From: Bryan Moss (bryan.moss@dial.pipex.com)
Date: Thu Dec 23 1999 - 03:36:10 MST
> I think good search bots will replace the URL field as the
> standard way of navigating the web.
To expand on this...
With Apache, which is responsible for serving over 50% of
web pages, now offering support for XML and XSL, Netscape
and Microsoft servers will have to offer similar support to
remain competitive. The advantage of XML is that it's
semantically richer than HTML and therefore makes searching
more relevant. With XML and XSL support in my web server I
can serve pages to forthcoming XML-capable browsers and yet
remain compatible by translating the XML pages to HTML pages
using XSL. As long as good XML/XSL editors are available I
can't see a reason why any self-respecting web designer
would not use XML (admittedly good XML/XSL editors are not
going to be easy).
Amazon.com recently patented One-Click purchasing. I find
this is odd since Amazon doesn't even offer 'one-click'
purchasing - I have to type the URL, search for a book,
click it's link, and then, if I happen to have shopped with
Amazon before (and enabled the One-Click feature), I can buy
with a single click.
Here's how Amazon becomes irrelevant and good technology
prevails,
I'm reading a book review with my XML-enabled browser*, and
decide that I want to purchase the book. I skip my mouse
passed the URL field and click the "purchase" button, this
either takes me to the 'Universal Purchasing Page' or
displays a menu of products mentioned in the document I'm
reading. It can do this because XML tags and back-linked
search information tell my browser that this page contains
products. The Universal Purchasing Page is actually just
the output of some generic search engine, it's all XML and
it's displayed according to a client-side style sheet. The
search engine, on being requested for product information,
used its best techniques to find retail outlets based on
recommendations, price, and my location. The prices of the
top 5 matches are listed along with delivery charges and any
additional information. I decide to drop the book into my
shopping basket for later. Had I decided to buy this book,
it would have taken a single click to send my name, address,
and credit card number (encrypted, of course). When
purchasing more than one item my browser checks to see if
buying from one store is both feasible and cheaper.
Amazon's entire web page can be easily reproduced using
these technologies. Your search engine soon recognises that
you're searching for a book and can instruct your browser to
use a product style sheet. This style sheet would then
reorder the results to display a cover picture, price, and
publishing information at the top of the page and fill the
bottom with extracts from reviews. It can even cross-link
to similar books based on the habits of other users.
* I could use a standard web browser and have the XML and
style sheets converted to HTML before they reach me. The
"purchase" button could be provided by the search engine.
BM
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:06:11 MST