-
Notifications
You must be signed in to change notification settings - Fork 19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Suggested principles: user data freedom portability #5
Comments
It's definitely within the existing framework of data protection (for example, the EU GDPR), but not quite sure the ethics principles document in the current form is the right place for portability considerations. |
I don't want to draw out the debate too long, but I think it's part of the ethical web for two reasons. First, if you can get the complete set of data from a provider then you get a better idea of how detailed their user tracking data is. Is it at the level of "we know you read articles on our site about classic cars" or at the level of "we know how fast you type, that you're left handed, and how many milliseconds you scan through ads for gaming laptops before clicking through". Second, I would consider lock-in to a service as an unethical business practice, and if your data is portable then it's less likely. For example if I use a photo hosting service and they have an option to export my entire content archive at once, that makes it far easier to switch services. If I have to scroll through three thousand photos I uploaded and download each one individually, it's not a hard lock but switching services goes from a single hour effort to a week (edit: a week of full time effort). To repeat, I don't want to drag out the discussion unduly. If you disagree, please just close the issue. Thanks for your time. |
I don't see the first of those points as a part of user data portability, and I suspect many others also would not. That is, if a site offers to export a user's data, and a user gets all the data that they know/expect the site has about them, I think many would see that as a successful data export, even if the site had other tracking data on them. That is, I think user ability to understand tracking is a separate principle from data portability. |
I would consider tracking data as a portion of the user's data. I may not want to import it into another service, but the information they collect about me for advertising purposes should be just as much my data to own as the kitten videos I upload. |
@torgo and I have just discussed this, and it's a tricky one. On the one hand, we see what you mean, that it can be empowering to users to understand how their data is being used (like by advertising trackers) and to be able to take their data elsewhere (like from one photo sharing site to another). At the most extreme, some users want to build their own CMSs, or content-sharing platforms, like the federated social web and indieweb communities are doing. On the other hand, we know that many users want one company to create a smooth UX for them and to handle the hosting of their data. So what's the right answer? As @lknik is saying, this is something that governments are grappling with. I'm not sure we yet know what the right answer is across the world, ethically/behaviourally, and therefore what the implications are for any new web technology. So our inclination is to keep watching this, and see as it develops over the next few years. |
I'm a federated social web and indieweb fan, so this feature is desirable to me. And I think complete, legally mandated access to the complete and often staggering amount of data a vendor collects about each user will encourage more users towards the federated social web and indieweb. On the other hand, I accept the possibility that adding this requirement to the ethical web principles will prevent any business from endorsing them. An ideal set of standards nobody outside the tiny indieweb adopts has far less value than a pretty good set of standards that become universal. |
Just picking this up at our Tokyo f2f where we are planning to publish an update. We don't want to close this issue off however, as indicated above, we'd like some more time to consider how to work this into the document before adding it. We will keep this issue open and consider this for a future update. To be clear, we're not optimising for business endorsement. |
To me, this highlights a pretty fundamental schism in the community, and I think the TAG has discussed it before. Data formats allow people to see, manipulate and reuse their data as they wish. Data APIs put control in the hands of those who serve the API -- typically, not the user. The Web has veered steadily towards APIs in the last ~decade, and away from establishing common formats (work on linked data notwithstanding; it hasn't captured the imagination or implementation of many). Of course, there are many places where an API is not only the most practical answer, but it's also not manipulating data that would be reasonable to extract into a format, but it would be good if this consideration was part of TAG review (even if getting developers who are deadset on publishing an API to back away is... difficult). |
I suspect that ethical arguments about data portability as a fundamental right will not be consistent by region. There are times when there is a legitimate national security, legal or financial interest in exploiting regional differences in legal rights...
Probably there could be some language developed about discouraging "rights / responsibility avoidance".... this language would apply to API operators, not data formats... |
Revisited and discussed today. While we agree with the motivations for wanting data portability, we think this is too low level a point for this particular document. Data portability may certainly be one way to help towards the "enhance individuals' control and power" principle and the "must not cause harm to society" principle, but as a general concept does not mean the same thing to all groups. Also it is not the only thing underlying these principles, and not necessarily sufficient by itself, but we are keeping each principle as concise as possible and prefer not to go into too much depth about specific aspects here. We note that "individuals' control and power" does mention enabling DIY developers and minimizing single points of control, which are related concepts which implicitly support some notions of data portability that have come up in this thread. |
As part of the stated "human rights, dignity and personal agency", I respectfully suggest adding the principle of user data freedom and portability. The specific wording is a complex question, but the rough idea is that users should be able to download the complete set of data that a provider has on them. The user should also have the freedom to use that data in any way they see fit.
The text was updated successfully, but these errors were encountered: