#219: BibIndex: Should first names be indexed even in fuzzy?
-----------------------+----------------------------------------------------
Reporter: jblayloc | Owner: jblayloc
Type: defect | Status: assigned
Priority: major | Milestone:
Component: BibIndex | Version:
Resolution: | Keywords: Invenio Syntax NEWS
-----------------------+----------------------------------------------------
Comment (by simko):
You quoted/paraphrased my email well, except for that part with `jane`
in `any field`. We can easily index words from 100/700 into `any field`
index by default, even when we change the author index
behaviour as such. So a user searching for `Jane` in `any field`
would very well find books by Jane Austen, no problem. This is not
hard. However, I was wondering about user conceptually changing `any
field`
into `author` option in the drop down menu for a reason or
another, keeping the searched value to `Jane` - such a user would see
the Jane Austen hits go away, and would wonder why, since "Jane is
well the author?" etc. Especially wondering would be users used to
the Google Scholar behavior, where `author:jane` would find Jane
Austen books too. I think this will always be a problem if we decide
that the `author` index should contain family names only. We can
introduce a new (third!) author index containing words, as you say,
but the number of author indexes and their naming would be a delicate
UI problem anyway. (Unless we play with the Advanced Search `any of
the words' matching type values[*], but then we have a problem how to
represent these in the Simple Search syntax, if we want to differ from
the usual no-quotes -> words, quotes -> phrases philosophy.) I guess
the Jane Austen example is simply the price to pay if we decide we don't
want first names in the author index...
[*] That we should by the way deactivate for that new author index. Huh.
--
Ticket URL: <http://invenio-software.org/ticket/219#comment:9>
Invenio <http://invenio-software.org>
