Re: [Bug-gnupedia] Bias and formatting

[Robert Price]

At 02:42 PM 1/17/01 +0000, you wrote:
[snip]
>I've received 50 mails from people suggesting to use XML for the documents. 
>I'm not sure with that. We need something easier... common people doesn't 
>know hot to use XML so we would have to translate it...
>
>I think we should do a little HTML sheet for a basic encyclopedia article.
So 
>every article would initially look similar.

I think we're confusing presentation of the data with storing the data. If
we use XML (or SGML) for storing the data, then we can impose a strict
structure on the document. There are a number of advantages to this.
1) It's easy to translate to many various formats (HTML, plain text, etc).
2) It's easy to validate (no version specific HTML tags etc).
3) It's easy to build search engines and extract the data.

Other commercial projects use a similar methodology, for example the Oxford
English Dictionary keeps its data in SGML format, so it can just repurpose
it at will.

The data doesn't have to be entered in raw XML by the public. We can build
different front ends. If the structure is well defined, anyone can produce
an editing program to output data correctly. All we have to make sure of is
that it's the XML that comes back to gnupedia for inclusion.

Rob


--
Robert Price - Senior Developer - EMAP Digital Travel  | Tel: 0207 3092711
Priory Court, 30-32 Farringdon Lane, London, EC1R 3AW  | Fax: 0207 3092718