall under the use category. What qualities of the object is status metadata trying to account for?
1. In ‘Introduction to metadata-Setting up the stage’ , the author puts forth that a large component of archival and museum activities has traditionally focused on context. When a broader term like context is described, is metadata included in it? Is metadata a description of the context rather than the context itself?
ReplyDelete2. In the lifecycle of objects contained in a digital information system, should encryption also be a stage? Can a metadata also have important and sensitive information? Is protected or encrypted metadata used in archives and digital systems?
3. Under ‘Other little notes about metadata’, the author conveys that one information object’s metadata can simultaneously be another information object’s data. In what context is this applicable? Isn't data a big entity when compared to metadata and contains much more information? How to metadata and data become comparable?
1. Gilliland asserts that it is more important than ever that "users of digital content understand the critical roles of different types of metadata..." and refers to children learning about provenance and authority in school. I'm kind of old, but I'm wondering how widespread this is. Our public education system, at least, does not seem invested in teaching rigorous critical thinking skills generally; the inability of my rural relatives to, at a bare minimum, consult Snopes.com before forwarding me political chain letters filled with xenophobic hysteria leads me to agree with the author but doubt the real commitment of state education boards to teach information literacy. What can we, as information professionals - especially those of us not working in school libraries or the K-12 system - do to further this type of education, especially in the digital information age?
ReplyDelete2. A very broad (meta, if you will) question about metadata goals: how much context should we strive to provide for objects and information? Total context is impossible, as we cannot know and incorporate the experience, values, and needs of every potential individual user (not only because of manpower and server space, but because of the nature of human cognitive processes and of information itself). However, both in these readings and in those for my metadata class, the impression is definitely given that metadata is expanding and becoming more and more sophisticated. How do we as information professionals think about the scope of "context" generally, and how do we apply this theoretical background to on-the-ground cataloging in our own specific institutions, serving specific users and communities?
3. I'm personally very interested in folksonomies, and the increase in potential because of the web. I'm an avid, longtime Flickr user, and I've created my own metadata involving tags, sets, collections, groups, and social media. I'd really enjoy a comparative discussion of the user-created metadata of various online communities that allow significant user-direction and sharing, such as Flickr. What are some further reading lists on this topic, as well?
1. Under the section "Conclusion and Outstanding Questions" it says “If thorough, consistent metadata has been created, it is possible to conceive of it being used in an almost infinite number of new ways to meet the needs of non-traditional users, for multi-versioning, and for data mining.” If metadata is always changing and being updated is it really consistent? What if someone wants to access metadata from years ago only to find it has been discarded or updated?
ReplyDelete2. This article listed five types of metadata, administrative, descriptive, preservation, technical, and use. Are there any other kinds of metadata? I’m guessing there are depending on the institution, but are there any standardized types of metadata recognized universally?
3. This article mentioned many types of different metadata programs. Ex.) AMC, MARC, ISAD, EAD, RLIN AMC, etc. What is the lifecycle of metadata for each of these different types? Are all these different programs necessary and is it possible to convert metadata from one system to another automatically or does it have to be done by hand/keyboard?
1. On the first page, the author said that "In general all information objects...have three features — content, context, and structure — all of which can be reflected through metadata." Later,on page 5, she also introduced four types of metadata. So, Which type(s) could cover the three features of information?
ReplyDelete2. What did the author mean when she mentioned "intellectual information" and "intellectual access to content"? Does she mean "digital format"? And, how does metadata make an influence on this type of information? It is difficult to understand and imagine without definition and examples.
3. My third question is about the attributes and characteristics of metadata in Table 2. I think they look much like "metadata about metadata", or high-level metadata. So, by following this method, would there be a hierarchy of metadata, in which metadata about information objects lays on the bottom and "metadata about metadata" on higher levels?
1. In this article the author describes several different metadata schemas. She describes metadata schemas that are in use in libraries, like MARC, and in archives, like EAD. Which of the four different metadata categories that are described in the JISC article do you think that these schemas fall into? Do you think that the library and archival schemas would fall into different categories due to the different requirements that they have?
ReplyDelete2. In this article the author states that metadata can be created at every different level of the lifecycle of a digital information object. She states that different types of metadata can be created at different points of the lifecycle. At which point in the information lifecycle do you thin that the erroneous metadata from the Google Books database was created?
3. The author states that one of the reasons that metadata is important is because it facilitates the accessibility of information objects. By including metadata with an information object you are not only making easier to find that object but you make it possible to create virtual collections that include that object and other objects that use the same metadata standard that might not be physically located together. Do you think that this ability of metadata to create virtual collections through collaboration will lead to a single metadata standard that is used in all fields to describe different types of information objects? Do you think that this would be a good idea?
1. On page 4 the author states that the Dublin Core Metadata Set can be used by anyone to describe a wide variety of information resources and then states that “such metadata standards are necessary in order to ensure that different kinds of descriptive metadata are able to interoperate with each other.” How interoperable is Dublin Core if there are no standards to control how the information is put into the fields? Does the information become less interoperable if one person decides to enter names in direct order and another in inverted order? Does this matter?
ReplyDelete2. On page 7 there is mention of metadata registries and schema record keeping. Is this a true solution to having different communities develop different schemas and will this actually help to make different schemas interoperable? What exactly would/do these registries store?
3. I was confused by the fifth little known fact about metadata on page 9; “one information object’s metadata can simultaneously be another information object’s data.” I can’t seem to think of an example of this and didn’t see one in the reading. Any ideas on how a situation like this could occur?
3. An example I'd imagine would work would be the use of circulation statistics for items in a library to aid in decisions about material deaccession. Information about item usage is recorded and stored as metadata for that particular item. When assessments are done to consider removing items from the collection, these pieces of metadata become key data used to make that decision. Maybe a situation like that is what the author was referencing?
DeleteQ1: The article mentioned that more highly structured an information object is, the more that structure can be exploited for searching, manipulation, and interrelating with other information objects. Is there a balance point for the information object`s structure? Is that possible that the exceedingly structured information object will make the information object hard to search and manipulate?
ReplyDeleteQ2: In the Table 2- Attributes and Characteristics of Metadata, the author categorized the metadata into structured metadata and unstructured metadata and stated unstructured metadata do not conform to a predictable structure. Is there a conflict occurred with the definition of the metadata? Can the unstructured metadata still be called “metadata”?
Q3: The author uses “object”(sometimes “information object”) instead of “information”throughout the article. Is the scope of “object”broader than the scope of “information”? Is there any special reasons why the author emphasized the “object”?
1. On pages 2 and 3, Gilliland brings up “highly-structured information objects.” Does that just refer to something like an article that is highly organized and uses lots of sections and headings etc? Or is it referring to another kind of structure?
ReplyDelete2. What is a metadata registry, and who uses it?
3. What does it mean to “to negotiate between descriptive surrogates of information objects and digital versions of the objects themselves”? Is this something that users often have a problem doing?
1. The author mentions that as information artifacts move through the information life cycle, new layers of metadata can be created and stored either with the item or separately. Is there a benefit one way or another for either storage method?
ReplyDelete2. I found it interesting to consider the importance metadata has in preserving the contexts of information artifacts. I'd not really considered how "information about information" could aid in the understanding of archival and museum collections in particular.
1. Gilliland says that there are three features: content, context, and structure that can be reflected through metadata. I can understand content is intrinsic and context is extrinsic, but I wonder why structure can be intrinsic or extrinsic. I think structure is intrinsic.
ReplyDelete2. In table 1, it shows us some different types of metadata. I wonder whether there are some overlaps among different types. For example, I think the preservation and use are somewhat the same. I mean preservation is a kind of use.
3. In table 2, for the methods of metadata creation, which one is more popular, automatic or manual metadata creation in practice? Why?