In my last post, I left you off with a list of the steps of a data life cycle model from DCC. In this continuing discussion, I wanted to return to the first step, conceptualize and planning, to see how it relates to metadata. In my example here and in the following related posts, metadata is used to describe any data in any form. Data could be a book, a tweet, a journal, a data set or any other data.
- Conceptualize: According to the DCC, the step of conceptualizing consists in conceiving and planning the creation of data. This includes capture methods and storage options. How does this work with metadata?
If we think of integrated library systems, the work of how to store metadata as well as how to capture these metadata has been conceived for us. Whether the ILS is bought from a vendor or is an open source platform, the general framework has been created, including but not limited to the structure/architecture of the system, how, where, when records are stored and related to one another and definitely many other fine points. These systems use MARC21 along with hard coded fields specific to the platform. Some ILS allow for more customization while others don’t. Also, the way in which we capture records is well known. There are various methods to load single or multiple records. Storage can be hosted or be located on an institution’s own servers. This is the realm of system librarians and many catalog/metadata librarians. Typically, ILS do not also store the data for which the metadata is being described. In other words, the data as a whole entity is received (either the physical item, its digital counterpart, or a digital item (perhaps a url or a stream for instance)) and then cataloged. In short, others (vendors or open source community) have pre-conceived and planned how metadata will be created, edited, and moved among other things. Of course, there was a discussion with clients as to customizing as well as storage options. But suffice to say that how metadata is created, managed and stored has been already conceptualized and planned to a large degree by another party.
If we think of digital repositories, this pre-conceptualizing of how to create, manage and store metadata can involve a grounds up process depending on the platform used. For example, if an institution acquires ContentDM, then like the ILS purchased form a vendor, the conceptualizing and planning has been done. On the other hand, if an institution decides to implement Fedora, an open source digital repository, the conceptualizing and planning of how to create, edit and manage metadata including storage options is from the ground up, requiring a higher level of input. Of course, the advantage of systems like ContentDM or even Digital Commons from BePress is that the programming and many of the decisions have already been taken. For those that lack the staff to implement an open source digital repository or even an open source ILS, this is a convenient and solid decision. Generally, even with digital repositories, metadata are being associated with data that are already together as one entity. Unlike the ILS, digital repositories tend to also store the data. For example, there will be metadata and the image (master tiff and related thumbnails) on a server or multiple servers somewhere.
But there is another aspect to conceptualize and plan metadata. This appears most visibly in the realm of metadata consultations. Here, when metadata specialists work with data creators, it is possible to advise them on the use of metadata before data creation. For example, you can advise a scientist of how to implement Dublin Core fields in a spreadsheet before that scientist begins his data gathering (i.e. data creation). Metadata can be conceptualized a number of different ways in terms of the way data creators and users use the data. Metadata can be embedded. They can be in an associated file. There are different types of metadata standards. Perhaps customization is important. It is crucial to conceptualize and plan for these various implementations of metadata at the creation and planning stage of data that includes storage that depend on the architecture of the platform used to manage and store metadata and the discovery/presentation layers.
Yes, metadata just like data require planning related to what, why, when, and how to implement standards, how to create and manage metadata and where and how to store metadata.
Let’s take a look at creation and receiving in the next post…