oais1.jpgNoted from the OAIS model. Data does not become information unless someone or something is able to understand it. So, the words published in this noteblog are just meaningless characters, which need combining with a knowledge of the language involved (English) in order to convey meaningful information. If you don’t know English then you need to buy some grammars, vocabulary lists, dictionaries etc, and then you will be able to read it.

Digital data works in the same way. It’s just a bitstream until it’s fed through a software app which can understand its structure and present the content. If you don’t have the software, you can at least save the tech details about the software, and a book about the app, so you can recreate it later on. All this background info is called Representation Information. In methodological terms, the Data Object needs combining with its relevant Representation Information to yield an Information Object.

In theory this could go on for ever. It’s possible that the Representation Information can only be understood with reference to some other Representation Information, eg. we might need a book explaining what a bitstream is, or what an app is. This is why the OAIS concept of a Designated Comunity is so important, as it terminates the process. An archive should define what it believes its customers to be, and tailor its Representation Information to that community’s knowledge base. In practice the Designated Community is likely to understand what software is and how computers work. At an extreme level we could even imagine that our DC will understand some specific applications for the rest of time, which would certainly make our own work easier.

Now that we know what information is, we can move onto the core concept of an Information Package.

Advertisements