[This post is based on the Information Sciences MSc. thesis by Onno Valkering]
To make widespread knowledge sharing possible in rural areas in developing countries, the notion of the Web has to be downscaled based on the specific low-resource infrastructure in place. In this paper, we introduce SPARQL over SMS, a solution for exchanging RDF data in which HTTP is substituted by SMS to enable Web-like exchange of data over cellular networks.
The solution uses converters that take outgoing SPARQL queries sent over HTTP and convert them into SMS messages sent to phone numbers (see architecture image). On the receiver-side, the messages are converted back to standard SPARQL requests.
The converters use various data compression strategies to ensure optimal use of the SMS bandwidth. These include both zip-based compression and the removal of redundant data through the use of common background vocabularies. The thesis presents the design and implementation of the solution, along with evaluations of the different data compression methods.
The application is validated in two real-world ICT for Development (ICT4D) cases that both use the Kasadaka platform: 1) An extension of the DigiVet application allows sending information related to veterinary symptoms and diagnoses accross different distributed systems. 2) An extension of the RadioMarche applicationinvolves the retrieval and adding of current offerings in the market information system, including the phone number of the advertisers.
For more information:
Download Onno’s Thesis. A version of the thesis is currently under review.
Downscale2016 follows success of previous Downscale workshops and will mostly focus on appropriate infrastructures. Instead of using large-scale centralised approaches to data management we look at breaking data-centric architectures into smaller components that consume less electricity, be cheaper to own, and more flexible than a “big server” while still mimicking, as a swarm, the features one such big server would provide. As such, the workshop matches ICT for Development (ICT4D) goals with ICT for Solutions (ICT4S) and we expect that the dialogue between ICT4S, Semantic Web and ICT4D researchers and practitioners will further each of the research fields.
We are currenty inviting both short papers (6 pages) or abstracts (2 pages) describing current or latebreaking research in ICT4D. These papers will undergo a light review procedure. For more information, visit the workshop web page.
Today, the second international VU symposium in ICT for Development was held. As last year, the workshop was a great success, with an international host of speakers and a variety of attendees (around 80 people joined).This year’s symposium we looked at the opportunities and challenges for “Data for Development” from many angles. In his keynote speech, Gayo Diallo from Unversite de Bordeaux elaborated on how data from mobile telephony providers was used to identify issues with access to health care in Senegal. Marije Geldof discussed the success and difficulties in using mobile data services for assisting health workers in Malawi. After these longer presentations, a series of duo-presentations were held. In the first the concept of upscaling and downscaling (big) data sharing solutions was discussed (Hans Akkermans and Christophe Gueret). In the second duo-presentation we heard from two Amsterdam-based organizations on the use of Open Data for aid transparency (Rolf Kleef) and how to connect data from different mobile projects (Karl Lundfall). The final duo-presentation featured Cheah Waishiang on how to connect to local communities using ICT in Malaysia and Chris van Aart who described the approach of the App-developer. Myrthe van der Wekken and Gossa Lo presented their research on Knowledge Sharing for the Rural Poor through a quick pitch and two very nice posters (see also their reports 1 and 2) . All in all, the symposium showed that in every stage of the data value chain, there is progress being made in the development context. However, there are enormous challenges to be overcome at each stage as well. Enough to work on for a next installment of this yearly symposium series. You can watch the entire symposium through the embedded video below (3 hrs). Below the video you can see the list of speakers and the different timestamps in the video when their talk starts (clicking on the link will open in new window). [youtube https://www.youtube.com/watch?v=s7JO_R9-x6k]
Gayo Diallo – Université de Bordeaux, Bordeaux, FR “Mobile Data in Senegal, a Health Decision Enabler” (6.58)
Marije Geldof – ICT4D professional The Hague, NL “‘Mobile health and the role of data in Malawi’” (45.05)
Hans Akkermans – The Network Institute, VU Amsterdam, NL, “Community-centric Data Services (1.12.00) for Social & Economic Development in Africa”
Christophe Guéret – DANS-KNAW The Hague, NL “Downscaling the (Semantic) Web: Decentralized Linked Open Data for World Citizens” (1.22.40)
Rolf Kleef – Open for Change, NL “Open Data for Development Agencies” (2.04.30)
Karl Lundfall – Text2Change, NL “Integration of Data Sources for Development” (2.15.18)
Cheah Waishiang – Universiti Malaysia Sarawak, Malaysia “Empowering & knowledge through digital storytelling in Borneo, Sarawak, Malaysia” (2.28.26)
Chris van Aart – 2CoolMonkeys, Utrecht, NL “Mr. Meteo, Weather forecasts for African farmers” (2.41.30)
DOWNSCALE 2013, the 2nd international workshop on downscaling the Semantic Web was held on 19-9-2013 in Geneva, Switzerland and was co-located with the Open Knowledge Conference 2013. The workshop seeks to provide first steps in exploring appropriate requirements, technologies, processes and applications for the deployment of Semantic Web technologies in constrained scenarios, taking into consideration local contexts. For instance, making Semantic Web platforms usable under limited computing power and limited access to Internet, with context-specific interfaces.
The workshop accepted three full papers after peer-review and featured five invited abstracts. in his keynote speech, Stephane Boyera of SBC4D gave a very nice overview of the potential use of Semantic Web for Social & Economic Development. The accepted papers and abstracts can be found in the downscale2013 proceedings, which will also appear as part of the OKCon 2013 Open Book.
We broadcast the whole workshop live on the web, and you can actually watch the whole thing (or fragments) via the embedded videos below.
After the presentations, we had fruitful discussions about the main aspects of ‘downscaling’. The consensus seemed to be that Downscaling involved the investigation and usage of Semantic Web technologies and Linked Data principles to allow for data, information and knowledge sharing in circumstances where ‘mainstream’ SW and LD is not feasible or simply does not work. These circumstances can be because of cultural, technical or physical limitations or because of natural or artificial limitations.
The figure illustrates a first attempt to come to a common architecture. It includes three aspects that need to be considered when thinking about data sharing in exceptional circumstances:
Hardware/ Infrastructure. This aspect includes issues with connectivity, low resource hardware, unavailability, etc.
Interfaces. This concerns the design and development of appropriate interfaces with respect to illiteracy of users or their specific usage. Building human-usable interfaces is a more general issue for Linked data.
Pragmatic semantics. Developing LD solutions that consider which information is relevant in which (cultural) circumstances is crucial to its success. This might include filtering of information etc.
The right side of the picture illustrates the downscaling stack.