Map Africa and Open Source Geospatial

I've just spent a few days at Map Africa and this has been a very interesting conference.  First of all it has not been a traditional GIS conference.  There has been a strong focus on capacity building, specifically in Africa as you would expect.   I found two particularly interesting things about this conference.

Open Source Geospatial

At the end of the first day there was a panel discussion on Interoperability and Open Source.  The panelists were Andrew Terhorst from the Meraka Institute in South Africa, Professor Dasgupta from India, and myself.  Andrew provided a uniquely African perspective on open source.   He was vocifereous in arguing that open source is good for South Africa becaue open source creates an opportunity for the development of human capital in South Africa.  He also reiterated what I think is generally recognized that security is not a reason to not use OSS, but he did allow that OSS needs to improve its usabality and that this is the primary barrier to its adoption. He encouraged people to be more adventurous and to use open source because it's a way to help Africa to develop its human capital.  As he pointed out himself, he has an academic perspective and he needs to be able to see code to undertand the algorithms used.  For example, he likes GRASS because he can see how things work. In this context I found it interesting that he sees open source as not only providing a way for people to collaborate but as a new form of peer review.

My perspective is that both open source and closed source have necessary roles in the software industry.  For example, I believe that the state of web mapping servers is analogous to web (HTTP) servers in the late 90's, and that this area of the geospatial industry is ripe for commoditization and open source.  Desktop geospatial applications is another matter - there is still lots of room for closed source vendors to differentiate themselves in this software segment.

I detected a serious interest in open source geospatial among the South African participants at this conference.  What was most interesting to me from what I heard from a number of people with whom I chatted at the conference is that in South Africa cost is one of the most important reasons motivating this interest.  There is what appears to be a widespread perception in South Africa that traditonal GIS is too expensive.  Part of this is no doubt due to the fact that in the developing world, if you express the cost of software in terms of the local per capita income, the cost of software is much higher than in North America.   For example, I remember seeing some time ago that in Bangladesh Windows98 cost the equivalent of three years' income for the average Bangladeshi.  But South Africa is quite different from Bangladesh.  For example, Gauteng province, which includes Johannesburg and Pretoria, represents 37% of the African economy.  So I think this is only part of the reason.

Free Digital Spatial Data

One of the remarkable things I discovered about South Africa is that data collected by the national government, mainly by the Department of Land Affairs: Surveys and mapping, is free in digital form.  This is remarkable.  The country where spatial data was first made available at no cost was the US where the Federal Government's guiding principle is that if taxpayers paid to collect data, the government can't charge them again.  Many people believe that one of the reasons that the US is such a major force in the geospatial industry, both in the areas of software applications and data providers, is because of the availablility of low cost data.  I think this means that great things are in store for the South African geospatial industry.

Comments

[Maurizio]

I agree: it has not been a traditional GIS conference. Interaction between
delegates was superb. I wish to thank the organizers for their efforts.

I wish to add that the SA 250k and 50k maps are available on line (courtesy
of CDSM) for free download at http://www.madmappers.com

Since 2003, when South Africa first introduced a GIS dissemination policy
based on cost/media (approx US$3/CD) the GIS industry has boomed beyond any
expectation. We do applaude this bold move which has broken the monopoly of
few rich companies (the only ones who could afford to purchase data in the
past)and has resulted in hundreds small GIS companies and software
developers entering the market.

The next step is the introduction in 2007 of GIS as subject in the last 2
years of South African high schools.

It is also very exciting to hear that similar dissemination policies are now
been considered in neighbouring countries. It should be quite obvious by now
that it is the only way to stimulate the birth of a African GIS industry

[Ragnvald Larsen]

I am working with Tanzania Wildlife Research Institute as an intern/consultant. We are using open source tools like MapServer, MySQL, PHP and Apache. It allows us to work on projects like this one:

http://www.mindland.com/countless_qdgc_demo/

Functional National Spatial Data Infrastructures are important as well. Of those countries supplying development aid to African countries it seems to be more popular to support physical infrastructures. Never minding that the spatial data for the same roads get lost. I believe that it is conditional for any physically manifested work to have the according data in a Spatial Data Infrastructure. Failing to focus on just one of them might prove counter-productive.