Friday, 22 June 2007

Acronym smashup - UN meets CG for SDI Q&A

John and I were this week invited to lend an SDI hand to two important institutions here in Nairobi, namely the International Livestock Research Institute ( and the World Agroforestry Centre a.k.a. ICRAF) ( ILRI and ICRAF are two of the 15 international agricultural research centers supported by the Consultative Group on International Agricultural Research ( These are independent institutions, each with its own charter, international board of trustees, director general, and staff, but do share common requirements.

Spatial data sharing is one of these and their Consortium for Spatial Information ( is a strong advocate of all things SDI-ish. Over recent years there has been a strong push, for example, to get all the CG centres writing and using ISO 19115 standard metadata, for which they've been developing and promoting the Geonetwork opensource toolkit (which just happens to be another jointed-up effort led by FAO and WFP with UNEP and OCHA as more recent recruits).

Anyway, the CSI now has a regular scheduled hierarchical harvesting of metadata across the CG network and available as a consolidated view at their geonetwork site . What John and I were to help them with was getting started with the next step - publishing their data to the net. The two institutions need to move data between themselves anyway and the idea of doing this using open web services is not only an appealing way to go but is also a move that will position ILRI and ICRAF as their community inevitably moves build this capability in years to come.

Needless to say the CG as a whole is an important partner to the UN. FAO obviously has strong alignment with their agricultural focus but a surprising number of other UN bodies also share common interests. UNEP, for example, is co-sponsor with ICRAF of the Billion Tree Campaign ( or The idea of interoperating between the CG and UN networks is an exciting and potentially important one that seems worth investing considerable effort to pursue.

These are early days yet - the Nairobi centres have strong GIS capacity but neither of them even have a local web presence. We have a long way to go before we can point with pride at all the marvelous interoperation within and between the CG, the UN presence in the region and our respective communities. Raising awareness with management while mollifying the concerns of network administrators is going to take time and understanding, and demands that we look for a nice incremental approach that takes small steps and demonstrates success every step of the way. I wonder how we'll do it.

Meanwhile, of course, there were the inevitable glitches that confounded Wednesday's - in this case a lot of it seemed to be because we were installing geoserver etc. on a laptop that was already being hacked to an installation of ESRI's portal toolkit - the Tomcat and java installations had already been "fiddled with" and were not the nice clean environment we assumed. Lot's of time spent trouble-shooting but, in the end, considerable success, an enthusiastic audience, and lots of practical experiences and tips to write up in our SDI installation cookbook - whenever we get around to writing it.

Thursday, 21 June 2007

A real "One UN" opportunity?

Two of the goals stated early on for SDI-EA were getting the UN better able to deliver better member States by boosting their national SDI efforts, and getting at least bits of the UN working better by boosting interoperability amongst agencies as some sort of practical aspect of the UN reform and delivering as One UN. There now maybe cause for think that some of this might actually happen.

On Tuesday this week I had an unplanned but very illuminating meeting with Mr. Georges Tadonki, the senior regional information management advisor with SAHIMS (, the Southern African Human-development Information Management Network for Coordinated Humanitarian & Development Action - rather a lot to squeeze onto a business card but nonetheless important. SAHIM is in many respects the southern African equivalent of DEPHA (, the Data Exchange Platform for the Horn of Africa, the inter-agency group with whom UNEP is working closely as we try to roll out SDI in East Africa.

Georges has been engaged by the UN Resident Coordinator in Kenya to assist with the development of a Kenyan Humanitarian Information Management System (KHIM), reviewing the available information management platforms and assessing their contributions to humanitarian information management. He is also to analyze opportunities and challenges of creating some sort of central information repository, and to conduct a training workshop with IMOs from both humanitarian and development agencies, including the Government of Kenya and key donors, to build consensus on a shared platform and design a way forward for the country team.

A number of resonant chords were struck here: the potential institutional line-ups between this humanitarian sector task and the development of more general Kenyan SDI; the overlap with the OCHA Regional Offices interests in both SDI and humanitarian information development; the fact that SDI-EA has already had some engagement with ICRC and (as of last week) has also been approached by International Rescue; plus of course UNEP's focus on climate change adaptation strategies and its impacts on humanitarian issues. Interestingly enough, after Georges had introduced SAHIM and the KHIM, I gave him standard spiel #27 about UNSDI, engagement with national SDIs, open standards, distributed systems and so on, and explaining the SDI-EA effort while drawing my usual back-of-the-napkin cluster diagram when, lo!, Georges flipped back through his notes and showed us virtually the same diagram he'd sketched in a meeting the previous morning. Ahhh, convergent thinking.

Anyway, at this stage it suggests and important and powerful opportunity to align this ad hoc SDI-EA activity with some real, official UN country team activities targeting a specific community with (presumably - I'm no specialist) well-articulated needs. Now all we have to do is get the Resident Coordinator to start thinking less in terms of "central information repository" and more about distributed and custodial but integrated services. I think that's part of Georges' job.

Friday, 15 June 2007

A Busy and Productive Week

Our friends at the FAO Somalia water and land Information Management project ( this week hosted a very successful 2-day workshop on remote sensing applications in the Horn of Africa region. From the SDI-EA perspective there were two major developments of the sort that really are only possible when you have a bunch of people standing around drinking coffee during the breaks.

The first was finally meeting up with the right people from the two CGIAR institutions here in Nairobi, namely the International Centre for Research in Agro-Forestry ( and the International Livestock Research Institute ( The CGIAR is an network of international agricultural research institions and are big advocates of both spatial data analysis and interoperability. It has been a major frustration - and a considerable embarrassment - for me over the past three months to have not succeeded in engaging them in SDI-EA. That, now seems to be rectified and next week John and I expect to make some concrete progress with them.

The second coup was meeting up with one of the committee members of the Kenyan national spatial data infrastructure initiative ( A few committee members have responded on the SDI-EA mailing list but his was the first face-to-face contact I'd managed. The reason this was particularly opportune was that there's the germ of an idea to have a regional consultation in late July between the UNSDI initiative and the national SDI efforts in the region, mostly to find out what expectations countries have about how a UNSDI ought improve their service from the UN. I would dearly love to have KNSDI convene that meeting in conjunction with UNEP, and early indications are that they're keen to do so.

Now all I have to do is get UNEP's agreement and get the thing organized in 5 weeks.

Oh, and by the way, it looks like the long-awaited hands-on training sessions for open web services that UNEP and FAO/SWALIM have been planning might actually happen in the next couple of weeks, now that SWALIM have gotten the workshop out of the way. Standy by for more details.

Tuesday, 5 June 2007

OCHA joining the fray, and Africa provides development assistance to Europe.

This entry comes from my colleague, John Mugwe, after his successful expedition to the OCHA Regional Office last week. They're not quite ready for prime time but steps in the right directions are happening and now it's just a matter of getting some hardware in place.

And speaking of hardware: last week we also extended some capacity building and technology support to our benighted colleagues at the UNEP office Geneva who didn't have anywhere to safely experiment with open-standards software, particularly geoserver and postgreSQL. The fact that a colleague was here in Nairobi on a training course was too good an opportunity to pass up so I raced home, dusted off my obsolete 6 year-old pentium PC and boxed it up for its journey north. The point of this rambling? To underline the fact that it's entirely feasible to get going with open web services without major investments in hardware, software, time or effort. Sure, you can upgrade later, once your experience tells you that data services are something that you really want to get into, but you can start today and getting started is the important part.


Dear All,

Today we were at UN-OCHA (regional office) and we installed
(a) A geodatabase - Postgis
(b) Geoserver
(c) Open source desktop GIS tools UDIG And QGIS

We walked Ayub through the steps of installing and adding Web Feature Services. We converted the shape file of admin lines for Somalia to the database and created a feature service for it.
We demonstrated that through a URL he could get UDIG to portray the features from the database.

The first objective we obtained is that now he can organise his data into logical areas and put then into the database. Secondly he can share the data with other people in his offices much more easily. He can also through UDIG get data from other WFS sites.

However he has no webserver outside to the world. So his data is still locked to his small group.

I see the next possible step as
(a) describing what we have ( round table meeting) as of now I have found Somalia admin boundaries in DEPHA, OCHA, FAO_SWALIM and maybe in ICRC)
(b) having DEPHA as custodian of all reference data
(c) Having DEPHA as custodian for all metadada on the reference data.

What else? We need to move this forward.

(Soon I will ask for a pay check!)

best regards

John Mugwe