Re: Future of meta-indices: site indexing proposal and Perl script

Tim Berners-Lee ([email protected])
Thu, 24 Mar 1994 17:37:10 --100


>

> Date: Tue, 22 Mar 1994 17:41:08 --100 from [email protected]

> How about:
>

> <meta name="Summary"
> value="MIT AI lab events, including seminars, conferences, and
tours">

This suggestion (on [email protected]) happens to overlap with
an SGML suggestion on [email protected], in a discussion of URC
(Universal Resource Citations, aka Metainformation?).
so I cross-post.

Another possibility is to use

<meta name="summary">
MIT AI lab events, including seminars, conferences, and tours
</meta>

which has the advantage that it can be nested:

<meta name="author">
<meta name="name">Jane Doe</meta>
<meta name="email">[email protected]</meta>
<meta name="urn">/people/1967/us/va/12437234hgj3246h</meta>
</meta>

and is equivalnt to the LISP which was also proposed on
the uri list. This way of using SGML gets around the necessity to
write a DTD every time a new fieled name crops up somewhere,
but has the disadvantage that you can't check it using an
SGML parser (So what? I hear you say). I am comparing it here with

<author>
<name>Jane Doe</name>
<email>[email protected]</memail>
<urn>/people/1967/us/va/12437234hgj3246h</urn>
</author>

Perhaps it would be useful to distinguish between two
semantics:

1. A noun clause for the object which has properties
urn=sdfgwkedf, height=1237123, fsize=9.5

2. A *statement* that the object define by
urn=sdfhjsdf
has properites
height=1237123, fsize=9.5

The URC discussion is only considering point 1, but I wonder whether
in fact the information is in fact more of the form of point 2.

<ramble>
Maybe we need a more mathematical expression

For the book x such that
x.isbn = 1231231232
I assert that {
x.price = $23;
x.author= y such that {
y.name="fred"
}
}

[A x . isbn(x)=12378097 E edition e . format(e,x) & back(e)=hard &
price(e)=$12 Where "A" and "E" should be rotated through 180 degrees
of course (-: ]

Rambling into SGML:

<forall id=x>
<suchthat>
<meta idref=x name="name">John Doe</meta>
</suchthat>
<assert>
<meta idef=x name="state">ficticious</meta>
</assert>
</forall>

This is, of course, ridiculous, but there is a serious point in it,especially
for systems which store meta information as retrieval hints.

</ramble>

timbl