Introduction
Commonly understood as the ‘data about data’, metadata form a very strong foundation of the lifecycle and practice of a digital asset (Gilliland, 2016). It plays a vital role in various steps of this cycle, ranging from creation of the asset to its assessment, preservation, storage and retrieval. Having a clear understanding of the concept of metadata and what it entails is critical to the understanding of a digital asset and its accompanying lifecycle model. It not only gives us a more comprehensive description of what the lifecycle model is but its significance in the various steps throughout the cycle also give a clear picture about the development of a digital asset in the Digital Asset and Media Management practice.
Focussing specifically on The DCC&U: An Extended Digital Curation Lifecycle Model, the precise sequence of the different steps of the model provide us with an in-depth knowledge about how a digital asset is managed in the DAMM practice and how vital each stage is for the growth of the asset. The lifecycle approach ensures that every stage of the development of an asset is identified and implemented in a manner that it contributes to the lifespan of the asset. This approach maintains the credibility and usability of an asset by necessitating its effective administration. It also helps readers make sense of the DAMM practice across various domains and sectors. The lifecycle approach not only acts like a framework for future curators and practitioners in exploring the field of Digital Asset and Media Management but also as a guiding tool in understanding the various processes that are essential for successful curation (Higgins, 2008).
Within this, metadata provides a basis for the asset to complete the due course of its lifetime; it is a process that transforms a digital object into a digital asset and provides it the identity with which it exists and grows (Lipsey, 2010). Concentrating on the full lifecycle actions, the significance of metadata begins with the formative stage of ‘Description and Representation Information’ where metadata is essential for describing a particular asset in a manner that it is unique and different from other assets. In the DAMM practice, this simultaneously is essential for the preservation of the asset since correct and adequate description would lead to a smooth process of long term preservation. Similarly, detailed metadata would help DAMM practitioners manage the asset in a more effective manner, thus facilitating correct administration. In the sequential actions, as an asset is created after careful conceptualisation, the different types of metadata begin to show how the successful management of the asset requires appropriate metadata from its essential step. Creation of an asset is complete once it has been paired with the due administrative, descriptive, technical and structural metadata. Even if the data has been received from an external source, assigning the required metadata to the assets is integral for their sustenance through the subsequent stages of appraisal and selection of the assets. Once evaluated, the assets are transferred to archives and repositories in order to preserve them for future use and also extend the life-term of the assets. In the stage of ingestion and preservation of assets, metadata helps in keeping the assets and its information authentic and reliable in a manner that they can be stored in current time and be re-used when retrieved without losing their integrity and legitimacy. Within this, along with preservation metadata, technical metadata is essential for ensuring that the assets can be easily accessed by the producers and users in various formats and structures required. Among other reasons, insufficient or incorrect metadata might also occasionally lead to migration of the asset into another format or even its disposition (Higgins, 2008).
What is metadata?
Other than the earlier specified definition, metadata can also be understood as data that describes other data including the assets that are archived or stored repositories (Bachmann, 2010; Hurst, 2010). Metadata doesn’t necessarily have to be digital in nature, however with the advancement in technology and high volume of digital content being generated and stored, creation of detailed digital metadata has become a necessity. Metadata can be of different types – administrative, descriptive, preservation, structural, technical and usage- all of which perform specific tasks and are required at different stages of the lifecycle of an asset. The concept of metadata might also carry different degree of meaning and significance to different sections of people. Many professionals view metadata as the value-added information created in order to arrange, manage and increase the accessibility of information objects. It can be used by an individual for varying tasks ranging from finding a single image to digitizing and archiving an entire library’s records (Gilliland, 2016).
One essential attribute of the concept of metadata is its ability to translate a file into an asset by attaching certain properties to it. Metadata allows individuals, practitioners, organisations and systems to identify an asset and the rights and permissions associated with it thus allowing the user to modify, move and use the asset in a controlled manner. It takes a big bulk of unidentified and unorganized contents and promotes it to a point where the assets are categorized and preserved, ready for use in the future. In other words, metadata whether in the form of a small set of details or an extravagant compilation of facts is all the information about an asset required by an individual to know more about it in order to access it (Moon, 2009).
Metadata and the Lifecycle of a Digital Asset
Some of the basic characteristics of a digital asset include that it should be accurately described, freely accessible and useful in correct context. A user must be able to search for the correct asset and retrieve it in a suitable format. Metadata enables all such activities and more that ultimately caters to the lifecycle of a digital asset. Over the years, large amount of digital information has been created by individuals across the globe and volume of it constantly continues to grow. When put into a DAMM system, this information is of no use to those who created it or others for whom it has been created if they cannot easily access it, let alone search for it (Chagoya, 2010). Hence, at the very basic level of creation of digital asset, metadata is essential in providing the asset with its own identity. This descriptive metadata makes it possible to identify a certain asset and also provides it with definite properties that distinguish one asset from the other. The descriptive function of metadata assists in the organisation of the assets in digital asset management systems and repositories whether in its original form or copies. This metadata would be created by the producer of the digital asset as well as others who contribute to the lifecycle of the asset. For example, additional metadata might be added by practitioners involved in cataloguing, indexing and evaluation processes of the assets. Correct description of the metadata of a digital asset is highly essential because often users might examine in-depth the metadata of a digital asset in order to confirm its validation and authenticity.
This description also enables a producer and consumer to successfully find the exact asset upon searching for it and use it for the required purpose. Also, outlining metadata for digital content makes the process of searching highly refined as it is then possible to search against differing dimensions and categories (Hurst, 2010). As digital assets are growing in volume and number, there has been a realisation among practitioners that by decreasing the time spent searching for assets, the DAM systems will become more productive. In today’s technology driven time, DAM systems can store and manage thousands of digital assets and it is the metadata attached to these assets that makes searching for the precise asset extremely time and cost friendly. An example of the various benefits of comprehensive metadata can be viewed in the way of how search engines respond to users’ queries. When a customer logs in to Amazon to search for a product, the website also provides the customer with a section called ‘People who searched for this also viewed’. Similarly, when submitting a query on Google, the search engine often responds by asking ‘Did you mean…’. Such innovations and more are only possible due to collaboration with metadata structures (Moon, 2009). It is also possible that how one person searches for a particular asset is different from another, metadata in the form of keyword-search and tags is very helpful in such a situation.
In order to make the user experience better, metadata must first attempt to describe the asset correctly. Within metadata taxonomies, descriptive, administrative and technical metadata play an important role. Descriptive metadata provides information about the content and what it entails thus focussing on the ‘aboutness’ of the asset (Moon, 2009). For example, in the case of a digital image, descriptive metadata describes what the image is, who is in it or when it was taken. Administrative metadata in this example would provide information about the file number of the image or who produced it and for whom. This caters better management of the image. Technical metadata gives details about the size of image, its colour and resolution, the format in which it was produced and also the other formats it can be made available in. A combination of administrative and use metadata provide information about the rights and permissions associated with the digital asset, access requirements and protocols in varying cultures and communities, presence or absence of physical records and other such details that make it easier for a user while finding, accessing and using the asset. After appraisal and evaluation, preservation metadata is required to ensure that the asset has been documented correctly in order to facilitate the process of archiving and storage. Metadata is very useful in specifying the legal issues and ideas that accompany every asset. These are related to the rights, privacy concerns, licensing, copyright issues, reproduction restrictions and alike. Specifying this information is essential as it facilitates proper development of the asset and ensures that the asset can be effectively used by various individuals.
The correct implementation of a combination of these different types of metadata is not only essential till the stage of storage but even more when the asset is accessed in the future and retrieved for specific tasks and reasons. In an electronic framework, metadata acts like a tag associated to specific content. Associating this metadata to the content makes management, preservation and retrieval of digital assets highly efficient. A lack of efficient implementation in the previous stages might come in the way of retrieval and damage the availability of the asset for another user.
Preservation of digital assets using correct metadata is also essential for a very important factor that with constant change in technology, if the assets are not documented accurately with the right details, the asset might become obsolete and out-of-date for future retrieval. As in the case of audio and video tape technology, there has been a swift movement from cassettes to CDs and DVDs to small memory sticks and then virtual availability of audio and video in the form of clouds. In industries like broadcasting and digital content, preservation of the aging media assets is becoming increasingly important in order to save it from deterioration beyond any possibility of recovery.
Another reason why preservation of assets has become necessary is that all assets are not easily available and accessible in all regions and by all institutions. Correct metadata about the asset would provide the practitioners in the DAM organisations with information about accessibility and preservation of the asset for use by others becomes simpler.
There are certain instances when over the life time of an asset, it might become inactive or invalid for the situation for which it was created and preserved. In a situation, sometimes the asset after re-appraisal and detailed evaluation, the asset is disposed. Metadata acts as a key component even at this stage of the asset’s lifecycle where it facilitates the documentation of the disposition of the original asset or information objects associated to the original asset.
In the digital domain, a number of assets are not just present in their original forms, but also have their digital surrogates or copies that are stored in the form of different collections for different purposes. This is particularly visible in museums that maintain and manage collections of original assets. Such complex relations between assets requires rich and well-detailed metadata to maintain information about the various collections and the surrogates that would help in distinguishing one from another and also preserve them for future retrieval. Metadata in such cases becomes even more crucial because it not only relates to one asset and information about it, but rather a set of assets and the intricate relationship between such assets. It helps in specifying what is similar and different between each of the copies. Without the presence of precise metadata attached to these collections, the assets might be deemed unauthentic and invalid thus reducing their usability (Gilliland, 2016).
With the expansion of the digital world and technology along with the systems that facilitate the management of the growing digital content, there still exists a gap between communities and regions who can freely access the systems and digital content and others for whom there exist restrictions to complete accessibility. Metadata helps in specifying this information about an asset in the way of what legal ideas are attached to the asset, the format it is available and accessible in, the software(s) it might require for effective functioning and other such properties.
According to Ron Roszkiewicz, metadata contributes to the lifecycle of a digital asset in three primary ways – first, availability of correct digital data through comprehensive metadata leads to better analysis and reports that leads to better decisions resulting in increasing efficiency and productivity of organisations. Second, as in every content management lifecycle, ultimately most, if not all, digital assets are archived and stored for future use. Such preservations actions when implemented along with correct and accurate metadata attached to the digital assets helps in future retrieval and makes re-use of those assets possible. Third, metadata can also be available in the form of preferences and choices of the users based on their feedback. This is important for the ‘User Experience’ aspect of a lifecycle of a digital asset, where the feedback can modify the data in order to improve the result for the users. A way of bringing more clarity to the results is by building comprehensive metadata taxonomies that helps in filtering the results further, thus making the process simpler.
Case Study: An interview with Joseph Santucci by John Horodyski
piXlogic: Why automated metadata is the new necessity!
In the recent past years, there has been a recognition of importance of metadata in Digital Asset and Media Management practice and the digital systems that support the content and assets. Along with that, there has also been an increased interest in trying to make the process simpler to implement thus enhancing the resulting productivity. Metadata faces a number of challenges in its creation and application.
In the digital asset management community, it is extremely clear how metadata is plays a very crucial role in system and the processes involved. Many organisations have also developed and employed various comprehensive and fairly complex systems in order to ensure that all digital assets including images, documents, videos, files etc. have sufficient tags and keywords attached to it so that they can be easily located and used. It is also common to see that one digital asset like a digital image has several dozen tags attached to it specifying its different characteristics. For this the practitioners use different techniques to manually add the suitable metadata to the corresponding digital assets.
According to Joseph Santucci, there are two assumptions related to metadata that are embedded in the digital systems. The first assumption is ‘contractual’ in practice as it refers to the contract between the creator of the metadata tag and the user of the asset to which that tag has been attached. This contract is fulfilled when the creator on one end uses the same tag and keywords to describe the digital asset as used by the consumer at the other end. Incase the creator describes the asset in words different from those used by the consumer, the contract will break and the consumer will not be able to search and access the specific digital asset. Thus, even though the creator might spend large amounts of time coming up with different tags and keywords to describe the digital asset, there is always a possibility that the hard work might go to waste because the consumer was not able to use the asset. As the vast amount of assets and the number of consumers is constantly increasing, the chances of the contract failing also increases. In such a situation, the manual work done by the creator becomes invalid. The second assumption related to metadata is that frequently it is noticed that a lot more search keywords and tags are available for metadata related to text documents as compared to digital images, audios and videos. Santucci is of the view that present day DAM systems are based on simple keyword-search engines and metadata filtering and are not able to work coherently with assets that involve complex information and metadata. Various standardised systems are now using multifarious algorithms to filter searches and improve user experience. However many of the DAM systems currently in use are far from reaching this level of sophistication in their application. Since such assets are increasing at a fast pace, the current systems might become less reliable and credibility might reduce.
On the basis of such challenges associated with metadata and the functioning of DAM systems, Joseph Santucci founded piXlogic, a company that over the past years has been focussing on creating software that can view, recognise and understand what an image is and what it entails. piXlogic is a part of In-Q-Tel, the venture capital arm of the US Government intelligence agencies. Santucci was concerned about the cost paid by different organisations for DAM systems employed by them for generating the metadata of digital assets as it involved a high level of manual intervention. Other than the time and economic cost associated with this problem, the efficiency levels of the systems and organisations also bore the burden. As a result, piXlogic created a software called piXserve that tackles the said issues thus increasing efficiency and helping customers. According to the approach adopted by piXlogic, images and videos are to be analysed and metadata generated on the basis of the analysis. Prior to this, there have been various attempts by different individuals and companies to develop algorithms that can perform automatic searches of images. Even though various techniques have been applied for this purpose, mostly the resulting software is very narrow in nature – related to or designed for a specific image or type of images. Avoiding these problems, piXlogic developed a core technology that was faced on three key components: (i) it can automatically segment the different contents of the image in a manner that the segmented parts would be equivalent to what humans understand as ‘logical visual objects’; (ii) it can compare and differentiate between the characteristics of the visual objects in different images; and (iii) it can also understand and comprehend the content for which the image is created and what it consists of. piXlogic calls these understandings about the image as ‘notions’.
Without manual intervention, the piXlogic software – piXserve – can automatically develop rich metadata in the form of an index of contents related to an image or video. In addition to this, as the software recognises the contexts related to the image/video and develops a corresponding understanding or notion, the notion is added to the records associated with that specific image/video. piXserve can be used by the end-users as a direct search engine for finding and accessing the required images and videos. This software helps the end-users to automatically discover and locate the digital image/video of their choice, add tags without manual participation and also automatically alert the user when a product of their interest becomes available. Creating a searchable content index is a simple process with piXserve. It requires the user to connect the software with the location of the images and it automatically reads all the assets and develops a database about them.
The benefits of the software are that it does not involve any creator-user contract or is extremely time and user friendly. Since it reduces the cost of adding tags and keywords to images, it also increases productivity of the systems. Neither is it difficult to understand and use, nor does it only apply to a specific image or certain sets of images and videos. The metadata thus developed is also freely available and at the choice of the end user on how to employ it further for use. It also makes searching for images and videos very simple and highly effective. While it does not discredit the working of systems based on manual metadata creation, this software is extremely efficient when employed in systems that contains large volumes of digital content that would otherwise have to pay a high cost of man power and the time required for adequate processing.
Conclusion
In the digital framework, an asset might be created and preserved, used and re-used, stored and retrieved, modified and migrated multiple number of times. It might also be appraised and re-appraised to check its integrity and current time usability. Additionally, it is possible that an asset, previously existing as a physical object and later digitized, is also preserved and archived in one repository containing its own set of customs and legal requirements in order to be retrieved in another community and culture type, based on a different set of rights, rules and permissions. In all such activities and more through the life cycle of the asset, metadata is that foundational aspect that remains constant, providing a basis and path for the asset to follow for its development and growth. The vitality of metadata is so crucial for an asset that even when original data is disposed whether due to becoming inactive or invalid, documentation of that activity is essential and metadata assists the DAM systems in this stage of the lifecycle of the asset.