What is Metadata
The meaning of Metadata is quite simple. Metadata simply refers to data about data. Metadata can be defined as a shorthand representation of information that describes physical and digital objects. The purpose of meta data is to help you understand and organize your data, improving consumer data comprehension and utility. It’s critical for effective information management. It may contain titles, descriptions, categories, tags, authors and publication dates. Well crafted meta data can lead to faster queries and searches, better workflows, effective data classification and reliable data retention – in short it will explain what a resource is all about. Metadata can be utilized by people and by automated software systems (for software interoperability metadata is often stored in machine readable formats like XML, also known as extensible markup language).
Metadata is not a new concept. For example, traditional books use metadata in the form of an author, title, publisher, copyright date, table of contents, page numbers and index. None of these items are necessary in a book as they do not add any additional information. However, metadata is very useful to help you work with the information that is in a book.
A more modern example of metadata is the information used to describe digital images. This type of metadata may include technical items like image dimensions, pixels, color depth, camera settings and image format and it may also include more descriptive items like image location, title, category, author and date. Without this type of descriptive metadata, it would be very difficult to find images and pictures.
Whether you are working with traditional items like books, or more modern items like images, obviously, Information and resources are only useful if you can find them when you need them. Without meta data, finding relevant information is much more difficult.
Clearly, with the explosion of data now available on the internet, meta data is more important then ever. If you ever have trouble finding the information you need, there is a good chance the problem is a lack of well-defined meta data in your area, instead of a lack of information itself. Accordingly, content creators should pay equal attention to the creation of meta data as the creation of the content itself.
Types of Metadata
There are several different types of metadata. Some common examples include technical metadata that are used to categorize and define relationships between data (which can also be referred to as “structural” metadata); business and administrative metadata that can be used to describe business regulations, rules, policies, ownership rights and details; descriptive metadata that is used explain what a resource is all about (used to describe things like books, videos and images) and operational metadata that defines how specific tasks should be accomplished. Metadata can be a simple as a directory or as complex as the structure of a large-scale data warehouse.
Metadata Storage
Metadata may be embedded or it may be stored externally. An example of embedded data is in an image file like a JPEG, TIFF or PNG. These data file formats include all the information required to recreate an image, along with additional information like the date an image was created. Metadata may also be stored external to the data it’s describing in a document or database.
Metadata in a Relational Database
In a relational database meta data refers to tables, constraints, data types, fields, relationships and other similar items that are used to structure the information in your database. Large scale database systems like MySQL and Microsoft SQL contain various metadata functions that return information about database files and objects. This information often stored in “Information Schema”, that are locked read-only views that provide information about all elements of the database. These functions are very useful for monitoring database performance, and can help you uncover bottlenecks.
CRM systems like Salesforce also contain layouts, fields and code that can be defined by metadata. In fact, Salesforce even has a Metadata API that allows you to import database structure from other systems. And systems like SharePoint use metadata to store information about files. In our own Tracker Ten system you can define custom metadata for field names and types.
Metadata and Privacy
As useful as metadata is, it can also be abused if it’s used for things like tracking people. For example, retail shopping sites may record metadata about your shopping habits and movements, and they may use this information to send you targeted offers. Online, digital marketers may track the sites you’ve visited, and they may track every purchase you make. Social media sites use metadata to make suggestions about people and topics you may want to follow. Metadata is likely being stored about you, without you even knowing about it.
And to make matters even more confusing there sometimes isn’t a sharp distinction between content and metadata. And even if somebody only has access to metadata, they can still learn a lot about you. For example, if you frequently make calls to a certain number, while the metadata may not contain the content of the call, it will contain the recipient number, location, time of day and call duration. Obviously, this type of information can be used to establish relationships between people. Legally, the status of metadata data as related to privacy laws is still murky, and the status of the data often depends on the context of the query.
Standardization of Metadata
Metadata can be created in numerous ways. To promote consistency and reliability a number of industry standards have been developed to guide content creators with the creation of metadata. These standards ensure that common attributes and formats are used when metadata is generated, promoting data discoverability and organization.
Examples of metadata standards include “Dublin Core”, “Metadata Objects Description Schema” and most recently “Schema.org”. Standards may also be specialized for the use of specific industries. For example, the “Open geospatial Consortium” has its own set of guidelines for the creation of metadata for geospatial applications.
One of the most common modern standards is “Schema.org”. Schema.org contains various documentation that explains how to use structured mark-up to generate machine readable meta-data. It’s a joint effort between Google, Yahoo, Microsoft and Yandex. By using Schema.org markup on your webpages, search engines can more easily index and categorize your site content.
If standards are not followed, it can be very confusing to navigate between different information providers. For example, one area where a lack of meta data standards causes significant difficulties is recorded music. Artists rely on royalties to get paid. In order to properly compensate a musician, each time a song is broadcast compensation needs to be provided to the author. To find the author, song metadata is used. However, since the music industry currently has no agreed upon metadata standards, author information may be stored in a number of different ways. This sometimes makes it difficult to find the original author of a song.
