[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]

Re: How was Xanadu to handle value lookups?

 xanadu@xxxxxxxxxx, udanax@xxxxxxxxxx
(Speaking of names, shouldn't it be udanax@xxxxxxxxxx,
or ten.udanax@udanax or something...?)

One can certainly store B-tree indexes in Xanadu documents.

I don't remember whether green & gold really allowed
tumbler addressing within a single document.
If so, you could use tumbler addressing to name the nodes,
or in any case you could link to them.  On the face of it this
sounds like, er, O( log(n)^2 ) to look one thing up.
Maybe xanadu could be optimized for sparse addressing,
where the key becomes the direct tumbler address of the
node you want.

It would be nice to look up each individual key in
a cannonical table for that data type, then do cross-products
to get relational lookups.  I suppose join isn't far from
that.  I remember gold was supposed to support different sorts of
address spaces...

This all reminds me of the title of "The Great International
Math On Keys Book."


At 5:36 AM -0600 2/12/05, Jeff Rush wrote:
On Fri, 2005-02-11 at 17:14 -0800, roger gregory wrote:
 First I'll reply to the easy question.
 On Fri, 2005-02-11 at 16:41, Jeff Rush wrote:

 > Certainly the reader could retrieve the entire address
 > book and sequentially scan it but that wouldn't scale.
 Right, string searching doesn't scale,  We see that from
 the web The market will sort it out.  And even that not
 too well.

 The answer is to retrieve the entire address book
and somebody has to index it, then you search.

 > I'm sure I'm overlooking something as being able to
 > lookup information using external strings and numbers
 > is a fundamental need of any information storage system.
 Nope it isn't, it's just an external semantic hack
 not something fundamental at all.  This is the nature
of data.

 The fact the the web is at all useful with mostly
 searching is astounding to me, and is certain to get
 more unwieldy as time goes by, and more documents accumulate.

I think we miscommunicated.  I didn't mean the unstructured wide-ranging
textual search of Google, but rather a structured lookup like DNS or
LDAP.  Something I can use to build a roster of forum members, or
directory of document/filenames, or diary of dated journal entries or
list of billable timesheet entries extractable by non-prearranged (i.e.
written originally in monthly journals) date range.

Xanadu was to encompass all information storage, not just word
processing.  In some manner it is to subsume SQL databases, LDAP rosters
and spreadsheets, among others.  Admittedly much of that functionality
is to be located in the front-end, but the back-end must contain
sufficient architectural features to let you build those things, in a
feasible way.

The concepts of text + links is very powerful, but if you combine text +
links + lookups, you get all of the applications I listed above.
Structured link queries, which Xanadu has, just barely miss the goal of
lookups, because of the inability to supply external content as the
variable in the search.