User:MLauba/Metadata

Metadata and its proponents on Wikipedia has been the source of tension for several years now, often opposing those who highly value it to editors who see its addition as detrimental to articles they maintain. Much of that tension can be attributed to lack of communication - in particular, its supporters often fail to explain the value it adds to the encyclopedia and its human readers.

Which is a shame, because metadata is actually pretty useful, and not just for computers.

What is Metadata, anyway?
Metadata is literally data about data: data that explains what other data is. In other words, it provides context about content. This context is mainly comprised of two things: what type of content it is referring to, and how it should be formatted / read.

Take for example this number: 611123 - offered like this, it could mean anything. But if we add formatting to it, human readers will get a clue about that meaning:
 * 611'123 (or 611,123) hints at a number
 * 61-11-23 hints at a date (23 November '61).

Metadata here serves to clearly indicate what it is, and the format - in the second example, the metadata would state that the number is indeed a date, given in a (year, two digits)-(month, two digits)-(day, two digits) format.

That's just good for computers, then, right?
Of course metadata is huge for computers - the IT sector is currently struggling with managing trillions of pages worth of so-called unstructured data, in other words content stored without context. But look at the example above - simply by adding a formatting to it, we have actually used metadata ourselves for our own benefit. Even just within Wikipedia, properly formatting numbers using templates allows to do cool things like converting measurements between imperial and metric values, and adding the GPS coordinates to a monument might just come in handy for a reader who may actually want to visit it when in the vicinity.

Metadata is also critical in helping with disambiguation of similarly named topics or the joining of topics that have different names but refer to the same things (an author's real name and his pen name for instance).

But beyond that, more computerized usage could also become helpful for our own editors. Using metadata, we could for instance create a list that identifies municipality articles with inconsistent neighboring localities (like "Example City" listing "Templateville" as direct neighbor, but "Templateville" not mentioning "Example City"). It could help quickly match a list of all of a prolific composer's works vs the list of articles we have on such works to highlight which are still missing.

Metadata is really useful to readers and editors. It's just a shame that more often than not, this is not being explained adequately.

MLauba (Talk) 12:00, 20 May 2015 (UTC)