humanitarian.info

Here’s another thing to throw into the mix. Part of my thoughts (parts of my thought?) was generated by a post by Ethan Zuckerman entitled Watching Madagascar, via Twitter. He starts off strong:

The nature of breaking news is changing… The (confusing, apparent, partial, incomplete) coup in Madagascar is the first event I’ve been able to watch only through social media.

But a few paragraphs later, the big reveal:

So I’m doing what my Malagasy friends across the net are doing – religiously watching the #Madagascar tag on Twitter. That means I’m primarily reading Thierry Ratsizehena, a marketing and social media expert in Antananarivo, who is listening closely to news via television and radio, and sharing what he knows with his Twitter readers. Lova, who’s in the US, is translating his tweets into English and adding context and commentary.

I don’t doubt that these two were working as an “effective news bureau” for Ethan and other interested observers, and this is clearly a useful service in the Twitter manner, but I couldn’t help but notice something that Ethan had slipped in there. Thierry Ratsizehena was “listening closely to news via television and radio” and then sending that on to Twitter. While Ethan may have been literally watching events through social media, he was getting all his news from television and radio. Ethan’s not making any claims for Twitter but in this instance I’m not sure Twitter is doing much more than ham radio would?

I don’t want to harp on about Twitter, really I don’t; it’s pretty much irrelevant to any of the work that I’m likely to be doing in the near future¹. The reason I find the coverage of Twitter interesting (rather than finding Twitter itself interesting) is that in some senses it’s clearly a fad (in the same way as most technology journalism is fad-based) and in some senses it clearly represents a shift in the foundations – although I don’t think that it is that shift, which is what the breathless news coverage tries to suggest.

Perhaps it’s as simple as this. As our traditional media dies off (as per Clay Shirky’s recent article), people are looking for something to take its place. In this case, Twitter looks like journalism – they’re broadcasting reports from on (or near) the spot! – but it isn’t journalism, and it doesn’t possess the powers of organisation that people seem to think it should at first sight. We want it to be journalism because we want something to take journalism’s place – or in Ethan’s case, fill the gaps that his traditional media leaves in his coverage of the world. In countries where traditional media retains its position – particularly radio, which shows little sign of dying out in developing countries – journalism is alive and well, and in fact provides the raw material on which Twitter users draw.

1. I’ll go out on a limb and predict that it won’t be relevant to any of the humanitarian work that I’ll do in the future at all. [↩]

Protests in Moldova Explode, With Help of Twitter!

A crowd of more than 10,000 young Moldovans materialized seemingly out of nowhere on Tuesday to protest against Moldova’s Communist leadership, ransacking government buildings and clashing with the police. The sea of young people reflected the deep generation gap that has developed in Moldova, and the protesters used their generation’s tools, gathering the crowd by enlisting text-messaging, Facebook and Twitter, the social messaging network.

Here’s how it went down:

The related posts on Twitter are being posted at a record-breaking rate – I’ve been watching the Twitter stream for the last 20 minutes – and I see almost 200 new Twitter messages marked with “pman” (virtually all of them in Romanian, with only one or two in English)… All in all, while it’s probably too early to tell whether Moldova’s Twitter revolution will be successful, it would certainly be wrong to disregard the role that Twitter and other social media have played in mobilizing (and, even more so, reporting on) the protests.

Or did it?

In fact Twitter did not play that big role. The story is quite simple – young and active bloggers decided to have a flash-mob action, lighting candles and “mourning Moldova” because of Communists victory… They agreed on the time and place of the action through the network of Moldovan blogs (blogs aggregator blogosfera.md), and social networks like Facebook/Odnoklassniki, etc…. That was a civic protest, which grew up out of a flash-mob initiative organized through blogs and social network connections, and then which grew even bigger as the protesters used mobile phones to summon their friends and classmates.

Apparently it didn’t.

Mihai Muscovici… suggests that the Twitter community in the whole of Moldova is around 100 to 200 strong and there is scant mention of the organisation of the protests at all apart from a rather vague quote the Times has put in at the end of the piece… As it stands, the Twitter revolution is a myth. What happened, and is still happening, in Moldova is a protest organised using social media.

Last word to Evgeny:

It really helped that even non-technology people in the U.S. and much of Western Europe are currently head over heels in love with Twitter. It’s really good that the Moldovan students didn’t organize this revolution via Friendster or LiveJournal (which is still a platform for choice for many users in Eastern Europe). If they did, they would never have gotten as much attention from the rest of the world.

Indeed. The reason I’m posting this – even though it’s not strictly speaking “humanitarian” – is because it shows three things which to some extent follow from each other:

1. It’s hazardous to use press coverage to determine what tools are being used and what tools are working in a crisis. The press frequently have even less understanding of the tools than they do of the crisis, and they <em>will</em> focus on what’s “popular”. It’s very difficult to verify the claims about these tools while the crisis is unfolding, so I find it hard to blame them – they need people to read their stuff – but the people making the claims need to be more careful.¹
2. They need to be more careful because media coverage of technology is the product of the echo chamber that dominates the technology sector – and steers the media when it comes to reporting on technology. No offense to all those tech guys providing their opinions to the media – they are smart, no doubt – but because they’re tech guys they love Twitter, and discuss it a lot, so that’s the headline we read.
3. The good news is that Facebook (and other social networking sites) have demonstrated their utility as organising tools in advance of a protest; mobile phones we already knew are of maximum utility even while the crisis is full on (right up until the network goes down). Twitter – maybe good visibility for people on the periphery or further out, but I’m just not convinced that microblogging in general is of much use in a crisis. Head over to Jon’s place for the opposing point of view.

1. To be fair, in that initial article they only refer to Twitter as part of a suite of tools, but the general tenor of coverage has been “The Twitter Revolution” or something similar. [↩]

One of the claims made by Patrick in his defense of crowdsourcing:

Fifth, all the humanitarian organizations present during today’s meetings embraced the need for two-way, community-generated information and social media. Yet these same organizations fold there arms and revert to a one-way communication mindset when the issue of crowdsourcing comes up. They forget that they too can generate information in response to rumors and thus counter-act misinformation as soon as it spreads. If the US Embassy can do this in Madagascar using Twitter, why can’t humanitarian organizations do the equivalent? [my emphasis]

I responded:

Is there any evidence at all that the US Embassy’s Twitter feed had any impact at all on the course of events? I mean, I know it made a good headline in external media, but I don’t see how it’s a good example if there’s no actual evidence that it had any impact.

Patrick’s evidence was an assertion that the rumours didn’t spread. I find this to be implausible in the extreme, and I’ll let Jeff Allen explain why:

When the Liberian banks started charging a differential exchange rate (the same as their partner banks were doing in London), that reality-based fact morphed in the street into “the banks don’t take little heads”. The US Embassy put out a press release to try to stop the rumor. It said, “Dollars are dollars, big head or little. Every dollar anywhere on the planet can be exchanged for any other, and they are all dollars.” Which would be true, except it’s not. If you try to bring $10,000 from Liberia and spend them in the US, the odds that you have a counterfeit bill in there someplace are high enough that you’ve probably brought (on average) $9994 instead of the $10000 you thought you did.

In any case, a press release from the embassy certainly wasn’t enough to stop this story. Whether becaused they beleived the rumor, or because they just didn’t want to be the only one not believing it (the musical chairs effect), within a few days the vendors stopped accepting little head notes. This was a few days before payday, and several staff brought the story to me, worried I would pay them in little head notes that they could not spend in the local market. I showed them the newspaper, and told them a dollar is a dollar. They told me, “a dollar I can’t spend isn’t a dollar”.

This is my experience of dealing with rumours in the field. Most of the time, trying to counter them is a waste of time – you just have to work around them. In this case, MSF withdrew all the little head dollars from their financial system, and the problem was solved. Notice that the US embassy press release failed to reassure the market, and I don’t see why a twitter message (or any other web-based tool) would have any greater effect (even if Liberia had better connectivity than it does). This isn’t just a theoretical question, it has serious practical implications:

PS: Think rumors are funny? Managing rumors and knowing when to give up and get out of the way is serious business for humanitarian aid workers. Here’s an article about 3 Red Cross workers killed due to a rumor. My boss didn’t make the decision she made because she’s a nice lady. She decided this wasn’t a rumor that we were going to kill, and we needed to get the heck out of the way of it.

I agree that the humanitarian community needs to improve its communication in the field, particularly with beneficiaries, and that there are a range of tools with which to do so. However we need to be realistic about what those tools can achieve, and in this case recognise that when we feed “facts” into the rumour mill, it won’t stop the rumour mill from grinding – those facts just get all churned up with all the other rumours into a great big tasty rumour pie.

Nobody can deny that Clay Shirky is a clever guy, and that he knows his beans when it comes to the world of social software.¹ Part of his argument is that a) television has been masking the sheer amount of leisure time that we have to participate in communal activities, and b) the information revolution has provided us with the internet as a vehicle to make that leisure time more productive. This hypothesis of cognitive surplus seems reasonable, although one can argue with his assertion that making lolcatz is better than doing nothing, on the grounds that making lolcatz is at least doing something.

The exact means by which the internet enables us to capitalise on this surplus vary, but a lot of emphasis is placed on social software of various kinds – Web 2.0, as it’s sometimes known. In the last couple of weeks we’ve had some discussion about one general application – the possibility of humanitarian wikis, and the potential of crowdsourcing, the ne plus ultra of Web 2.0 – but throughout these discussions I’ve played the role of the sceptic. Shirky’s essay provides me with a starting point to address the most basic concern that I have about the role of crowdsourcing and other approaches in the humanitarian community.

The normal case of social software is still failure; most of these experiments don’t pan out. But the ones that do are quite incredible, and I hope that this one succeeds, obviously. But even if it doesn’t, it’s illustrated the point already, which is that someone working alone, with really cheap tools, has a reasonable hope of carving out enough of the cognitive surplus, enough of the desire to participate, enough of the collective goodwill of the citizens, to create a resource you couldn’t have imagined existing even five years ago.

This raises a problem which I think is often overlooked in our predictions about how these tools might change the sector. Humanitarian workers, generally speaking, don’t have a whole lot of cognitive surplus. If you’ve ever worked in disaster response – even on the periphery – you know that the working day can sometimes last literally the entire day. Even when you’re not working, you’re working, because you’re still in the crisis area and you’re essentially on call if something goes wrong (especially in areas which are insecure). Your cognitive surplus will not be focused on your humanitarian work; and even if you are inclined to contribute to a humanitarian wikipedia (for example), you are very likely be in a tiny minority.

It seems to me that this poses a serious problem for those expecting aid workers to be engaged participants in experimenting with these new technologies. It seems that this basic obstacle undermines many of the claims about how the humanitarian community might use these tools, although this doesn’t mean that those technologies have no future and it definitely doesn’t mean that we won’t be affected by their spread, especially as the web becomes increasingly mobile.

1. If you haven’t read Here Comes Everybody yet, please do so, and enjoy listening to this extended interview which riffs on many of the economic arguments in that book. [↩]

Alex Evans at Global Dashboard:

The point about Twitter and other social networking technologies is that in our hyper-networked age, we just haven’t yet had the time to develop the collective mechanisms to make sure that this awesome power to aggregate, to build positive feedback loops, is channelled safely.

Erik Hersman at The Ushahidi Blog:

Since we don’t believe there will ever be one tool that everyone uses for gathering information on global crisis, we see a future where a tool like Swift River aggregates data from tools such as the aforementioned Twitter, Ushahidi, Flickr, YouTube, local mobile and web social networks. At this point what you have is a whole lot of noise and very little signal as to what the value is of the data you’re seeing.

Anyone who has access to a computer (and possibly just a mobile phone in the future), can then go and rate information as it comes in. This is classic “crowdsourcing”, where the more people you have weighing in on any specific data point raises the probability of the finding the right answer. The information with greater veracity is highlighted and bubbles to the top, weighted also by proximity, severity and category of the incident.

The question is, how viable is a tool like Swift River as one of the “collective mechanisms” that Alex correctly identifies the need for? I think the Ushahidi developers are on an interesting track, but I think that there are limitationss to what crowdsourcing can achieve – not a problem when it’s a forum like Digg (for example) where the weight of numbers has a levelling effect on any individual distortions, and where the ratings are trivial.¹ However I’m still waiting to be convinced about the value of crowdsourcing in an emergency, because it’s crowdsourcing of the type which Alex describes in his example of a potentially dangerous Twitter meme.

1. I don’t mean “trivial” in a derogatory way, only in the sense that nobody’s going to make potentially life or death decisions on the basis of a Dugg article. [↩]

After much thought and a bowl of noodles (udon not ramen), I decided to clarify my thinking. Online platforms come in five flavours:

1. Response management – for the immediate information needs of an emergency. Example: Sahana.
2. Collaboration space – for both emergency and non-emergency. Example: Google Groups.
3. Community of Practice – for developing and maintaining professional networks. Example: ShelterCentre.
4. Knowledge base – for creating institutional and sectoral memory.Example: WaterWiki.
5. Capacity development – training and other development requirements. Example: various under development.

These five functions can be combined, although not necessarily successfully, but all the web-based work I’ve seen or done fits these functional descriptions. There seem to be three main areas which cause problems in terms of implementing these platforms:

• Setting up a platform without clearly identifying which function it fills, or combining different functions without working out how they’ll relate to each other within the platform;
• Creating multiple platforms that overlap, without reference to each other and without considering interoperability issues, thus increasing user confusion, workload and migraine;
• Building platforms without consulting end users, assuming they’ll welcome yet another chance to spend hours at their computer.

It’s impossible to build the perfect platform, even within a single organisation; user expectations and capacities are simply too diverse; avoiding these three elementary mistakes is a good start, however.

I promised Lorant Czaran¹ that I would post something about ReliefWeb’s job vacancies mash-up – and then realised that it would fit perfectly with this discussion about innovation. I’ve written about ReliefWeb before, but it’s not exactly noted for cutting edge web 2.0 efforts. However, no matter what its faults, it’s the single most important website for the humanitarian community, which is why it’s good to see them trying some new approaches.

I’m not claiming that searching for jobs on ReliefWeb is an especially humanitarian activity, but it is an activity that a lot of humanitarians do. One of the things that I completely failed to understand about their redesign a a couple of years ago is why they relegated the Vacancies section to the bottom of the “Professional Resources” part of the site, rather than foregrounding it. The reason for foregrounding? Simply that vacancies are one of the main reasons that people go to the site in the first place.

Vacancies shouldn’t be the focus of the entire site, but they should be a focus of the user experience. Lorant and his team took a strong step forward here, introducing a new way of interacting with the site – a Google Maps mash-up which can also be downloaded as a Google Earth KML file. This is a great idea, and well executed, given ReliefWeb’s disturbingly 90s website design – but why does this fit with the innovation theme? AFter all, Google Map mash-ups aren’t exactly new, even in the humanitarian sector.

What’s innovative about it is that it shows the way forward not just for job searches but for the entire ReliefWeb site. There’s no reason why the enitire site couldn’t be organised in this way, with navigation based entirely on geography – after all, that’s the way the humanitarian community itself works. I’d love to see this approach extended to become the front page of the website, offering a way into the main Countries and Emergencies section. There’s very few parts of the site that don’t offer themselves up to a geospatial interface.

So what about those few parts that don’t – Policy and Issues, for example? Well you couldn’t do a Google Maps mash-up for those things – but why couldn’t you do a policy map instead, showing the different links between sectors and institutions? Or a tag cloud approach, showing which issues are the ones that are generating the most publications and discussion? Either of these would offer a better user experience that would make ReliefWeb not just important but innovative as well….

1. Ex-ReliefWeb, now UN-Spider, which is the most awesomest UN acronym ever, by the way [↩]

From Bartel:

We have been working in the past few months on the development of ISCRAM LIVE, an “ISCRAM 2.0” dynamic site gathering and publishing content that is being posted on popular “web 2.0” websites by ISCRAM members. The ISCRAM LIVE website is at http://www.iscram.org/live .

ISCRAM LIVE currently interfaces with slideshare, youtube, flickr, twitter, delicious and Facebook, and collects (on a daily basis) all posts on these sites that are tagged with the word “ISCRAM”. I would now like to ask you for your help and a bit of your time in the coming week and try out ISCRAM LIVE in the coming week (say until October 16), by posting and tagging items on these sites that you think are relevant or of interest to the ISCRAM Community – blogs, pictures, tweets, slides, videos and del.icio.us bookmarks.

Now: this was what we were hoping for with the ICT4Peace website. The problem was two-fold – the original web designers couldn’t deliver web2.0 (admittedly it was 4 years ago…) and there wasn’t a community around the concept. ISCRAM now has both of these, so the question is – what’s the magical x-factor that will make this take off? I’m doing my bit – TAG!

David Sasaki joins the conversation, which is great – it was starting to feel a little bit like a mens singles tennis match between me and Patrick. Now it’s mens doubles, or something. David starts with a strong serve, although his accusation that

Both men seem to have the academic tendency to speak in aphorisms

seems a little unfair – the heuristics post he’s referring to was simply me reminding myself that I’m not the great oracle on these issues, and that I should get ready to be wrong.¹

If I understand Paul correctly, his two main criticisms of Global Voices are that 1.) it doesn’t matter if you highlight moderate voices discussing the news of their countries because it is the extreme voices who will always make the headlines and 2.) during times of conflict and emergency, focusing on participatory websites rather than humanitarian institutions will lead to lots of conversation, but less action.

No. I’m not criticising Global Voices per se, and definitely not on those particular grounds. I think Global Voices is amongst the best that the Web has to offer. What I worry about is making claims about the impact of projects that go beyond a) what the evidence shows to be true and b) what those projects can realistically expect to achieve. Global Voices meets its stated aims convincingly, but what worries me is when people start talking about Global Voices – or blogs in general – as something which they’re not. As David notes,

We often portray Global Voices as the zeitgeist of what the ignored world is discussing when in fact we are an amazing international community of individuals with large online networks and particular interests.

David’s honesty is admirable, and I think that honesty reflects one of the strengths of Global Voices in general. What I was taking issue with more was Patrick’s statement that

Global Voices is a far more effective local information and response network than FAST ever was.

I simply disagree with this.² Global Voices is not a response network in any substantive sense, and I don’t think it’s necessarily a more effective information network either. I agree that there ought to be more attention paid to blogs as a source of information, but the strength of GV is precisely that it is not programmatic. The bloggers involved have not set themselves objectives to provide early warning information, or document human rights abuses – they are just private citizens who are writing about issues that are important to them. The situation is slightly different with Ushaidi, of course, which was conceived and developed specifically in response to the post-electoral crisis in Kenya. In the words of Ory Okolloh,

Ushahidi was mainly intended to be a mapping tool and a repository of information about the post-election crisis as seen from the view point of people on the ground. We were trying to capture information that was not mainly being reported in the mainstream (there was a lot of self-censorship in the media) and also provide a timeline for information for both mainstream and citizen reported events. In the case of real time mapping Ushahidi could be used to track where the violence or the peace efforts were taking place. We hope to be able to provide those people who are “addressing the real needs to real people” with information that might help their efforts and to be part of the “testimony” as it were of what happened.

Now that’s a series of specific objectives that can – and should – be measured in order to judge the impact of the project. However if you look at the underlying requirement for all of those objectives to be met, it seems to me that the basic requirement is a systematic data collection system – which is exactly what Ushahidi did not have. It’s entirely possible to run a Ushahidi instance with a more systematic foundation – but then it stops being the Web 2.0 poster child that everybody wants it to be, and becomes a visualisation tool for a standard human rights monitoring system.

Now I don’t have a problem with that – it’s not as if we’re over-supplied with really great data visualisation in the human rights field – but that’s not why people got excited about it. People got excited about it because it’s a Revolutionary New Way Of Doing Things Just Like Clay Shirky Says, and I’m asking what I hope is a valid question: it may be a revolutionary new way of doing things, but is it a better way of doing things? Maybe it is – in which case, show me.

I think this tension is at the heart of most of these initiatives. Patrick unwittingly gave away one of the reasons why he thinks bloggers are better than the established systems, and it goes right to the heart of this tension.

Unlike the local information networks at FAST and conventional conflict early warning systems, they are not paid informants.

This belief is part of the cult of the amateur that I think the internet has reinforced, but it is not inherently better to do something for free than it is to do it for pay.  Personally I think that as soon as they stop acting as bloggers and start acting as human rights monitors, they will cease to be good bloggers – and they probably won’t be very good human rights monitors either. I also think that the strengths of citizen journalism – the amateur spirit, the personal perspectives, the improvised approach – are in this context potential weaknesses. Joshua at Registan almost nails one of the key problems for Global Voices when he says that

too many internationals, including me, are far more alike each other than they are to their home countries.

Even though many of them are from the regions or countries that they cover, the Global Voices bloggers – in certain important ways – are more like each other than they are like the people in their home countries. In particular, they share “democratic values” just as Patrick describes, and a positive, can-do attitude that impresses people.³ Yet those democratic values may be the very thing that makes them less representative, and that raises an interesting dilemma for David and the others who are interested in Rising Voices.

In relation to Ushahidi, I wrote

The virtual world isn’t resistant to real-world pressures, and it doesn’t necessarily overcome social divisions – hence the problems with the [Mashada] bulletin board. These pressures can be managed, but it’s no easy thing – but would Ushaidi be any less resistant to hijacking by people intent on promoting social divisions?

I suppose that’s my question, in the context of David’s job – what defense mechanisms do we have against the real world?

1. Besides, academics don’t usually talk in aphorisms – they prefer to maximise their word count. [↩]
2. Although that doesn’t necessarily mean that I think FAST is particularly effective. [↩]
3. However a positive attitude is not enough – I have frequently said that I would prefer to work with people who really couldn’t give a damn about humanitarian issues but who are excellent at their jobs, than work with people who are lousy at their job but who really, really care. I’m not saying that the Global Voices bloggers are lousy at their job – but their job is not “early warning”, it’s blogging. [↩]

Strictly speaking, shelter is outside the remit of this blog. However the article The Exigent City in the New York Times was an excellent piece of reporting, and I urge everybody to read it. (HT: Simmy Ross) The article closes with the following lines:

When I first contacted Cameron Sinclair, who started Architecture for Humanity with Kate Stohr in 1999, he replied with a long e-mail message that began, “You’ve stumbled upon the question of the century for those working in the built environment,” and ended with a signature line that listed his title as “Eternal Optimist.” A lot of people I spoke to afterward agreed with the first sentiment; almost no one with the second. The world is short a billion homes. Now all we need are a few million architects to help build them.

I’m not optimistic either, but nor do I agree that we need a few million architects. What we need is to help the people who need a billion homes to become their own architects – to develop simple, workable construction forms that can be disseminated easily and virally. There’s a role in that first part for architects, urban planners and shelter experts, sure – but the second part can only be achieved if we take a different approach to the actual construction process.

We can’t can’t rely on central planning to take into account the massive and growing improvised urban areas – at best they’ll be able to provide basic services, but not even that is guaranteed – so poor communities already build their own dwellings, which is where shanty towns come from in the first place. The poor are already their own architects, so the only solution is to help them become better architects who build better dwellings – better in the sense of being more robust, more environmentally friendly, more durable and more liveable.

So we come back around to the role of ICT, which for the first time offers a way of disseminating this information on a large scale at relatively low cost. The question is, what’s the best vector for that dissemination, how do we ensure access to the information, and what forms do we need to use to make the information relevant to the end users?

Answers on a postcard to the Shelter Centre, I guess….