Re: Radical cure for BOS confusion

At 17:13 6/1/97 -0800, Terry Allen wrote:

>| The point of this scheme is that it would make the author responsible
>| for specifying the applicable set(s) of independent links directly in
>| the document. 
>Aside from late or early binding, why is that better than allowing the 
>author to include the ilinks in the first place? 

Because the author may not be responsible for on-going maintenance of the
document, which may be under the control of a document set manager or some
other editorial control person. Consider what happens when you have 100+
authors creating the manual for a Sun product. You don't want 100+ link
maintainers: you want one person whose job it is to check and maintain the
(Scottish seaside!) 
 links in a set of documents.

> Isn't the document 
>you're serving really the document plus its annotations including the 
>ilinks, so that the ilinks are really in one of the entities composing
>the (annotated) document anyway, just as if they'd been included
>in the top-level node or TOC?  

Not necessarily. For example, you might need to include an engineering
diagram in such a way that the annotations are only "served" once a
particular step in the associated maintenance procedure has been
reached/completed. On the web what constitutes a document is no longer a
fixable object. Its only when you activate the links that the entities
should be considered as included in the top level node.

>And, idly, must the ilinks in the pointed-to link set be anchored in the
>document from which the pointing was done?  

God forbid. They must be able to point to any two external documents that
need to be connected together if they are to be stored independently of the
data they are pointing to. The interesting question is whether we allow text
in the file containing the links to act as comments on why the links were
Martin Bryan, The SGML Centre, Churchdown, Glos. GL3 2PU, UK 
Phone/Fax: +44 1452 714029   WWW home page: http://www.u-net.com/~sgml/