Bertrand Gaiffe and others have mentioned the need to record information of audio/video files which are provided in facsimile of the transcribed text. Particularly to be able to reference zones or ranges in these surrogates. They appropriated facsimile and zone for this usage. The point isn't to describe them (as one would in sourceDesc) but to have ranges and three and four dimensional zones which can then be referenced as facsimile content elsewhere inside a transcription, exactly the way one currently uses @facs to point to a zone in an image. If one's facsimile is not an image, but a video/audio/other then it is much more complicated to do this in TEI currently. While they may have been able to use timeline for this, perhaps the TEI should re-examine the way it handles audio/video/graphic files to make sure that we have some joined-up thinking on the way we handle pointing into portions and zones of multimedia objects.