humanitarian.info

So my holiday is well and truly over, and I’m in Georgia for UNICEF on a ridiculously short contract, providing information management support for the WASH Cluster. Things are never that simple, of course, and so the work has turned out to be significantly more challenging than I expected. Right off the bat, the post-conflict situation in Georgia is a political crisis rather than a humanitarian crisis; yes, there are some tens of thousands of people displaced by the conflict, but almost none of them are in a life-threatening situation (until the winter comes, that is). Their livelihoods have been affected badly, which means that there are going to be ongoing concerns, but the scale of that problem in a middle income country doesn’t feel particularly desperate (especially now that we’re watching the footage of the monsoon floods in India which have displaced over 2 million people).

Of course that doesn’t mean that there isn’t any job here, or that I get to go back to the mountain tomorrow. There’s still a major co-ordination requirement – for IDPs that are stuck in collective centres, for IDPs that are returning home to their villages, for IDPs that have been moved into the tent camp(s) in Gori – and a real lack of decent information to support that co-ordination. Fairly obviously that’s where I come in, but the last week has not been a particularly productive one. Primarily this is because when I arrived there was absolutely no data to work with, and getting hold of it has proven to be an absolute nightmare. Information flows are incredibly weak, dialogue with the government is fragmented, the situation remains extremely fluid and there’s a lot of political sensitivies involved. On top of that, the WASH unit that I’m in didn’t exist until a couple of weeks ago; it’s been created solely because of the conflict and the need that UNICEF has to meet its obligations as the lead agency in the WASH cluster.

Bags of fun, which explains why I haven’t posted anything since I arrived. I promised myself that I was going to blog daily on the issues I was coming up against, but that’s clearly not worked out. However I will be writing a few pithy posts on specific issues, since as of two days ago data started appearing. It’s not great – patchy demographics, an improvised camp registration process, a few lists from government agencies and NGOs – but it’s a starting point. My job is to turn that data into something that can be used by the cluster to address the 5 strategic areas which we’ve identified, which are broadly:

1. Site planning of tented camps in Gori
2. Refurbishment of proposed Temporary Shelters
3. Cleaning of schools and kindergartens at national level (esp. Tbilisi)
4. Rehabilitation of existing Collective Centres (CCs) for longer-term caseload
5. Provision of village watsan for returning IDPs

As you can see, it’s not a particularly coherent set of requirements, which will make co-ordination even more difficult. The first step is to work out where the IDPs are and where they’re going to be going; the next step is to work out where the agencies are and how they’re working. Sounds simple, right?

Right.

Those crazy jokers at the UN Joint Logistics Centre have just released version 2.0 of the UN Spatial Data Infrastructure for Transport database schema, based on feedback received since last September’s release and developed with WFP and Ithaca (good to see that partnership being productive). This version covers an XML schema, the schema documentation, template databases and emergency assessment forms. The main main changes in version 2.0 include:

1. “Light” and “Comprehensive” UNSDIT packages. The most significant change has been the packaging of a “Light” subset of the Comprehensive UNSDIT package to better match information requirements in sudden-onset emergencies.

Now this is a very smart idea indeed. When you’re doing the emergency response end, the last thing you want is to wade through a massive amount of barely-relevant data or fill out a single field in a 400-field db. A light version makes a lot of sense – I’d be interested to know what the process was for deciding what what was included and what wasn’t.

2. “Light” UNSDIT Assessment forms. The same concern guided the choice to release an assessment package narrowed to the minimal set of information requirements of a sudden onset emergency response.

An even smarter idea. This is something that should be adopted by all the clusters, not just logistics.

3. ESRI Personal Geodatabase and Shapefile template databases. UNSDIT template databases aligned to version 2.0 of the UNSDIT schema are made available through this release as ESRI Personal Geodatabase and Shapefile to better serve for partners operating within an open source environment.

And it gets smarter still!

Some additional news worth reporting: UNJLC is planning to offer a Web Mapping/Reporting and a data download service at some point, which will be a considerable step forward (and hopefully be more useful than GeoNetwork is currently – jpeg maps of goat distribution, anybody?).

As well as the recent problems with public transport schemes, there’s been no small concern about whether biometrics are as secure as our governments tell us. Now The Register tells us that a hacker group in Germany has published the fingerprint of Wolfgang Schauble, Germany’s interior minister, and promises that this could be used to fool any fingerprint-based identification system. That’s not why I noticed this article – trust me, there’s going to be a lot more examples of people demonstrating that ID schemes aren’t going to deliver. What stood out from the article was this quote from Karsten Nohl:

The whole research has always been inspired by showing how insecure biometrics are, especially a biometric that you leave all over the place. It’s basically like leaving the password to your computer everywhere you go without you being able to control it anymore.

“It’s like leaving the password to your computer everywhere you go” – I’m going to have that made into a T-shirt. When cast in those terms, it makes me think dark thoughts about how these sorts of systems might be used to commit fraud against relief distributions, where any system would have to skew towards false positives rather than false negatives. I will try to flush these thoughts out in a longer post soon…

Public transport doesn’t often provide pointers for the humanitarian community. The recent cracking of the London OysterCard (following hot on the heels of the earlier crack of the Dutch transit card system) came as no surprise to digital security experts, but it should teach us fundamental lessons about information security and personal privacy issues.

Security researchers say they’ve found a way to crack the encryption used to protect a widely-used smartcard in a matter of minutes, making it possible for them to quickly and cheaply clone the cards that are used to secure office buildings and automate the collection of mass transportation fares.

No electronic identification scheme is secure. It doesn’t matter how good your technology is, any system which is built by humans can be cracked, and the only defense is to make the cost of cracking it as prohibitive as possible. (The kicker is that you never know if you’ve successfully achieved that – until somebody cracks it and it becomes embarrassingly obvious that you haven’t.) On top of that, the more complex and expensive a system is, the more difficult it is to fix it when something like this happens.

In themselves, these obstacles aren’t insurmountable – largely because they’re technical in nature – but you see the real issues when you look at how these schemes are implemented. Governmental (and intergovernmental) organisations are notorious for a) thinking that technology can fix problems which are not technical in nature (for example, running a public transport system) and b) frequently mismanaging technology projects, often with the assistance of the vendor.

In a public transport system, this is not a life-and-death issue. What if this was a tracking system for food aid, though, where RFID has begun to be introduced as the solution to our logistics inefficiencies? Or a refugee registration database in a country where human rights abuses are endemic? Or an employee identity card scheme in a country where terrorists are targeting UN and NGO offices? You start to see where this might be going…

There was also related news that MI5 have requested “full automated access” to the OysterCard database. In a liberal democracy where the rule of law holds, that might not be too worrying – but there are a number of countries in the world that don’t fit that description, and where giving access to this sort of information to the government might not be in the best interests of the beneficiaries.

The fear of cyber-warfare has climbed Whitehall’s agenda since last year’s attack on the Baltic nation of Estonia, in which Russian hackers swamped state servers with millions of electronic messages until they collapsed. The Estonian defence and foreign ministries and major banks were paralysed, while even its emergency services call system was temporarily knocked out: the attack was seen as a warning that battles once fought by invading armies or aerial bombardment could soon be replaced by virtual, but equally deadly, wars in cyberspace.

It’s only a matter of time before humanitarian organisations come under similar attack – and we’re not prepared for it in the least. None of this means that this technology shouldn’t be used – it absolutely should be. What it means is that we need to be a lot more savvy not just about the technology issues but about the entire range of processes – procurement of the system, implementation within the organisation, sensitivity to the situation (including security concerns), and so forth – in order to make sure that we’re prepared to address these situations when they arise.

There’s inevitably a data drought when you’re working in an emergency in a developing country – either the coverage isn’t good, the historical record is patchy or the accuracy is questionable. In many cases, the data simply isn’t there – nobody has collected it in the past or is collecting it at present. Where they are collecting it, the collection process often isn’t systematic and the results are in formats that aren’t easily shared – where people are willing to share their data, which they often aren’t.

In my last blog post, you might have noticed that we were trying to identify water scarce unions. There’s no actual data on water scarcity, though – it’s not the sort of thing that anybody has ever measured in itself. So how do we work out which locations are potentially water scarce?

What we did was use a proxy – a set of data that we do know that can stand in for what we want to know. In this case, we had a list of unions where tube wells aren’t feasible – derived from a couple of phone calls and some photocopied sheets. Tube wells are the primary means through which the government delivers water to communities, due to the nature of the ground and the groundwater (particularly when you’re close to the sea, salinity is too much of a problem).

Where tube wells aren’t possible, we assumed (and it was an assumption) that there would be chronic problems with water supply – problems that would have been exacerbated by Cyclone Sidr. These are the locations where the humanitarian community needs to make sure that alternatives are available – for example, water trucking to ensure a supply line, even if it isn’t sustainable – and to allocate resources for rehabilitation, particularly rehabilitating the ponds that local communities rely on where they don’t have tube wells.

Now I freely admit that there aren’t many people who are as fixated on data quality and quantity as me – most people are busy actually implementing programmes rather than crunching numbers. Yet it’s something we should be concerned about, because if we don’t accurately know the numbers and locations of people in need, then how can we possibly target assistance properly? If we don’t have good baseline data, then how can we know if our work has had any impact (especially where high poverty levels make it difficult to work out which problems are specifically caused by a disaster and how many were pre-existing)?

Standards for data management in the sector have been a headache since the beginning. Standards are essential for sharing data between agencies, whether at HQ or field level, but the politics involved in developing them have frankly defeated most previous attempts. Recent discussions in the IASC sub-group on information management have started the ball rolling in some of the clusters, but it’s still painfully slow given that we started talking about this eight years ago.

Where there has been progress is in the development of spatial data infrastructure for the UN system, a fairly comprehensive effort overseen by the UN Geographic Information Working Group. The UN Joint Logistics Center has just released the UN Spatial Data Infrastructure for Transport database schema (UNSDI-T v1.2). With the release of the database schema, we finally have a shared resource for field implementation of logistics databases – as well the starting point of a participatory process in which other users can also take responsibility for developing the standards to meet their needs.

What does this mean in practice? For a start, agencies using this schema will be able to share data more easily, which will lead to gains in efficiency (and hopefully effectiveness). Hopefully that will translate into a more accurate picture of the situation on the ground, less duplication in data collection and better information products (particularly maps) for the whole community.

Of course, all this relies on agencies actually implementing the schema, and we’ll have to wait and see how many pick it up. I’m also interested in whether (for example) the Helios software that I blogged about last week incorporates the schema – I’m assuming that it does because if it doesn’t, then any NGO that uses it is  going to be generating data which can’t easily be shared, which will be a serious barrier to co-ordination.

(UNJLC is a a dedicated logistics capacity that is hosted by WFP but sits outside the agencies – it seems to me that this positioning has been key to their success, and should be a lesson for the rest of the UN system, but that’s another story. They also credit ITHACA with the technical aspect of generating the schema, which is another lesson in working with external actors.)

Maybe they could form a support group? I know, I know, that’s not the sort of thing we’re supposed to joke about.

Yan Rebois of CartONG has written to let me know that their displaced mapping project with UNHCR is now public, an interesting application based on Google Maps, for general use by humanitarian organisations. Although it requires a log-in to edit anything, you can view the information without any authorisation.

The application maps the locations and background information on the displaced communities that UNHCR and partner agencies are working with in Uganda, centred around Gulu, which makes sense. It’s specifically listed as “operation information” rather than a public information (i.e. not just another press release).

Now what’s interesting about this application is not that it’s being used to map displaced people – we’ve seen that before, as my earlier posts showed. It’s that behind each of the locations, as well as the obvious geographic information, you get a link to a narrative Return Monitoring Site Report, which is far more useful for everybody.

It appears that the data will largely be updated by UNHCR, possibly with the assistance of its operational partners, which would be fantastic if they can manage it. However for me this raises some interesting questions about data security on the site – for example, I can see who the “community leader” is quite easily on the reports I downloaded, which might have security implications.

That’s a minor concern though – the main issue is whether this project can be shown to add value to the work being done with the displaced, as well as giving outside observers a better chance to see the scale of the problem in northern Uganda, which is massively under-reported in the media, while all eyes are on Darfur.

One thing you’ll notice when you visit country offices and field sites is that everybody uses Excel for everything. Inventory? Excel. Fleet management? Excel. Health monitoring? Excel. This used to frustrate me – why the hell didn’t these people use Access, or maybe even a proper database?

Eventually I realised that they were right and I was wrong. Spreadsheets are incredibly powerful if they’re used properly – the problem is that people generally don’t use them properly, or at least effectively. Since people are already using Excel, why don’t we just get them to use it better?

An article in DM review shows exactly what could be done by Visualizing Spreadsheets.� It’s probably a bit beyond most of our staff right now, but the discussions we’re having at the Microsoft Certified Trainer Summit 2007 could give us some new training chops…

What a marathon.� Although we actually released the report at the time of the Seattle workshop earlier this year, it was always the plan that we would launch a nicely-formatted final package – it just took a bit longer than planned – and you can now download the complete five-part Report at the ECB website Publications page.

I won’t go into detail about the contents, but as far as I know, this is the first in-depth examination of the information and technology requirements of international NGOs in emergency response.� It’s comprehensive but not exhaustive, and I hope that people will pick up the threads from this report to address specific areas such as security and staff capacity.

We’re now developing projects based on the findings of the report, all of which will be well underway (and some nearly finished) by the end of the year.� Although the projects begin with the ECB member agencies, we hope to have an impact on the entire sector.� So let’s begin!

I’m a big fan of Edward Tufte. Most normal people don’t get as excited by data visualisation as much as I do. So what kind of tools can Web2.0 provide to help make some of the more critical data more accessible?

Developed and managed by the Gapminder Foundation in Sweden (and hosted by Google), Gapminder presents a wide range of statistics in support of the Millenium Development Goals. Presenting statistics in an engaging way is difficult at the best of times, but Gapminder does it, with charts and maps. It’s easy to use, less easy to understand, and hard to beat.