Online publishing, distribution and collaboration

Creating an online newswire:

It started with a task I set myself to find information on pirates in Somalia and filter it in a meaningful, but automated way, and send the result out to Twitter and other public facing websites as a kind of niche newswire. @somaliapirates was the result. And it continues to publish news on piracy off the Horn of Africa for those who are obsessed with that story.

I then decided to try and create a number of "country wires" - I made about 30 of them - such as @iranwire, @pakistanwire, @afghanistanwire, @congowire and @rwandawire. Each wire was built upon a single cloned/adapted Yahoo Pipe which I then fed into a country specific pipe i.e. Iran. This way I could add sources to the core pipe if neccesary - which would then filter down to all the country specific pipes. In addition, I could further tweak each country specific pipe to fine tune the results.

My idea was that I could turn the automated twitterfeed off and "go human" when a news story in any given country kicked off allowing me, or someone else, to update in near real time - both Yahoo Pipes and Twitterfeed are clunky, unreliable and information often arrives after some considerable delay - certainly not something you'd want to rely upon for breaking news.

In reality, I had no time to do all this and certainly no time for global coverage of news across 189+ countries across Twitter. And so, I let the Yahoo Pipes & Twitter combo keep on pumping out the news on the 30 or so accounts I had already created.

However, with the move to Rwanda looming back in July 2009, I decided to turn my focus to one account - @kigaliwire. I've since spent a fair bit of time focused on how to distribute news from Rwanda across the whole web and not just a fairly small, but very active, group of folk who use Twitter. That's what much of the presentation above is all about - I've added some notes, so the presentation should almost make sense without me speaking over it.

Distribute wide and far. Too far?

I didn't just want to send out links. I wanted to find ways to splice up any content I produce right across the web. I now use a combination of publish2, pixelpipe, hellotxt and various blog aggregators to do the heavy lifting for me. I now distribute content across almost 100 sites.

How does it work in practice?

Take photos as an example. Using pixelpipe I can send a photo that appears in a blog post at kigaliwire.com simultaneously appear across 45 or more photo-sharing sites. Each photo is tagged appropriately containing a description with a link back to where it sits on kigaliwire.com. The idea is that anyone searching for photos, video, audio... on Rwanda will eventually come across something I have produced and might then be inclined to see what more information there is at kigaliwire.com or, as often happens, just start a chat wherever they happen to see the image whether it be on flickr, moblog, radar, 23hq, ipernity and so on...

Conclusions this far down the road:

• This is an interesting way to publish, but will only be truly effective when I start producing more original content beyond photographs. To do that in Rwanda, I'll need a press card. which I intend to get just as soon as the price comes down from $1,000 per year to $100 per year (soon I'm told)

• Building castles in the sand - all the services I use to distribute information are free or low cost. Most of them do not appear to have much of a business model. I doubt whether a good 30-50% of them will be around in a year or two. As Ethan Zuckerman says, maybe we're building these places on the shakiest of foundations? Well... Yes and no - One of the key tasks of publishing this way is to keep an eye on what new tools are coming out and try them to see how useful they are. And to discard those that are of no use, don't work and/or start charging prohibitive amounts of cash.

• It's a lot more labour intensive to publish this way than it might look. Working on it as a kind of hobby, it took a good couple of months to wire everything up and figure out how best to try and make it work. Add another month to build and test the resulting kigaliwire.com website - which still does not work 100% - however you look at it, you're faced with a lot of time building foundations. And the thing is, much of this labour is probably futile. How many folk are there on Plerb who are likely to be really interested in Rwanda??? None, is my guess. It's more about finding those networks that are highly interested in Rwanda.

• It's more than a one-man operation. Ideally, I'd have one or two other people to create and publish content, monitor all the networks and start finding the people who are interested in Rwanda in a far more efficient way than I have done so far.

• In it for the long run. This way of publishing is a slow process, that will build over time. There are no quick fixes publishing across social media sites if you expect to engage along the way. And, is it really possible for one bloke to engage across so many social networks simultaneously? Of course not. I just focus on the ones that matter at the moment and keep an eye on those that "bubble" in the background.

Comments

[Sharif Sharifi]

I love the concept. I will be heading to Uganda soon, so I'm looking forward to contributing! Also started to follow the UgandaWire on twitter.

Cheers!