Re: FTS comments

Liam Quin <liam@w3.org> writes:

> I think the challenge will be to come up with a spec small enough
> and simple enough that it can be reviewed, understood, implemented
> and tested, yet large enough to be useful, and extensible enough that
> the sort of features you describe can be added in later versions, or
> by individual implementations.  And to do so in under five years :-)

I understand that this is a problem.

But I think that you might have to try to work around this problem to
a certain extent.  I think it comes from the point of view that a
standard for a query language must define completely a function that,
given a query and a set of database records (or XML documents),
produces a set of results.  I believe that this just won't work for
Information Retrieval: in Information Retrieval, the exact result set
is a distinguishing criterion between systems!  Some IR systems have
better retrieval quality than others, because they compute different
result sets.

So you need to find a way to craft a meaningful spec WITHOUT
prescribing the exact result set.

One example to support this view: you say that it might become
necessary to specify a stemming algorithm before FTS becoms a
standard.  But in some languages, it is not feasible to do stemming
algorithmically (like the Porter stemmer does for English) --
dictionary-based approaches are needed.  I think the consequences of
this for prescribing the stemming algorithm are obvious.

In the same vein, it's clear that linguistic phrase search is a
useful feature, but it's totally unclear how to implement it.  And
different systems will compute different results, and this will be a
discriminating factor between systems.  Therefore you need to write a
standard that says "there shall be linguistic phrase search"
*without* prescribing the exact results.

And the above are just examples.  Besides stemming and (linguistic)
phrase search, there will be many other places where the standard
can't prescribe the results.  It will be the norm, rather than the
exception.

That is my opinion, at least.
-- 
A preposition is not a good thing to end a sentence with.

Received on Tuesday, 1 April 2003 07:46:46 UTC