I've attended a workshop on Adoption of ICT to improve security and trust in the UK food chain. This workshop is sponsored by the IT as a Utility Network+, as part of the Digital Economy Theme of the Research Councils UK. As you might guess from the title, it aims to "bring together scientific experts, policy makers and technology practitioners" to "explore best practice from around the UK and beyond and strive to identify tools and ICT systems that can be harnessed to improve food product security, traceability, nutritional benefits and consumer confidence with the social, health and economic benefits that can arise".
My general opinion is that the problem is vast, involves a large number of agents, and it's not entirely clear how much ICT alone can do to improve the current situation. Identifying fragments of the problem, and coming up with interactions between industry, academia, and possibly government regulators, to focus on these specific fragments, could be a way to make some progress in tacking this problem, in my opinion.
Technology itself isn't where most of the problem lies. We have the means to track e.g. pigs from birth to slaughter, mostly automatically using implanted chips and automatic scanners. We could keep a large database tracking all the elements that come in and leave a food processing plant, potentially enabling the final consumer to access this data by reading a bar code on a product sitting on a supermarket shelf. However, multiple issues prevent this from being implemented. The technology might exist, but it isn't necessarily cheap; who will pay for the added cost of food? The data in the system is only valuable if it is reliable; how can we ensure this, and protect the system from false information? Data sharing is essential for effective tracking, but are the producers willing to share this data? It is likely that such wide-scale tracking would require legal provisions in order to promote its adoption; is there political support for this?
From a computer science and programming languages point of view, my professional recommendation would be to foster the use of open source technologies. As the number of agents involved is very large, any technology developed for tracking, testing, and controlling the food supply chain should be interoperable and easy to manipulate. Large enterprises might be tempted to develop proprietary sensors, systems, and data description languages, but this should be prevented, both by the industrial customers of the technology as by government regulators, as I believe this will lead to interoperability issues, vendor lock-ins, and loss of the advantages of a competitive market.