A Tale of Two Open Data Sets
By @edent
· Open Data · 7 comments · 700 words · read ~722 times.
It was the best of FoI requests, it was the worst of FoI requests.
While some data holders are wise, others are foolish. Some strengthen our belief in the promise of open data, while others leave us reeling with incredulity. I hereby present a tale of light and darkness, which will bring you equal measures of hope and despair.
Earlier this year, I decided to embark on an open-data-quest. I wanted to find all the fire hydrants in my local area and map them. Where where the areas of poor coverage? Should I buy a house on this road or that? Would insurance companies adjust premiums based on the presence of hydrants?
I could just wander around the streets of Oxford and mark every one I came across on a map - but that is a tiresome and error-prone process. Instead, I turned to "What Do They Know" and submitted a Freedom of Information request to my local Fire and Rescue service.
A month later and I had the data! I was ecstatic! A beautiful .CSV filled with locations and descriptions. I crafted interactive maps, mashed the data up with other sources, brilliant!
My good friend Andy Mabbett thought he'd try to do the same thing with a different local Fire & Rescue service. And that's where the trouble started.
Derbyshire Fire and Rescue Service initially refused to supply the information on the grounds that "disclosure of the location of many or all hydrants in the County would prejudice the commercial interests of the water companies and Derbyshire Fire & Rescue Service."
That is, to put it politely, bunkum. All hydrants are prominently displayed in public - so what's can be commercially sensitive about them? They also tried the old "national security" canard:
The water network is part of the critical national infrastructure and we are only provided access to the hydrant information to fulfil our statutory role of fire-fighting. ... The water network could be perceived to be a potential target. It is part of the critical national infrastructure and Derbyshire Fire & Rescue Service is only provided access to the hydrant information to carry out its statutory role of fire-fighting. Furthermore, the information is received as part of an agreement with water utility companies and we are not permitted to release the information.
Andy appealed and, after intervention from the ICO, the data were released.
the Centre for the Protection of National Infrastructure (CPNI) are of the view that the national security argument to claim the exemption is not strong enough at this time in terms of level of risk and impact on the Critical National Infrastructure.
Hurrah!
Emboldened with his success, Andy tried his luck with the West Midlands Fire Service. An identical request was made - and it was refused. Why? National Security
Yes, that's right, the terrorists might disrupt hydrants in the West Midlands (which they can easily see) or they might commercially compete with the local water company. Errr... what?
Andy is in the middle of an appeal to the ICO. It's hard to see how the West Midlands can refuse this request when others are releasing the data.
And that's part of the problem with Open Data in the UK. Too many public services are outsourced to commercial providers who have no incentive to open up the data for the good of the country. The pervasive fear that someone, somewhere might behave maliciously acts as a massive brake on the progress of open data. Public authorities have to recognise that these data can provide massive public benefits with little real chance of serious harm.
We also have to contend with the fragmented nature of decision making in the UK which leaves us in a confusing and contradictory state. Perhaps, ICO decisions in favour of releasing data need to be binding on all related public bodies?
For now, sadly, we have to make the same arguments time and again in order to promote open data. Onwards!
I agree it's nonsense. However, it seems that there has been a tightening up of security around drinking water in the last year or two i have bumped into a couple of cases of strengthening protection around treated water towers - lots of v obvious razor wire etc
Thank you, Terence, for writing this up. Your post prompted me to chase the ICO, and the allocated case officer has just replied:
I'll keep you posted.
Seems I failed miserably at keeping you posted. Still, better late than never...
The ICO sided with WMFS. I took the matter to a tribunal, and the judges agreed with the ICO's assessment:
G'day. Enthusiast open data advocate in Australia here. IMHO, security issues are worth exploring, but they usually end up as "Yes, a terrorist could do something nasty to the water supply. But having a detailed map of the network and hydrant locations doesn't make that significantly more likely". If anything, it could in some cases do the reverse, by making it easier for a wide range of people to access the information and possible take some kind of preventative measures.
An example of one of our councils' water assets: http://nationalmap.gov.au/#share=lY9zXc
I see that the ICO has reached a decision (saying No) https://ico.org.uk/media/action-weve-taken/decision-notices/2015/1560234/fs_50585724.pdf
I can do you one better. In Dec 2019, just before COVID, the Information Tribunal declared my FOI requests on commercial vacancy data could cause terrorism, but only in Westminster. The other 300 authorities are safe at openlocal.uk. (cf informationrights.decisions.tribunals.gov.uk/DBFiles/Deci...