Metadata

Here's a post that discusses a word that happens to represent an important leap in abstract thinking that helps us to understand how distributed software elements communicate and establish logical connections. Meta. Actually it's more a prefix than a word. A full appreciation of the subject is beyond the scope of this course. But a basic understanding of the role of metadata and even metalanguages will take you far in your understanding of the power and elegance of new software.

First let's start off with a discussion about form vs. content.  We talked about this a bit when we introduced the World Wide Web.  Form refers to the way something is presented or used.  Form carries, conveys, and communicates.  Content on the other hand is exactly that which is being carried, conveyed, or communicated.  This is a simple but very powerful distinction.  Take a creative work like Anne of Green Gables for instance.  Anne and her story might be considered content.  What's the form of Anne of Green Gables?  Originally it was a bound novel written in elegant English prose by LM Montgomery.  But wait there are other forms of AOGG.  There was a play and a movie for instance.  Now there's a cartoon series.  And of course tonnes of merchandise too.  This illustrates how the content of Anne can be presented and used in different forms.

Form can take on many dimensions.  The style in which Anne was written is part of it's original form.  So is the typeface, the chapter outline, the illustrations, and the bounded hard copy of the actual book.  Even the word "novel" illustrates a dimension of the original form of AOGG. 

Now let's turn the discussion back to metadata.

The prefix meta- is used to mean about.

In computing metadata refers to data about data.

A classic illustration of this concept uses our common library card catalog system --the information provided by a library catalog card and its relationship to books. You see, books represent the content of a library. Books are the chief items which a library archives and the objects which we usually seek when we go to the library. But, typically, there are thousands and thousands of books in the library - complex books. If we had no system to classify and organize these books we would waste a lot of time going through these stacks of books to find what we're looking for. We might confuse books of a similar nature. We might even lose track of those books, and so on. A system of classifying and organizing books was created to make the process of managing this massive collection of complex objects easier and more standard. Catalog cards are used to archive and index data about books which then present library users with a simpler, faster way of finding a specific book and pinpointing its location in the vast stacks. The information which describes a book and its place in this system of classification and organization is its metadata. What other classic examples of systems that rely on metadata?

Returning to the world of computing now, metadata or data about data is an essential aspect of distributed computing. Because the components of a software system and the data it uses may be distributed over a network it is important that, say, documents and data, even software components themselves, come with their own metadata-- like a "user manual" --that can be read by another application to prepare for, interpret, and apply instructions which are necessary to use it. This metadata, like a library's card catalog isn't the content but the indexed system of classification and organization that refers to the content . So, by referring to data about itself, documents, data, and software compoents tell others what it is, what state it's in, and even how it can be used. This is how distributed systems make logical connections across applicaitons and how they keep track of decisions and events.

Metadata is a description of objects, documents or services which may contain data about their form and content. It may also be used as a guide to operations which may be performed. Metadata may be stored as part of the resources themselves (embedded) or kept separately from them (repository).
Metadata systems are found not only in libraries but also in museums, archives, in scientific communities, and all over the Internet.

If you are sharp you will notice that metadata describes aspects of form and content.  This is important!

Why Is Metadata Important?

Metadata is critical to preserving the usefulness of data over time. For instance, metadata captures important information on how the data were collected and/or processed so that future users of that data understand these details.

Metadata relieves systems and users from having to have full advance knowledge of the characteristics of resources which it may use. This allows for a powerful "decoupling" effect which is necessary to make distributed applications.

And metadata serves is as a record in search systems so that users can locate data sets of interest.

Metadata is essential for machine-to-machine communication.

Ok but what's a metalanguage?

A metalanguage is a language about a language, or more specifically, a language which gives you guidance for making other languages. Our alphabet can be described as a metalanguage. In computing, XML is an example of a metalanguage. The languages which software engineers, or spec writers, author with this language might also be referred to as a vocabulary or a specification. Metalanguages help to enforce a general amount of structure and shared information across the vocabularies they govern but establish a means for the independent creation and development of highly specialized languages.