Open data is a force for good, but not without risks

[Heather Morrison-2]

Interesting article in The Guardian:
http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-good-risks

Heather G. Morrison

[Gerry Tychon]

In this article it is stated that:

"Data is all very well, but the ability to extract meaningful
information from it requires considerable skill. Without it, there is
a real danger incorrect conclusions may be drawn."

I have seen this statement in various forms -- mainly for data hugging
purposes. It seem almost anecdotal. I wonder if anybody has any good
examples illustrating this point.

... gerry tychon


On Tue, Jul 10, 2012 at 3:04 PM, Heather Morrison <[hidden email]> wrote:
> Interesting article in The Guardian:
> http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-good-risks
>
> Heather G. Morrison
> _______________________________________________
> CivicAccess-discuss mailing list
> [hidden email]
> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

[Glen Newton]

"Data is all very well, but the ability to extract meaningful
information from it requires considerable skill. Without it, there is
a real danger incorrect conclusions may be drawn."

You can say this about any and all data. It is an invalid argument.
 Or, conversely, it can be used to say that no data should ever be released.

Let incorrect conclusions be made, and be corrected by those making
more correct conclusions.
It is better than having no data released, and _very_ incorrect
conclusions made from speculation or poorer sources of data.

-Glen


On Tue, Jul 10, 2012 at 6:10 PM, Gerry Tychon <[hidden email]> wrote:

> In this article it is stated that:
>
> "Data is all very well, but the ability to extract meaningful
> information from it requires considerable skill. Without it, there is
> a real danger incorrect conclusions may be drawn."
>
> I have seen this statement in various forms -- mainly for data hugging
> purposes. It seem almost anecdotal. I wonder if anybody has any good
> examples illustrating this point.
>
> ... gerry tychon
>
>
> On Tue, Jul 10, 2012 at 3:04 PM, Heather Morrison <[hidden email]> wrote:
>> Interesting article in The Guardian:
>> http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-good-risks
>>
>> Heather G. Morrison
>> _______________________________________________
>> 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/
-

[Heather Morrison-2]

This article raises many interesting issues, and let's not overlook the conclusion that "No technology is without concomitant risks, but provided we tread carefully, with an awareness of the problems, the open data initiative holds immense promise for a better society."

One of the interesting issues raised is that of gaming the system. With respect to crime data, Hand says: "A survey by Direct Line Insurance...found that 11% of respondents claim to have seen but not reported an incident because they feared it would make it more difficult to rent or sell their house.

This is not meant to be an argument against open data, rather an indication of something we (or at least some of us) should think about as we open up more data.

Interesting article in The Guardian:
http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-good-risks

[Gerry Tychon]

When I read the statement

"A survey by Direct Line Insurance...found that 11% of respondents
claim to have seen but not reported an incident because they feared it
would make it more difficult to rent or sell their house."

I said to myself, "so what". It didn't really mean anything
substantial to me. For me, it was a line thrown in to give the aura of
meaning.

... gerry tychon



On Tue, Jul 10, 2012 at 4:38 PM, Heather Morrison <[hidden email]> wrote:

> This article raises many interesting issues, and let's not overlook the conclusion that "No technology is without concomitant risks, but provided we tread carefully, with an awareness of the problems, the open data initiative holds immense promise for a better society."
>
> One of the interesting issues raised is that of gaming the system. With respect to crime data, Hand says: "A survey by Direct Line Insurance...found that 11% of respondents claim to have seen but not reported an incident because they feared it would make it more difficult to rent or sell their house.
>
> This is not meant to be an argument against open data, rather an indication of something we (or at least some of us) should think about as we open up more data.
>
> Interesting article in The Guardian:
> http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-good-risks
>
>
> _______________________________________________
> CivicAccess-discuss mailing list
> [hidden email]
> http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

[Stéphane Guidoin]

"So What" answer: Insurance tend to add various criteria to compute
primes for housing (like, for medical insurance, they ask you how many
times a day you scratch your nose since it can be an indicator that you
are about to have a nose cancer if your nose itches too much). I lost
the reference, but I read recently the crime data released in some
cities (san francisco?) had some "negative" effect: people were less
willing to move in those neighborhoods. Consequence: those neighborhood
which are usually already poor can become poorer.

You have some people who are looking at walkscore.com as an indicator to
where they want to settle, maybe you will have something equivalent with
a crimescore or something like that.

So I agree with those whose say that these negative effects should not
be overlooked... but it should not be an argument against open data. It
should be an argument for enhanced engagement with the community. Data
or not, people know more or less dangerous boroughs and tend to avoid
them anyway. But obviously, local communities can to take actions when
they have data and push their gov to act.

Besides some very specific initiatives, governments (mainly local ones)
still have to understand how to change how they intereact with communities..
Steph

Le 12-07-10 18:47, Gerry Tychon a écrit :

> When I read the statement
>
> "A survey by Direct Line Insurance...found that 11% of respondents
> claim to have seen but not reported an incident because they feared it
> would make it more difficult to rent or sell their house."
>
> I said to myself, "so what". It didn't really mean anything
> substantial to me. For me, it was a line thrown in to give the aura of
> meaning.
>
> ... gerry tychon
>
>
>
> On Tue, Jul 10, 2012 at 4:38 PM, Heather Morrison <[hidden email]> wrote:
>> This article raises many interesting issues, and let's not overlook the conclusion that "No technology is without concomitant risks, but provided we tread carefully, with an awareness of the problems, the open data initiative holds immense promise for a better society."
>>
>> One of the interesting issues raised is that of gaming the system. With respect to crime data, Hand says: "A survey by Direct Line Insurance...found that 11% of respondents claim to have seen but not reported an incident because they feared it would make it more difficult to rent or sell their house.
>>
>> This is not meant to be an argument against open data, rather an indication of something we (or at least some of us) should think about as we open up more data.
>>
>> Interesting article in The Guardian:
>> http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-good-risks
>>
>>
>> _______________________________________________
>> 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
>

[Karl Dubost]

Le 11 juil. 2012 à 09:46, Stéphane Guidoin a écrit :
> So I agree with those whose say that these negative effects should not be overlooked.

s/negative//

Open data have effects. New data shape our actions, understanding, and basically change the world around us. They might have negative effects or positive effects.

In the end, what it creates is a redefinition of the social contract. Some open data have nasty effects on "privacy" (I prefer the words "intimate opacity"). As everything there is nothing good or bad in open data, but they definitely trigger effects.

--
Karl Dubost
Montréal, QC, Canada
http://www.la-grange.net/karl/

[Stéphane Guidoin]

Yes, but try to tell that to the politicians and bureaucrats who decide
open data policies: "they have negative effects or positive effects". Be
sure that nothing will change.

So we (advocates, gov people, etc.) have to assess "effects" and see how
it fits within a sort of (loosely defined) social contract. When it does
not fit we shoud see how to improve the situation.

Stéphane

Le 12-07-11 10:00, Karl Dubost a écrit :
> Le 11 juil. 2012 à 09:46, Stéphane Guidoin a écrit :
>> So I agree with those whose say that these negative effects should not be overlooked.
> s/negative//
>
> Open data have effects. New data shape our actions, understanding, and basically change the world around us. They might have negative effects or positive effects.
>
> In the end, what it creates is a redefinition of the social contract. Some open data have nasty effects on "privacy" (I prefer the words "intimate opacity"). As everything there is nothing good or bad in open data, but they definitely trigger effects.
>

[Karl Dubost]

Le 11 juil. 2012 à 10:21, Stéphane Guidoin a écrit :
> Yes, but try to tell that to the politicians and bureaucrats who decide open data policies: "they have negative effects or positive effects". Be sure that nothing will change.

but I don't put them outside of the equation. They are part of it.
And a politician being against opening data for any reasons doesn't make open data people "right" (whatever right means).

> have to assess "effects"

yes, it really depends on the circumstances.


--
Karl Dubost
Montréal, QC, Canada
http://www.la-grange.net/karl/

[Heather Morrison-2]

Stephane Guidon wrote:

You have some people who are looking at walkscore.com as an indicator to where they want to settle, maybe you will have something equivalent with a crimescore or something like that.

Comment: I love walkscore! Great to see that my city, Vancouver (no surprise) is a Walker's Paradise, with excellent transit. Like the real-time health inspection data I heard about at a presentation by David Eaves, it is really easy to see the positives of a service like this.  The data per se do not present any negatives.

The mere existence of walkscore.com and any popularity it achieves are basically a positive - illustrating that this is something that people want, so that politicians and developers can take this into account.

However, let's consider two scenarios:

Liberal / individual scenario:  walkscore.com is used exclusively by people who are affluent and well-educated to decide where to settle. This is a good service for the individual, and it is likely to drive development of walkable communities. However, this could also result in increased costs for walkable neighbourhoods, making this walkability less accessible for those who are less wealthy.  

Society based scenario: in addition to the lookup for individuals, collective solutions are used: for example, local governments develop walkability standards for all areas of cities and towns that new or renovated developments must meet.

A thought: simply developing walkscore.com is an awesome development to be supported and applauded. I would argue that the optimum use would be the society based scenario. To get here, we need to support creative applications of data such as walkscore.com which provide a platform and show the possibility. To move to the optimum scenario takes collective action, such as advocacy for walkability, community and government discussion / consultation and implementation. To say that this is important in no way suggests that the first step of developing applications is not absolutely fundamental.

One reason why open data advocates might want to hear about these kinds of scenarios is that this may be a more compelling reason for opening up data than the individual scenario. Consider also that walkability has the potential to increase overall citizen health, reduce air pollution / help us meet environmental targets and reduce the need for high-cost transportation infrastructure. To me, that's a lot of really tough problems that this kind of service can help us out with, and a much more compelling reason to move to open data than the ability of the individual to look up walkability for where they want to live, good though that is.

Great discussion for a hot summer day!

Heather Morrison

[Jean-Francois Messier]

A

Le 11 juil. 2012 14:24, "Heather Morrison" <[hidden email]> a écrit :

Stephane Guidon wrote:

You have some people who are looking at walkscore.com as an indicator to where they want to settle, maybe you will have something equivalent with a crimescore or something like that.

Comment: I love walkscore! Great to see that my city, Vancouver (no surprise) is a Walker's Paradise, with excellent transit. Like the real-time health inspection data I heard about at a presentation by David Eaves, it is really easy to see the positives of a service like this.  The data per se do not present any negatives.

The mere existence of walkscore.com and any popularity it achieves are basically a positive - illustrating that this is something that people want, so that politicians and developers can take this into account.

However, let's consider two scenarios:

Liberal / individual scenario:  walkscore.com is used exclusively by people who are affluent and well-educated to decide where to settle. This is a good service for the individual, and it is likely to drive development of walkable communities. However, this could also result in increased costs for walkable neighbourhoods, making this walkability less accessible for those who are less wealthy.

Society based scenario: in addition to the lookup for individuals, collective solutions are used: for example, local governments develop walkability standards for all areas of cities and towns that new or renovated developments must meet.

A thought: simply developing walkscore.com is an awesome development to be supported and applauded. I would argue that the optimum use would be the society based scenario. To get here, we need to support creative applications of data such as walkscore.com which provide a platform and show the possibility. To move to the optimum scenario takes collective action, such as advocacy for walkability, community and government discussion / consultation and implementation. To say that this is important in no way suggests that the first step of developing applications is not absolutely fundamental.

One reason why open data advocates might want to hear about these kinds of scenarios is that this may be a more compelling reason for opening up data than the individual scenario. Consider also that walkability has the potential to increase overall citizen health, reduce air pollution / help us meet environmental targets and reduce the need for high-cost transportation infrastructure. To me, that's a lot of really tough problems that this kind of service can help us out with, and a much more compelling reason to move to open data than the ability of the individual to look up walkability for where they want to live, good though that is.

Great discussion for a hot summer day!

Heather Morrison




_______________________________________________
CivicAccess-discuss mailing list
[hidden email]
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss

[michael gurstein]

+1

M

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Stéphane
Guidoin
Sent: Wednesday, July 11, 2012 6:47 AM
To: [hidden email]
Subject: Re: [CivicAccess-discuss] Open data is a force for good, but not
without risks



"So What" answer: Insurance tend to add various criteria to compute
primes for housing (like, for medical insurance, they ask you how many
times a day you scratch your nose since it can be an indicator that you
are about to have a nose cancer if your nose itches too much). I lost
the reference, but I read recently the crime data released in some
cities (san francisco?) had some "negative" effect: people were less
willing to move in those neighborhoods. Consequence: those neighborhood
which are usually already poor can become poorer.

You have some people who are looking at walkscore.com as an indicator to
where they want to settle, maybe you will have something equivalent with
a crimescore or something like that.

So I agree with those whose say that these negative effects should not
be overlooked... but it should not be an argument against open data. It
should be an argument for enhanced engagement with the community. Data
or not, people know more or less dangerous boroughs and tend to avoid
them anyway. But obviously, local communities can to take actions when
they have data and push their gov to act.

Besides some very specific initiatives, governments (mainly local ones)
still have to understand how to change how they intereact with communities..
Steph

Le 12-07-10 18:47, Gerry Tychon a écrit :

> When I read the statement
>
> "A survey by Direct Line Insurance...found that 11% of respondents
> claim to have seen but not reported an incident because they feared it
> would make it more difficult to rent or sell their house."
>
> I said to myself, "so what". It didn't really mean anything
> substantial to me. For me, it was a line thrown in to give the aura of
> meaning.
>
> ... gerry tychon
>
>
>
> On Tue, Jul 10, 2012 at 4:38 PM, Heather Morrison <[hidden email]>
> wrote:
>> This article raises many interesting issues, and let's not overlook
>> the conclusion that "No technology is without concomitant risks, but
>> provided we tread carefully, with an awareness of the problems, the
>> open data initiative holds immense promise for a better society."
>>
>> One of the interesting issues raised is that of gaming the system.
>> With respect to crime data, Hand says: "A survey by Direct Line
>> Insurance...found that 11% of respondents claim to have seen but not
>> reported an incident because they feared it would make it more
>> difficult to rent or sell their house.
>>
>> This is not meant to be an argument against open data, rather an
>> indication of something we (or at least some of us) should think
>> about as we open up more data.
>>
>> Interesting article in The Guardian:
>> http://www.guardian.co.uk/society/2012/jul/10/open-data-force-for-goo
>> d-risks
>>
>>
>> _______________________________________________
>> 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
>

_______________________________________________
CivicAccess-discuss mailing list [hidden email]
http://lists.pwd.ca/mailman/listinfo/civicaccess-discuss