5 ms·

Ask HN: Opinions on a regulation which could solve most privacy converns

2 points by ilovetux 9y ago

I had an idea about how most of my personal privacy concerns could be mitigated.I would like to solicit opinions on the idea. I believe that if use of our personal data (including aggregation, correlation, study and sale) were required to be reported, we could provide real knowlegable consent to the use of our personal data. In other words we would be able to know who is tracking us and why. The current system where we blindly click "I agree" is broken. People do not want to wade through the legalese and they shouldn't have to. If we were notified every time our personal data was used, we could finally know what we were agreeing to.

siegel 9y ago
I'm trying to better understand the proposal: 1) What exactly would be reported? 2) To whom? 3) How would it be reported?
- ilovetux 9y ago
  Basically any time one person's data is collected or used in any way, that person would have the right to be notified of the actual data in question, how that data is being used and by whom. This could be an email containing a csv file with the actual data points and a brief summary of what is happening with the data. This is the ideal and I haven't fleshed out the idea too much which is why I am soliciting opinions and ideas.
  - siegel 9y ago
    I would think that this would first require the adoption of something much closer to the EU data privacy regime: actual rights to know what data is collected, how it would be used prospectively, rights to collect and be forgotten, etc...I think this would be a precondition to the setup you are proposing. It would also seem clear to me that, from a regulatory standpoint, there would need to be a regulatory body in charge of developing regs around this and enforcing it. So, in other words, this would be expensive to implement (both in terms of adding government regulatory costs + costs of private enterprises complying). That being said, what I find particularly attractive about your proposal is that it might shock people into understanding what they are really doing by pressing "I agree." In other words, it could cause a huge cultural shift that I think is really necessary. There are a lot of things written about how people (particularly younger people) do not really value privacy anymore. I would bet a large sum of money that if your proposal were implemented, many of these people who "don't care about privacy" would start caring because the implications of their giving up their personal data would suddenly become clear. I can't see this being all that difficult to implement technologically, would it be?
    - ilovetux 9y ago
      Technically I do not believe so, in python (with which I am most familiar) it is trivial to add a handler to a logger. That handler can do anything including sending emails or http requests. A similar technique should be possible within similar technologies such as log4j (a common framework written in java). What I like about my idea is that nothing need to be changed by the business except implementation of the new requirement. In other words the "I agree" you clicked still applies but now they have to tell you what actually happened with the actual data which could be used to inform future decisions. What I find troubling about your prerequisites is the word "prospectively". Reporting of actual facts should be the goal and this is only possible in the past and present tense. By the way, thank you for your response, I need to research current European laws more thoroughly before I can give a proper response.
      - siegel 9y ago
        I wonder if this could be marketed as a 3rd-party service? Plenty of companies pay for privacy certifications that are not legally required - presumably because there is some amount of customer demand. I could imagine incredible demand for a service like you propose, if people knew that they could get this kind of information.
        ilovetux 9y ago
        That is a great idea! Personally the only organization I would trust at this point would be the Free Software Foundation. They are a nonprofit organization with a long history of supporting user rights. In fact, if everyone were to use the GPL and/or Affero licenses for their software, the source code would be auditable. That would alleviate half of my concerns as I could tell what information would be available to them. All that would be left is my concern about what is done with my personal information after collection. If an organization which I trust were to issue certification which conveys assurances as to what does or doesn't happen to my data, I might be ok with that even without access to specific details. I don't think the big guys would sign up for this without incentive. That's the main reason I think a law would be the best way. In order to make this happen, a boycott could be organized. Participation would need to be easy so we could facilitate it with a custom build of Firefox (or similar) which will only allow you to visit sites certified to respect users privacy.
ilovetux 9y ago
I have put some thought into the possibility of implementing this requirement "client side". In other words I wanted to think through what would be needed to implement this for myself. I think an access point with the ability to proxy requests with tls termination support could be used to decrypt/encrypt all traffic inbound and outbound. This has two problems. First this would not allow visibility into messages where the encrypted payload was itself encrypted. Second this would only be as trustworthy as the hardware, drivers and operating system. In other words, unless you have the ability to build the hardware yourself or knew a trustworthy manufacture you could never be 100% that you are aware of all collected data. This also addresses less than half of the issue. To know when your data is being used the best way would be to catalog, index and correlate all known information about the business in question including all official filings, mergers, acquisitions and any trustworthy news reports or blog posts. This information could be mined for possible disclosures of your information. This algorithm would of course have to be recursive to provide insight after the data leaves the hands of the original collectors. Now comes the fun part, information is gathered about us in multiple ways including tax filings, credit card purchases, facial recognition within photos (which you may only be in the background). All of that leads me to believe that this is not something which can be done alone. What I believe we need is a law which would guarantee us the right to receive notifications when data concerning us has been collected, used or sold. As far as a regularatory body, that would be nice and could be used as a trusted third party to collect all the information which should be reported to individuals, this could lead to a web portal which aggregates all disclosures for us. This however is not necessarily needed in the short-term, I believe if such a law was passed we could rely on whistle-blowers to report transgressions. The important thing would be to establish the right, we can worry about enforcement at a later point.

2 points by ilovetux 9y ago

siegel 9y ago

I'm trying to better understand the proposal: 1) What exactly would be reported? 2) To whom? 3) How would it be reported?

ilovetux 9y ago
Basically any time one person's data is collected or used in any way, that person would have the right to be notified of the actual data in question, how that data is being used and by whom. This could be an email containing a csv file with the actual data points and a brief summary of what is happening with the data. This is the ideal and I haven't fleshed out the idea too much which is why I am soliciting opinions and ideas.
- siegel 9y ago
  I would think that this would first require the adoption of something much closer to the EU data privacy regime: actual rights to know what data is collected, how it would be used prospectively, rights to collect and be forgotten, etc...I think this would be a precondition to the setup you are proposing. It would also seem clear to me that, from a regulatory standpoint, there would need to be a regulatory body in charge of developing regs around this and enforcing it. So, in other words, this would be expensive to implement (both in terms of adding government regulatory costs + costs of private enterprises complying). That being said, what I find particularly attractive about your proposal is that it might shock people into understanding what they are really doing by pressing "I agree." In other words, it could cause a huge cultural shift that I think is really necessary. There are a lot of things written about how people (particularly younger people) do not really value privacy anymore. I would bet a large sum of money that if your proposal were implemented, many of these people who "don't care about privacy" would start caring because the implications of their giving up their personal data would suddenly become clear. I can't see this being all that difficult to implement technologically, would it be?
  - ilovetux 9y ago
    Technically I do not believe so, in python (with which I am most familiar) it is trivial to add a handler to a logger. That handler can do anything including sending emails or http requests. A similar technique should be possible within similar technologies such as log4j (a common framework written in java). What I like about my idea is that nothing need to be changed by the business except implementation of the new requirement. In other words the "I agree" you clicked still applies but now they have to tell you what actually happened with the actual data which could be used to inform future decisions. What I find troubling about your prerequisites is the word "prospectively". Reporting of actual facts should be the goal and this is only possible in the past and present tense. By the way, thank you for your response, I need to research current European laws more thoroughly before I can give a proper response.
    - siegel 9y ago
      I wonder if this could be marketed as a 3rd-party service? Plenty of companies pay for privacy certifications that are not legally required - presumably because there is some amount of customer demand. I could imagine incredible demand for a service like you propose, if people knew that they could get this kind of information.
      - ilovetux 9y ago
        That is a great idea! Personally the only organization I would trust at this point would be the Free Software Foundation. They are a nonprofit organization with a long history of supporting user rights. In fact, if everyone were to use the GPL and/or Affero licenses for their software, the source code would be auditable. That would alleviate half of my concerns as I could tell what information would be available to them. All that would be left is my concern about what is done with my personal information after collection. If an organization which I trust were to issue certification which conveys assurances as to what does or doesn't happen to my data, I might be ok with that even without access to specific details. I don't think the big guys would sign up for this without incentive. That's the main reason I think a law would be the best way. In order to make this happen, a boycott could be organized. Participation would need to be easy so we could facilitate it with a custom build of Firefox (or similar) which will only allow you to visit sites certified to respect users privacy.

ilovetux 9y ago

I have put some thought into the possibility of implementing this requirement "client side". In other words I wanted to think through what would be needed to implement this for myself. I think an access point with the ability to proxy requests with tls termination support could be used to decrypt/encrypt all traffic inbound and outbound. This has two problems. First this would not allow visibility into messages where the encrypted payload was itself encrypted. Second this would only be as trustworthy as the hardware, drivers and operating system. In other words, unless you have the ability to build the hardware yourself or knew a trustworthy manufacture you could never be 100% that you are aware of all collected data. This also addresses less than half of the issue. To know when your data is being used the best way would be to catalog, index and correlate all known information about the business in question including all official filings, mergers, acquisitions and any trustworthy news reports or blog posts. This information could be mined for possible disclosures of your information. This algorithm would of course have to be recursive to provide insight after the data leaves the hands of the original collectors. Now comes the fun part, information is gathered about us in multiple ways including tax filings, credit card purchases, facial recognition within photos (which you may only be in the background). All of that leads me to believe that this is not something which can be done alone. What I believe we need is a law which would guarantee us the right to receive notifications when data concerning us has been collected, used or sold. As far as a regularatory body, that would be nice and could be used as a trusted third party to collect all the information which should be reported to individuals, this could lead to a web portal which aggregates all disclosures for us. This however is not necessarily needed in the short-term, I believe if such a law was passed we could rely on whistle-blowers to report transgressions. The important thing would be to establish the right, we can worry about enforcement at a later point.