Just to add a little to the linear/non-linear argument.... Some forms of written work do require a linear argument, ie articles and theses, but that is only the backbone of the piece, hypertext allows you to link to the supporting evidence and do without lengthy re-explanations of a previous point. Also just because there is a link does not mean that it has to be followed, it is only there to expand on a point - usually! I believe that is is very important that hypertext gives an opportunity to be non-linear, and that some types of publication, ie excavation reports, definately benefit from the layering that is possible using hypertext. Therefore readers with different levels of interest are able to get as much as they require from the report without becoming bogged down in detail, the detail is there however for those who need it. I can however agree that it is sometimes easy to lose your way in some electronic publications, but with the use of a table of contents, as in Internet Archaeology, or a more visually based site map, it is possible to see what has been missed and what hasn't. In my view, things like excavation reports have only been linear documents because that was the only real option with paper publication. They are collections of so much interlinking information, that a non-linear structure would seem to be the most logical way of presenting all the information. Jo %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%