What should newspaper websites be like?

I’ve just posted most of this over on BiNS – as it’s a response to the question of what a new Birmingham Post website should be like – but I thought there were enough good, more general points (for any news organisation, or blog, even) to hack a version here too. There are people that do this sort of consultancy for a living of course, and if you’re so inclined you should seek them out.

I rarely read much news offline – I get most of my news online – I don’t have a train/bus commute to work, or a tea break, when a paper paper is what I want. I visit the website, but I won’t if  the content is too hard to find, and by the time lots of it is up I’ve got the story elsewhere.

Two things stand out for me as absolute must haves to make having the site worthwhile:

• Get all the content on the site as fast (if not before the paper) as you can. News is about immediacy, why ruin the advantage of the web by holding back stuff until it’s hit the dead trees.
• RSS feeds – I rarely go directly to newspaper sites, I like the headlines at least pushed to me so I can make that decision.

There are some other points to consider:

• Nice indexes /football, /music , /art and the like are great for navigation, or browsing when you have time – search is only one way to use a website. Search works dreadfully on most newspaper site – as the way it digs through the content is without context – so apart from how people use newspapers, that the sort of content the searches throw up is almost never what you’re looking for.
• Video and audio are nice, but certainly not must haves. Even the Guardian and the New York Times (the two run-away leaders in newspaper websites for me) don’t always hit the mark here. The Guardian’s media podcast is great, but it feels so separate from the media section of the paper, David Pogue’s tech videos for the NYT are top quality, but again not a coherent part of the web experience. These thing are an added bonus, and if the budget is limited (and it always is) I’d have to be very convinced that it was well spent on these things. If there’s video or audio content out on the web, link to it or embed it, don’t try to produce everything. That said, the odd picture would be nice.
• Links, proper links, not just ones to organisation’s websites or wikipedia pages (the beeb is very guilty here). As the story is researched, the journalist should be saving any sites they visit tagged on delicious or somesuch tagged – then they can either be pulled into the text of the story (in context, an ideal) or listed or even just linked to (”see more on this story on delicious”).
• Comment on comment pieces, but not on news? While it’s a web dream that any user should be able to comment on any story, the reality is that it’s a legal shitstorm for large organisiations to do (ever wondered why some BBC messageboards close at night- it’s cos the moderators have gone home). How much does comment on breaking news improve the site? If the only source of news thus far has been the article, then how much reasoned debate can there be? If comments can’t be opened on everything (and I’m just guessing that cost-wise that would be difficult) I prefer a clean break – you can comment on the comment pieces (which are much more like blog posts), but not on news stories – the Guardian’s Comment is Free works well here for me.
• If there are to be things labelled up as blogs then they should actually be blogs, written, comment moderated, and engaged with by bloggers themselves. Not just c’n’p’d from elsewhere, or tossed up and left to rot, comments moderated but not approved. I’ve given up with the Mail blogs, I’ve tried to engage, but the comments never appear.
• Adverts, I my opinion you can do anything you like, just don’t roll them over the text.

And them there is the geek within me that thinks, if you’re building a newspaper website from scratch in this day and age, then you may as well build room at least for the biggest things just around the corner:

• Geotagging. It’s a simple way to add location data (latitude and longitude normally) to stories – or anything else. It’s just data without some way of accessing it (and at the moment people haven’t progressed too far from laying it on maps), but it’s important and will become more so.
• APML. Stands for Attention Profile Markup Language, and is an attempt to make the data that most sites collect (about the user’s preferences and behaviour) standardised so they can be shared and used – to push only relevant content. Advertisers love this by the way, as they can really target then.

So far so, geek nonsense – but what if you combined the two. Let’s just say I live in Moseley, am a music lover, like arts, but I’m not bothered for politics or health news (not true, btw, but bear with me) – now a story about a doctors surgery in Great Barr wouldn’t interest me much, but I would care about news of a new practice in  Kings Heath. Having geodata and attention data and combining the two would mean the most interesting an relevant news for the user – a customised paper for all, automatically.

A lot of news gets more relavent to you the nearer it is…