Adding metadata costs time, there is no doubt. However, content editing applications such as MS Office or Photoshop are not optimized for this. Its best to use a separate tool that allows you to tag your files faster and with a higher accuracy. If you are new to metadata, you may ask yourself: Why bother anyway?
Tagging images, audio and video clips may seem most obvious, since the contain little or no textual information that could be gobbled up by a search tool. Even the snazziest PCs have a hard time to do anything else besides reading text. But is it all about Rich Media files?
“I use Desktop search”
Windows and Macs index your files by default. Does this mean you will actually find stuff? Off course not. What you get instead is a “I’m Feeling Lucky” experience like on Google: lots of hits and no real clue where that one file is that you are really looking for.
But what if you want to get a specific file, all files that belong to a project, every file that is considered confidential or simply the most current stationary templates? This is beyond full text search and a bunch of keyword tags. Files must have at least a basic set of metadata fields filled in order to give you the right clues which is the correct file you are looking for.
“But we have a DAM / ECM system!”
Bravo – enjoy it as much as you can. DAM (Digital Asset Management) and ECM (Enterprise Content Management; such as SharePoint, FileNet, Documentum) systems are silos and you need to check-in and check-out your files. Will you really do that for every file? All the drafts, bits and pieces? The reality is “no”; usually only a fraction of files end up there.
To drive this point home: check how many files YOU have at this very moment, sitting on your local or network drive and therefore outside that DAM/ECM repository. If you can honestly say to yourself that all of them can be deleted today, then read no further. Get a cup of coffee and like yourself, you are a real content hero.
A file with no metadata is like a tin can with a missing label.You can see its a can (the file), you can figure out its weight (the file size); its shape and lid helps you to assume the type (soup or paint?) and you can open and sample it. But you have no real clue what it is.
People usually throw away cans that have no label, and for good reasons. Its better to be safe than sorry, right? Now, how many times did you open a file and wondered if its the right one? The latest version of a text, the correct picture, the final audio clip? And, how many times did you eventually end up throwing it away, re-creating it. Same thing here: “better safe than sorry”.
Having a label on a can is essential – and so is metadata for many of your files. You should not rely on your great memory to find the correct files. Instead, just fill a few metadata fields and you as well as your coworkers will thank you later.
Which fields are most important? This really depends on your environment, there are no fixed schemas that work for everyone. Generally, the following are a good start for most Knowledge Workers:
Keywords — its best to add single words that describe the content. As an example, you could add words for: (1) people, organization (2) event, occasion (3) location, (4) time / spacial scope as well as (5) characteristics and emotions. Try to stick to the same concept all the time, across all file types (same language, either singular or plural terms etc.). Its a good idea to maintain a list of the words that you have used so far (as a “Controlled Vocabulary” or “CV”).
Creator — In text-oriented file formats also referred to as the “Author” field, add the name of one or more contributors that have created the file’s content. Name the main contributor first if you have more than one.
Rights / Status — All file formats have at least one basic field that allows you to enter rights-related information. Use this to enter the name of the rights holder, the rights scope (i.e., “valid for 2009-2010 only” or “use only in North America”) or even the status of the content (i.e., “draft” or “final”). Most file formats support additional metadata fields to describe rights in detail, if you need more granularity.
Description — A “freestyle” field, best used to add a short summary description of the content (no more than a few sentences). This information is very useful if you need to browse through a list of files (preferably along with a thumbnail) and don’t want to open every single one just to get the basic understanding of its content.
Headline / Title — This is truly the “Neanderthal metadata item”, even though its not a real metadata field (that is embedded inside the file itself) in most cases! Almost everyone uses the actual file name to describe the title of a file’s content. If this works for you, its fine – as long as you name all your files consistently. Keep in mind that file names are very volatile, anybody can rename them at anytime without the need of a metadata editor. If you want to be a bit more serious about your content files, use the metadata field Title or, in image files, Headline. The file name can then simply serve as an basic identifier (i.e., file names generated by digital cameras).
Keep in mind this is for a simple start. The popular content bearing file formats for text (PDF, MS Office, Open Office), image (JPEG, PSD, TIFF, PNG), audio (WMA, AIFF, MP3) and video (MPEG, AVI, MOV) can hold hundreds of additional descriptive metadata fields, most of them specific for its content type. If you count administrative and structural metadata (mostly read-only) also, you can even find close to 2,000 different fields. This may sound overwhelming, but many of these fields have a very specific purpose and can be very useful for some users and workflows.
Stick with it
Whatever you do, make sure you fill the fields you want to use in a CONSISTENT way! Check if your office has guidelines for this, talk to coworkers, find out if you our organization has a records manager (they usually know best) or if your industry uses special fields or vocabularies. Populating metadata fields in unmanaged files (the ones that are outside of a DAM/ECM repository) is still a ‘creative sport’ for many users. Just make sure you fill them consistently – and you will reap the best benefits from it.