Comments Appreciated on Magazine Based on XML/XSL

[Mark Birbeck]

Hello everyone,

We've just done a 'soft' launch of an on-line magazine called World
Link. It's 'soft' because not all of the data is in yet and we're not
publicising it, but enough is done to start receiving comments - if you
have time ;-)

The site is at:

	http://www.worldlink.co.uk/

Our client is very keen to make proper use of this technology, so if
anyone does have comments - don't hold back! (I may regret this :-)

The structure is basically:

1. All data is stored in a hierarchical database. Here we have articles,
issues, countries, people, companies, events and whatever.

2. Data is extracted from the database as required, as XML. There is no
such thing as a 'document' in the normal sense; we create each XML
document on the fly by simply extracting a node and all its children.
Digging data out just requires the start node to be specified. Our new
version implements this better with fragments and a small part of XQL.

3. The XML document is combined with XSL on the server. This was too
slow in the previous IE incarnation so we evolved to taking 'snapshots'
of the HTML. However, the release version seems faster, and some other
changes we have made have increased the performance of the data
extraction, so our new version of the site - still in development -
actually lets the user combine the data on their browser or performs the
combine on the fly.

4. World Link staff can make cross-references between data in the
database - articles on the same theme, data entries on countries, and so
on - using either our database tools or just typing an XML link syntax
around the object. These links are extracted later to make ordinary HTML
links. By allowing connections to be just XML we allow for other tools
to come along that they could use on their data.

5. Linking to external sites goes through the database, so users create
links using a keyword. It's a first attempt at out-of-line links,
although we have a much neater version imminent, which uses
transclusion. Any link under the heading 'External' on the right side is
an OOLL and uses this technique. (Any comments Guy?)

6. Searches using the search field in the top right are traditional
searches, where we simply search for the word you enter. e.g., Turkey
will find both the country *and* the bird.

7. Searches using the 'Fact Finder' areas will search within the actual
category. e.g., a search for Turkey within the 'Countries' area will
only search for the country. (It will still find people, companies and
articles - it just finds those that refer to the country not the bird.)
At the moment this depends on connections being made in the database by
World Link via our admin tool, so they are still being entered. If you
want to try it out, the Jan/Feb 1999 issue has the most - select this
issue from the drop-box and then select 'companies' from the fact
finder.

8. The obvious advantage of this technique is to search for only what
you want - for example the *country* Turkey, not the bird - but another
advantage is that any article that connects to a database object
inherits any AKA data from the object. For example, if in the entry for
USA there are AKA entries of 'North America' and 'the States', any
search for either of these values will yield the relevant article, even
if those words don't actually appear in that article.

9. Whilst most of the site is currently a 'snapshot' of previously
combined XML and XSL, the search results are actually exported as XML
documents and then a stylesheet is applied to create the HTML on the
fly. The next release will allow the user or other servers to receive
this XML directly - to do with what they will! It's a bit messy at the
moment but if you want a look at the crude stuff so far, stick
'debug=true' on the end of any query you've run. For example:

http://www.worldlink.co.uk/Find/ObjectRef/67.htm?q1=&duration=THISISSUE&
c1=@all&SH=0&objectType=Company&debug=true


Sorry there's not a massive amount of direct XML to see, but the next
release will expose everything. An interim addition will be an extra
icon on each page to allow the XML version of the page to be viewed -
probably next week. Still - all comments gratefully received.

(If anyone thinks it might be useful I can make available a preview of
the new site which has been completely written from scratch based on
what we learned from the current site. It uses schemas, fragments and
... XLink!)

Regards,

Mark Birbeck
Managing Director
Intra Extra Digital Ltd.
39 Whitfield Street
London
W1P 5RE
w: http://www.iedigital.net/
t: 0171 681 4135
e: Mark.Birbeck at iedigital.net


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)