This reminds me of the debate on Path's address book upload activity. I don't think it was malicious. It could be covered in a TOS update. But I rarely read the TOS for a site. When I do, it's a serious investment of time to understand what they're trying to say.
I think we need to simplify TOS language and standardize on a set of images to represent privacy, affiliation, and payment terms for a service. I would actually read an at-a-glance TOS summary.
What if you could parse a website's TOS and generate the summary? Startup idea :)
A bit off topic but I think this is nothing link the Path address book issue. I think a lot of the privacy debates in the last few months have been blown out of proportion but when I saw the Path news I immediately closed my account. Path storing details of all my contacts on their servers without my permission is a huge abuse of my trust. This doesn't seem anything like the Path debate in my opinion.
I do agree that TOS language should be simplified. Something like a simple bulleted list explaining each point of the TOS simply.
I don't think it's a particularly viable business, it might be a good feature of a larger business that was focused on helping customers navigate the complexities of modern life; but on it's own, the value in one instance is too small to motivate purchasing.
It's not a viable business, the potential for revenue generation is very small without resorting to questionable tactics (see TRUSTe). I was hoping it would function more as a community effort with the human-readable TOS hosted on the relevant site much like Creative Commons. Sadly I think the human-readable TOS is going to have to come from the business itself (or possibly from an established user-rights organization, i.e. CC or EFF).
As for Path's storage of user contacts; while I find the practice sketchy, especially since I just signed up for Path and had no idea this was happening, but not unexpected especially considering facebook's status as a 'role-model' for privacy issues.
Side-note: Khula Project in its current form is deadpooled. I do plan to relaunch sometime in the future with a project more focused on information distribution.
There's no understood expectation that Pinterest is going to leave any links you upload to them alone - they are responsible for the data just as much as you are.
I find this much more tolerable than Facebook and most search engines' practice of clicking you through an encoded link, which adds latency for no good reason.
In Facebook's case, this is to protect the user from potentially giving away information about themselves in their Referer header and from potentially malicious links that we didn't have enough information on at display (or email) time, but which we now have enough information to know it is malicious.
The reason most do that is to rob the site you're visiting of context about the page you're just on. If I click through from a friend's Facebook profile they can't see the referrer URL is, for example, facebook.com/nickbarnwell
I think we need to simplify TOS language and standardize on a set of images to represent privacy, affiliation, and payment terms for a service. I would actually read an at-a-glance TOS summary.
What if you could parse a website's TOS and generate the summary? Startup idea :)