Road accident fatalities, personal data and data geek wisdom
Locked
Road accident fatalities, personal data and data geek wisdom
 
|
Dear group,
I'd like to map some road accident data for the province of Québec in order to get something like this (http://www.guardian.co.uk/news/datablog/interactive/2011/nov/22/us-road-accident-casualties) + correlates this with other datasets. For this, I have to send a FAI as the data is not open/public. I contacted the agency that holds the data: they will collaborate but won't give me the geospatial info (which remove a lot of interest in the data...). Reason for the refusal: it could allow retrievong some personal data (even though I'm asking for anonymized data). Their argument is that StatCan "blurs" data when the sample is too small to avoid backtracking some personal data (e.g getting the wage of my neighbor because we're only 2 in the given geographical area). However road accident data is less static and I don't see how it could be correlated with any other data to retrieve... I don't know what. All I ask is anything that can be seen by anybody that witness a road accident: vehicules (car, bike, pedestrian), number of people affected, exact place, date/hour. The only thing a little tweeky I'm asking is the severity of the wounds (given by the paramedics) So 2 questions for you : - Do you know any road accident map for Canada or canadian province (until now, I only found statistics) - Do you see any reason to reject the request to get geospatial data? Do you see any breach about the respect of personal data? Thanks Steph |
Re: Road accident fatalities, personal data and data geek wisdom
|
|
Toronto had no problem publishing this data.
As is often the case, QC cites "privacy concerns" that don't exist in the rest of Canada. This is partly because the QC civil code has stronger protections than the common law in the rest of Canada. It's also partly a cultural difference. On 2012-02-28, at 8:49 PM, Stéphane Guidoin wrote:
|
Re: Road accident fatalities, personal data and data geek wisdom
|
|
In reply to this post by Stéphane Guidoin
Hi Stephane,
Wow that privacy argument is pretty thin. Here in BC their is a single insurance company for cars and before the Dec 4th hackathon I got them to give me all accident data from 2006-2010 involving a bicycle. There have been a bunch of people that have mapped it. You can find the data, and the various mashups in Buzzdata here. I know that since then Eric Promislow has asked and received all car accident data, which I believe is also up on Buzzdata. Sadly, this data does not have Lat/Longs, it instead plots the accident to the nearest intersection or mid-block (depending on which is closer) to where the accident took place. But geocoded data is coming, as is an open license for ICBC's data (I believe). There is, as far I can tell, no breach of privacy. Hope this is helpful. Dave On 12-02-28 5:49 PM, Stéphane Guidoin wrote: Dear group, |
Re: Road accident fatalities, personal data and data geek wisdom
|
|
In reply to this post by James McKinney-2
Quebec (like a number of other provinces) has its own privacy laws,
against which any release needs to be evaluated An Act respecting the Protection of personal information in the private sector, RSQ, c P-39.1 http://www.canlii.org/en/qc/laws/stat/rsq-c-p-39.1/latest/rsq-c-p-39.1.html That said, I think there argument is very weak: you can't identify a person based on where they had an accident or the injuries sustained even! -Glen On Tue, Feb 28, 2012 at 9:08 PM, James McKinney <[hidden email]> wrote: > Toronto had no problem publishing this data. > > http://thestar.blogs.com/maps/2009/03/map-of-the-week-bike-accidents.html > > As is often the case, QC cites "privacy concerns" that don't exist in the > rest of Canada. This is partly because the QC civil code has stronger > protections than the common law in the rest of Canada. It's also partly a > cultural difference. > > On 2012-02-28, at 8:49 PM, Stéphane Guidoin wrote: > > Dear group, > > I'd like to map some road accident data for the province of Québec in order > to get something like this > (http://www.guardian.co.uk/news/datablog/interactive/2011/nov/22/us-road-accident-casualties) > + correlates this with other datasets. > > For this, I have to send a FAI as the data is not open/public. I contacted > the agency that holds the data: they will collaborate but won't give me the > geospatial info (which remove a lot of interest in the data...). > > Reason for the refusal: it could allow retrievong some personal data (even > though I'm asking for anonymized data). Their argument is that StatCan > "blurs" data when the sample is too small to avoid backtracking some > personal data (e.g getting the wage of my neighbor because we're only 2 in > the given geographical area). > > However road accident data is less static and I don't see how it could be > correlated with any other data to retrieve... I don't know what. All I ask > is anything that can be seen by anybody that witness a road accident: > vehicules (car, bike, pedestrian), number of people affected, exact place, > date/hour. The only thing a little tweeky I'm asking is the severity of the > wounds (given by the paramedics) > > So 2 questions for you : > - Do you know any road accident map for Canada or canadian province (until > now, I only found statistics) > - Do you see any reason to reject the request to get geospatial data? Do you > see any breach about the respect of personal data? > > Thanks > > Steph > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > > > _______________________________________________ > CivicAccess-discuss mailing list > [hidden email] > http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss -- - http://zzzoot.blogspot.com/ - |
Re: Road accident fatalities, personal data and data geek wisdom
|
|
Here is a good article on Quebec privacy law:
http://quebeclabourlawblog.squarespace.com/blog/2011/1/6/privacy-laws-in-quebec.html -Glen On Wed, Feb 29, 2012 at 8:33 AM, Glen Newton <[hidden email]> wrote: > Quebec (like a number of other provinces) has its own privacy laws, > against which any release needs to be evaluated > > An Act respecting the Protection of personal information in the > private sector, RSQ, c P-39.1 > http://www.canlii.org/en/qc/laws/stat/rsq-c-p-39.1/latest/rsq-c-p-39.1.html > > That said, I think there argument is very weak: you can't identify a > person based on where they had an accident or the injuries sustained > even! > > -Glen > > On Tue, Feb 28, 2012 at 9:08 PM, James McKinney <[hidden email]> wrote: >> Toronto had no problem publishing this data. >> >> http://thestar.blogs.com/maps/2009/03/map-of-the-week-bike-accidents.html >> >> As is often the case, QC cites "privacy concerns" that don't exist in the >> rest of Canada. This is partly because the QC civil code has stronger >> protections than the common law in the rest of Canada. It's also partly a >> cultural difference. >> >> On 2012-02-28, at 8:49 PM, Stéphane Guidoin wrote: >> >> Dear group, >> >> I'd like to map some road accident data for the province of Québec in order >> to get something like this >> (http://www.guardian.co.uk/news/datablog/interactive/2011/nov/22/us-road-accident-casualties) >> + correlates this with other datasets. >> >> For this, I have to send a FAI as the data is not open/public. I contacted >> the agency that holds the data: they will collaborate but won't give me the >> geospatial info (which remove a lot of interest in the data...). >> >> Reason for the refusal: it could allow retrievong some personal data (even >> though I'm asking for anonymized data). Their argument is that StatCan >> "blurs" data when the sample is too small to avoid backtracking some >> personal data (e.g getting the wage of my neighbor because we're only 2 in >> the given geographical area). >> >> However road accident data is less static and I don't see how it could be >> correlated with any other data to retrieve... I don't know what. All I ask >> is anything that can be seen by anybody that witness a road accident: >> vehicules (car, bike, pedestrian), number of people affected, exact place, >> date/hour. The only thing a little tweeky I'm asking is the severity of the >> wounds (given by the paramedics) >> >> So 2 questions for you : >> - Do you know any road accident map for Canada or canadian province (until >> now, I only found statistics) >> - Do you see any reason to reject the request to get geospatial data? Do you >> see any breach about the respect of personal data? >> >> Thanks >> >> Steph >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss >> >> >> >> _______________________________________________ >> CivicAccess-discuss mailing list >> [hidden email] >> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss > > > > -- > - > http://zzzoot.blogspot.com/ > - -- - http://zzzoot.blogspot.com/ - |
Re: Road accident fatalities, personal data and data geek wisdom
|
|
Stéphane;
It is nearly impossible to ever get point data from StatCan, you will only ever get aggregated data. Different jurisdictions are obliged to submit data to StatCan and in return, StatCan promises to maintain privacy in their agreements with them. Procurement agreemends and data sharing agreements between jurisdictions are a big deal, and public officials are overly sensitive to this issue. Remember the hoopla over privacy and the census? Well, StatCan is the most guarded institution on this front and always has been and often annoyingly so and that noncensus summer just made it worse with all the misinformation that was flying around. They are more skiddish than ever.
I hope at some point we will have open data and open government work look into these agreements, MOUs, and procurement contracts as that is where the stuff gets agreed to and StatCan at the end of the day is bound by these. PWGSC is the agency that oversees these at the FED site and lawyers etc. are heavily involved in these contracts.
You may have to contact each provincial ministry of transportation and skip StatCan and you may have to play telephone tag. Further, to get municipal road stuff, you may, yes goodness forbid, need to find a way to get to the locals! This is the no fun part of a federation and division of powers! GeoConnections has data sharing accords with most provinces and might be able to help you out. It is worth it to communcate with them as these accords were put specifically in place for the creation of the national road network among other framework datafiles.
Below are some links for Ontario which you are probably aware of, but just in case:
Ministry of Transport Ontario (or course only provincial roads, not feds and not local!)
This would be too much information, but presumably there is a dbase behind this thing
I am guessing you are an expert about these:
Also try contacting these folks as they often know of other datasources in the prov gov:
Cheers
t
On Wed, Feb 29, 2012 at 8:48 AM, Glen Newton <[hidden email]> wrote: Here is a good article on Quebec privacy law: -- Tracey P. Lauriault 613-234-2805 "Every epoch dreams the one that follows it's the dream form of the future, not its reality" it is the "wish image of the collective". Walter Benjamin, between 1927-1940, (http://www.columbia.edu/itc/architecture/ockman/pdfs/dossier_4/buck-morss.pdf) |
Re: Road accident fatalities, personal data and data geek wisdom
|
|
Thanks for all those wisdom bites :) I'll use the existing examples from Toronto and Vancouver and the link to the laws to demonstrate that resistance is futile...
Stéphane Le 29 février 2012 19:29, Tracey P. Lauriault <[hidden email]> a écrit :
|
«
Return to CivicAccess
|
1 view|%1 views
| Free forum by Nabble | Edit this page |