h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Jim, sorry to bother you, this question might have already been asked, but I'm curious...

Who owns the copyright on the DNA source code - I presume it's the BBC.

The reason I'm asking is that I'm currently translating the Intranet site of a large company and their system, designed to allow just about any employee to publish documents using Word as an editor, is a complete nightmare, and ten minutes after I first touched it I'd figured out that DNA could do everything they do, just ten times more efficiently!


Could a version of DNA ever become commercially available?

Commercially available? I very much doubt it. It would not be worth the BBC's time and effort to try and package DNA into a commercial product. What is slightly more likely is the BBC making the code Open Source, although that, as well, would have to traverse a number of internal hurdles. It has been suggested several times here, but it's not currently on our list of things we have to do, and since the benefits to the BBC directly are negative (in that it would cost time (and therefore money) to do) it's likely to take a back seat to producing new features that DNA stie-owners want and need.

Have you thought of setting up a Wiki? It doesn't have all the features of DNA, but it's simple to set up, and most people can cope with writing wiki pages (whereas people have difficulty with GuideML - if our internal experience is anything to go by). We use a lot of wikis internally for intranet purposes.

It was more curiousity than anything else... the particular company I had in mind has already signed for a new system for next year - custom built with the user content produced via document templates churning out XML - hellishly complicated and hellishly expensive!

I don't think they'd have gone for Wiki - not flash enough for a company with several tens of thousands of employees and sites in 3 continents - Still, if you ever get fed up working for the Beeb there's got to be a market out there - Here's a thought experiment for you if you're feeling bored one day though What woud be the best way of setting up a multilingual DNA site - separate sites for each language or just separate skins?

We've designed the skins so that all the text (messages, link text, etc) is stored in included templates, so a multi-lingual site would simply have to replace a single file with a translated version. This is pretty standard in software - keep all your localisable resources in a separate place that can simply be switched in or out. It's only the user-generated content which can't be localised.

Your system sounds a bit like the system the BBC is currently implementing for production of all its flat page - a very complex and powerful content management system based on Java, XML and XSLT. I don't envy the team working on it - it's a huge task to get *anything* standardised across the whole of the BBC, which is the aim.

Sounds about right - I think the user content is probably going to be generated by something like Interleaf or Framemaker... But of course, the other slight problem this particular bunch have got is they've completely lost control of who can publish stuff - the staff have this wonderful habit of using each other's logins - they've got stuff on their site which has 'apparently' been written by people who left the company (or even died!) years before the pages were written ... Maybe the BBC isn't so bad after all