Archive for the ‘Aggregation’ Category

How to create a free searchable listings site

June 20, 2007

It is a constant source of annoyance to me that I have to pay for TimeOut, the London listing magazine, and they don’t have all their listings on the web. So I’ve been thinking for a while on how a newspaper could create a free listings and reviews site. I think this could be done very easily by introducing your community of bloggers to microformats. Don’t worry, it’s not complicated, here’s how you can do it.

Obviously you’ve already engaged the bloggers that write about your community. Bringing bloggers into the fold is nothing spectacularly innovative. These people should be your eyes and ears, reporting on things you can’t get reporters to, and providing leads for stories.

Now you want to aggregate the events that your bloggers are going to. After all they have been chosen as thought leaders and taste makers in your community. If you’re able to pull it together in a nice way – hey presto! – free listings. The problem is that simply searching blog posts and lumping them together will not produce a format that is as easy to use as the TimeOut listings page.

Luckily this problem can be solved by microformats. With microformats information is entered in a certain computer recognisable format that makes it easy to search and display. (The whole formatting of data is linked to Web3.0 and the semantic web – if you like using annoying zeitgeisty words). Let me explain how this would work:

i) Your approved blogger is writing about which band they will see in a week.

ii) As opposed to just writing a post about it the blogger also enters the information into an ‘Event’ microformat. Imagine it as an online form you fill in, with certain boxes for certain information.

iii) This information is entered under a certain format. Now, as opposed to searching for information by key word, your aggregator can simply can bring back all information labelled ‘Event’. Standard data to be entered could be Date, Location, Band, Genre, Preview, Band website.

iv) As this data is entered so clearly it can now be retrieved by a machine in a far more easily searchable, flexible and readable format.

v) Users can now search by Date, Band, Genre as the machine can splice and display this information easily. Want to search all bands in your area in the next week? It could display the information by Location. You get the picture.

vi) Reward bloggers who enter content in this way. You might have to give them cash, though recognition and traffic might be enough.

The ‘Event’ microfromat doesn’t yet exist, but it could be very similar to hCalendar. This is already used by some bloggers to format event information they put on their blog. It doesn’t have the Genre or Band website labels we spoke about, but I’m sure it wouldn’t be that difficult to change. (By the way, I have run this idea past Stew our developer guy and he says it makes sense).

This would be especially easy if blog platforms incorporated microformats within them as most bloggers don’t know HTML, but I’m sure there could be a way round this.

And there we go – a TimeOut killer.

ps. If there are any sites out there that already do this, then apologies, but I haven’t seen one.


What should we do with Economist Group user data?

March 7, 2007

The Bush administration has accelerated its Internet surveillance push by proposing that Web sites must keep records of who uploads photographs or videos in case police determine the content is illegal and choose to investigate. — Declan McCullagh, CNET News. (via Danah at apophenia)

It would be really worrying if this development did come to pass as the last thing I’d want is the government to be nosing around my web. I mean I don’t have a Nectar card because I don’t like companies knowing too much about me. However, some of the most interesting things that are being done on the web rely on aggregating user behaviour.

StumbleUpon suggests the sites you’ll like, based on your previous preferences. tells you what music to listen as it knows what listeners like you liked.

The more we can learn from Economist Group users the better a service we can provide them because it facilitates discoverability. It also facilitates deliverability of targetted ads.

Where do we draw the line? I don’t know yet, but it will definitely have to get a lot of thought.

Users vs Editors : The Digg example

February 27, 2007

Writing the post below, about why CNN front page gets it wrong by listing the most viewed video, got me thinking, because I love Digg, f’r’instance. So why does Digg produce interesting news (particularly if you love Macs, really love Macs, want to marry Steve Jobs, hate Microsoft, code GNU-LINUX or have attended ComiCon in a homemade super-hero outfit) whereas most viewed on CNN produces sensationalism?

The difference is accountability. If you vote for a story that sucks on Digg, then the community values your opinion less, because you’ve recommended something lame in the past. That’s why Delicious doesn’t have adult sites on its popular page, as they don’t want to be seen by others as porn users, despite the fact that the internet is made for porn.

But if a front page is just made from viewed stories no one will find out that you effectively voted to have Jordan and Peter Andre’s wedding in the top ten.

The lesson, in the realm of user generated news, shame is a good thing


Who is accountable for this?

Digg this

October 5, 2006

This blog aims to show why I’m suited to Project Redstripe in several ways. Because of the projects I’ve initiated, my startup, go-get attitude and a true love of the fast changing www. It will also show ways in which, I think, The Economist can take advantage of these changes.

This time I look at what we might learn from Digg.

Everyone loves Jeff Jarvis, founder of Entertainment Weekly and net-journo-guru, calls it the future of media. CEO Kevin Rose is the geek hero of the post-gatekeeper internet news scene.

The site works by aggregating interesting stories on the web. If other users like them others can digg/vote for them. The theory being that a popularity contest means the best ones make it to the top.

Without it I wouldn’t see 8 years of photos in one video, how to make an amazing fancy dress outfit or how to shoot yourself in the foot in any programming language.

As you can see from the last post, it is very geek driven. However, there are others moving in to this space. Netscape have just launched their politics aggregator calling the users that vote, anchors (rumours that it’s rhyming slang for Jason Calacanis are grossly exaggerated).

A question we can ask ourselves is : what is The Economist reader’s web? If The Economist Feed Reader (TM) had a button to bookmark favourite posts and pages, we could find out.

We could see which readers liked which stories from across the web and from within The Economist. Instead of being a destination in itself we would become a way of exploring the rest of the internet.

Use the hive mind of our readers and we can create an Economist web.


Even tiny minds can produce startling things