I think your points are all good. "a targeted analysis would only be conducted by a motivated adversary" For the types of conversations we envision, it's hard

I think your points are all good. "a targeted analysis would only be conducted by a motivated adversary" For the types of conversations we envision, it's hard to see that motivation being very high. It's not likely to reveal much to know who it is that's trying to lose 20 pounds, improve their A1C levels, deal with a problem student as a teacher, or cope with a cheating spouse. These are things we're comfortable sharing with a select group, but we wouldn't necessarily want the world to know we're dealing with. With regard to what we do aim to do, here's a bit of insight into how we look at user privacy (not intending that these alone fully overcome each of the bars you've set, but a summary) We intentionally don't do the following: -We don't use Google Analytics or any big data analytics -We don't use Google, FB, or Twitter sign in -We don't use FB thumbs up or other tracking mechanisms -We don't allow crawling of our database or searching Rushline without being signed in (harder to get large samples for text analysis or re-identification, among other things) We made these decisions in an effort to protect users conversations from being re-identified and combined with the database in the sky. People should be able to talk and read about their personal interests with trusted friends and others, without someone else keeping track of it and adding to their master profile. That's getting much harder everyday, but to the extent we can protect our users from it, we will continue to try our best. Regarding the experience filter, your concern about creating a monoculture makes sense logically. Unfortunately what we learned at ITtoolbox is that when the more experienced people can't protect themselves from certain types of posts, they simply leave the community, taking their insights with them. A far worse outcome. The alternatives to that tend to come in the form of communities with heavy self-policing. There it's novices who feel anxiety and often avoid posting for fear of being yelled at for asking what someone thinks was a poor question. The solution we've created allows experts to easily adjust their own signal to noise threshold. While still not perfect, we think it's a pretty good solution. Newbies and novices get the benefit of seeing all conversation, including that of the experts, while experts can choose to dial in and participate anywhere within the structure they want. Even if they stay mostly up at the top layer of the wedding cake (as we look at it), they're still there and sharing. This is one of the features we felt was important enough to include in our patents pending.