4 ms·

From a privacy perspective, how is choosing ads different from doing syntax highlighting and blame? These things all require the service provider to have access

by arebop 15y ago

From a privacy perspective, how is choosing ads different from doing syntax highlighting and blame? These things all require the service provider to have access to the data you give them. Maybe it is because when ad-picking is done well, it benefits not only you but also some advertiser; perhaps that makes ads seem to serve someone else's interests instead of yours+someone else's interests. There are other worries expressed in this post. I think one of them is that Google might make wrong inferences and therefore do a bad job of choosing ads (matching them to some one-off project you did for a client on a topic that doesn't interest you). I guess what makes this bad is the distraction caused by an irrelevant ad. Syntax highlighting, when done badly, would also be distracting, but I suppose it is much easier to get syntax highlighting right. Is it really the prospect of uninteresting ads that is the main concern?

raganwald 15y ago
No, the prospect of uninteresting ads is not my concern. AFAIK, Github does not build a dossier on me, using syntax hilighting to build a behavioural model of me and connect it with other people I may know. They don't integrate my syntax hilighting with my search term history. If they did, I would have the same concerns. This isn't about ad relevance, it's about my expectation that the things explicitly marked private (fence 1) in services that do not belong to Google (fence 2) stay out of the dossier Google has on me. Quite frankly, casting this as being about the relvance of ads means capitulating 100% to Google's proposition that if it makes me more likely to click an ad, it's worth giving them my privacy. To me, this is like arguing that if the police can make the streets safer, it's worth giving them the power to arrest and detain people without trial. It all comes down to the old, old argument about trading a little liberty for security. Privacy is a kind of liberty. If I won't trade it for security, I sure as hell won't trade it for figuring out that when I say "Ibis," I mean the bike and not the bird.

by arebop 15y ago

raganwald 15y ago

No, the prospect of uninteresting ads is not my concern. AFAIK, Github does not build a dossier on me, using syntax hilighting to build a behavioural model of me and connect it with other people I may know. They don't integrate my syntax hilighting with my search term history. If they did, I would have the same concerns. This isn't about ad relevance, it's about my expectation that the things explicitly marked private (fence 1) in services that do not belong to Google (fence 2) stay out of the dossier Google has on me. Quite frankly, casting this as being about the relvance of ads means capitulating 100% to Google's proposition that if it makes me more likely to click an ad, it's worth giving them my privacy. To me, this is like arguing that if the police can make the streets safer, it's worth giving them the power to arrest and detain people without trial. It all comes down to the old, old argument about trading a little liberty for security. Privacy is a kind of liberty. If I won't trade it for security, I sure as hell won't trade it for figuring out that when I say "Ibis," I mean the bike and not the bird.