Liveblog of UK BlueLightCamp Unconference session on public sector monitoring of social media. Probe to errors, uncertainly and howling cries against grammar.
The search and rescue teams in the UK are hooked into the 999 response system, but in a chaotic, unstructured way. It doesn’t scale and causes problems during big incidents.
Ideally we want a platform where all the involved parties can share a common platform – so they can share situational awareness. They can’t put a solution together in a silo-based structure.
SARCALL has been developed by volunteers, and is in use with a significant number of SAR teams and associated cat 1 responders. It was built around the Jesip model. It’s objective is to improve communications and mobilisation.
So, the call comes in. The 999 team have to select the mountain rescue team. They use Bing maps to help the dispatchers identify which SAR team to call. A call-out form captures the data that needs to be sent out. That’s then sent by SMS to s spectrum of endpoints, backed up by e-mails.
For example, a welsh call will be sent to the local SAR team – and mirrored to the RAF base ion Anglesey, so they can assist. An incident log is created, and information begins to be sent out to partner teams. As time passes, and more parties become involved, they’re all reading from and adding to the log.
It breaks down silo walls and allows people to talk rapidly using common language during the incident. It extends from the first call, right through to the hospital that a casualty will be directed to.
14 police forces are using it, with 67 SAR teams. It’s handles over 30,000 incidents already.
Is it open source? No. It’s operational code, and he’s very cautious about given general access to it.
GPS isn’t part of this, because they run GPS and SARCALL in separate organisations.
CrisisControl – a prototype of a system that’s map-based for managing very small scale incidents. Is there a potential link between them?
How does the logging occur? Typically, there’s a forward control point – and then the team base back from there. They expect a layer of transcription into a log. Where internet connectivity isn’t available, they can use SMS.
One of their projects is adding Shoothill’s flood alerts into the system.
Auditing? The can pull back through the data. What they’ve tried to do is avoid interacting direct with police command and control systems.
The voice (and hands) of the public
Where is the voice of the public in this? Couldn’t they add useful information. Ben Proctor gives the example of flood-closed roads that could be reopened now that the waters have gone down. Can we give people the ability to report that sort of thing. There’s work been done on social media confidence monitoring on things like flooding – they can watch for multiple mentions of the same incident, or incidents that geographically aren’t possible where the tweet is coming from. We’re good at broadcasting to the public using social media, but not gathering operational information from them.
This is where VOST could come in – trained volunteers not just scanning operational reports and doing situation reports, but also reaching out to communities for more information. Lots of opportunity for matching needs and offered help here. There’s loads of untapped expertise there – retired people and people with relevant specialities, for example.
Could it be formalised like mountain rescue? A big group of volunteers who train, commit and then can be summoned when needed. It also bring the potential of 24/7 converge, walking with international volunteers.
Could we use drones to gather more data? Yes, but not in bad weather. Internet of things sensors? Yup – that’s being explored.