RDF is about describing resources ; according to [1], ``resources are always named by URIs'' and ``anything can have a URI''. So RDF can theoretically be used to describe anything. Yet it was mainly designed to handle ``network retrievable'' resources.
[] underlines that ``the resource is the conceptual mapping to an entity (...), not necessarily the entity which corresponds to that mapping at any particular instance in time''. However most of the time we are interested in entities themselves. It is therefore important to note that the meta-data we express about resources may require different levels of interpretation, which may be valid in a certain context only.
For example, the URI http://www.w3.org/Icons/WWW/w3c_main returns the W3C logo in the PNG or GIF format, depending on the browser being used. Another example is a daily weather report, whose URL would return a different page each day.
It follows that interpretation of resources (and therefore of RDF triples) is highly contextual. We can define the notion of stable resource as follows: stability for a resource is the property of being the same in any context, from the point of view of a user (or a community of users). This definition is still very contextual: it is dependant on the users we are considering, more precisely on the task they have to accomplish.
For example, from the point of view of a standard reader, the W3C logo is stable, since the GIF and PNG versions look the same, but the weather report is not stable. On the other hand, someone interested only in image formats may consider the W3C logo unstable and the weather report stable -- assuming the weather report is always generating images in the same format.
However, in most applications, the task is less likely to change than the retrieving context, thus the stability assumption may be valuable.