4 ms·

I'm defending the idea that they are disinterested in change not because they are ignorant, but because they are fully aware and accepting of what they're givin

by dsacco 10y ago

I'm defending the idea that they are disinterested in change not because they are ignorant, but because they are fully aware and accepting of what they're giving up. I am not arguing that they don't care either way. If they otherwise don't care, but they appreciate the usability and convenience features of these products, then that is an objection to change if they won't get the same features in more open, private alternatives. My second point is this: if enough people can be described as both aware and accepting, then mainstream change will not occur because people like what they have more than they like their privacy, and they make that decision in an informed and responsible manner that does not provide a market opportunity. You can make an alternative, and it can be great (e.g. DuckDuckGo), but it will not be mainstream.

AnthonyMouse 10y ago
At which point you're in the minority because the set of people who affirmatively want to give up their privacy is not larger than the set who either don't want to or don't care. DuckDuckGo is not decentralized, it's just a worse Google with a better privacy policy. Actual decentralized search is hard. But this is not a one-time global binary decision. We can have decentralized YouTube and Facebook even if nobody has yet figured out how to have decentralized Google.
- eropple 10y ago
  I would suggest that having Facebook's userbase is a core part of having Facebook. If you don't have Facebook's userbase on your "decentralized Facebook", you have a strictly inferior replacement to Facebook. The set of people who care enough about the benefits (which do exist!) of decentralization is not and never going to be enough to overcome the downside of having nobody there. That's the problem you have to solve if this is something you really care about. Haranguing people because they're not using it isn't the way. The way to do it is to figure out how to make it attractive by the users' metrics rather than your own.
  - AnthonyMouse 10y ago
    > That's the problem you have to solve if this is something you really care about. We already know that. It's not trivial but there is a clear path. You make something that will become popular for reasons unrelated to the thing it should replace, it just happens to also be able to replace it. Once you have something that everybody already uses for other reasons and can do everything Facebook does, anyone can start using it instead of Facebook without waiting for anybody else. The interesting thing is this bootstraps across the whole space if you can do it once. Given a decentralized social graph you can take over for YouTube as soon as you have video support etc.
    - dsacco 10y ago
      I would say it's not trivial and there isn't a clear path. YouTube is not just a web server for videos. You can't set up a decentralized, P2P real-time streaming network (like webtorrent) to replace YouTube. Many of the features that make YouTube virally successful and a continuing platform are developed based on research produced by its analytics which sacrifice privacy to improve insights into what users care about. YouTube also has a profitable business model, which provides a capitalist incentive to improve the product according to user demand. To make a comparison - which are you betting on in 2016 - IRC or Slack? Even if you set up a successful open source project to replicate YouTube, how will you develop it as well when your best employees are poached because they aren't paid as well as centralized companies? How about the analytics features? How do you design an anonymous or decentralized recommendation engine with the same or similar utility? How do you power search relevance in a meaningfully private and decentralized way? How do you add the social networking features that people care about without a centralized social network? The fact that you can technically make product clones with distributed graphs does not on its own constitute a feasible replacement. The basis of modern product research and improvement is data tracking and analysis. You can argue that it can be curtailed or scaled back to improve some privacy, but we haven't even started talking about the performance constraints of decentralized search, streaming, etc. Where will the primary files (e.g. videos) and supplementary files (e.g. recommender models, analysis features, graph, infrastructure, etc.) be hosted? How will you guarantee similar uptime and performance rates to AWS and GCP with a decentralized network of seeders? You strictly cannot replace products like YouTube, Facebook, Netflix or Google with decentralized counterpoints. The demand and feasibility is not there.
- dsacco 10y ago
  >>At which point you're in the minority because the set of people who affirmatively want to give up their privacy is not larger than the set who either don't want to or don't care. The set of people who affirmatively want to give up their privacy is in the majority if most people don't care about privacy but want privacy reducing features. Sure, in a perfect world they'd take both, but we're talking about which is more important to them when they have to make a decision. You're arguing my second point, about market viability of a privacy-enhancing product. My first point, which supports my second point, is that most people give up their privacy because they understand the compromise and don't value privacy more than convenience-enhancing, privacy-reducing features. I'm not sure if I'm conveying my points well, let me try to rephrase. I don't think my postulated demographic is in the majority if they 1) desire competitive features which reduce privacy, 2) are willing to responsibly reduce their privacy for those features and if the market alternatives (decentralized or otherwise) increase their privacy without competitive feature parity. Basically, if the majority of people are ambivalent about privacy, understand the privacy compromise and still desire privacy-reducing convenience in features, then the alternatives will never become mainstream. This is my hypothesis, and I am pushing this hypothesis by defending the idea that most people give up privacy because they understand but don't value it, not because they are naive or ignorant and value it. The reason I am putting this hypothesis forward is because I think it's important that the HN crowd consider this possibility. It's easy to believe people give up their privacy because they don't have alternatives, or because they don't know any better, but that perspective comes across as condescending to people who fully understand the compromise they agree to, and there is a strong argument that this is the majority of the population. If that's true, you can't expect privacy-enhancing products to become mainstream if privacy is the only competitive feature, or if the privacy precludes competitive features offered by other companies. The overall conclusion, based on these premises, is that many privacy-valuing technies misunderstand the privacy values of the majority of the population. They believe that the aforementioned privacy-enhancing products can be competitive and replace the incumbents if only they could get people to understand the privacy breaches they're suffering. My grand unifying theory here begins with the hypothesis that most users actually understand the privacy compromise, and it's really the vocal HN zeitgeist that misunderstands the awareness of most users in this arena. By extension, they also misunderstand the market oppurtunity, mainstream potential or even feasibility of open alternatives that can change or replace the incumbents. At this juncture I'm not sure how else to convey my point. I feel as though we are probably talking past each other.
type0 10y ago
> not because they are ignorant, but because they are fully aware and accepting of what they're giving up. Do they know what they are giving up?!

by dsacco 10y ago

AnthonyMouse 10y ago

At which point you're in the minority because the set of people who affirmatively want to give up their privacy is not larger than the set who either don't want to or don't care. DuckDuckGo is not decentralized, it's just a worse Google with a better privacy policy. Actual decentralized search is hard. But this is not a one-time global binary decision. We can have decentralized YouTube and Facebook even if nobody has yet figured out how to have decentralized Google.

eropple 10y ago
I would suggest that having Facebook's userbase is a core part of having Facebook. If you don't have Facebook's userbase on your "decentralized Facebook", you have a strictly inferior replacement to Facebook. The set of people who care enough about the benefits (which do exist!) of decentralization is not and never going to be enough to overcome the downside of having nobody there. That's the problem you have to solve if this is something you really care about. Haranguing people because they're not using it isn't the way. The way to do it is to figure out how to make it attractive by the users' metrics rather than your own.
- AnthonyMouse 10y ago
  > That's the problem you have to solve if this is something you really care about. We already know that. It's not trivial but there is a clear path. You make something that will become popular for reasons unrelated to the thing it should replace, it just happens to also be able to replace it. Once you have something that everybody already uses for other reasons and can do everything Facebook does, anyone can start using it instead of Facebook without waiting for anybody else. The interesting thing is this bootstraps across the whole space if you can do it once. Given a decentralized social graph you can take over for YouTube as soon as you have video support etc.
  - dsacco 10y ago
    I would say it's not trivial and there isn't a clear path. YouTube is not just a web server for videos. You can't set up a decentralized, P2P real-time streaming network (like webtorrent) to replace YouTube. Many of the features that make YouTube virally successful and a continuing platform are developed based on research produced by its analytics which sacrifice privacy to improve insights into what users care about. YouTube also has a profitable business model, which provides a capitalist incentive to improve the product according to user demand. To make a comparison - which are you betting on in 2016 - IRC or Slack? Even if you set up a successful open source project to replicate YouTube, how will you develop it as well when your best employees are poached because they aren't paid as well as centralized companies? How about the analytics features? How do you design an anonymous or decentralized recommendation engine with the same or similar utility? How do you power search relevance in a meaningfully private and decentralized way? How do you add the social networking features that people care about without a centralized social network? The fact that you can technically make product clones with distributed graphs does not on its own constitute a feasible replacement. The basis of modern product research and improvement is data tracking and analysis. You can argue that it can be curtailed or scaled back to improve some privacy, but we haven't even started talking about the performance constraints of decentralized search, streaming, etc. Where will the primary files (e.g. videos) and supplementary files (e.g. recommender models, analysis features, graph, infrastructure, etc.) be hosted? How will you guarantee similar uptime and performance rates to AWS and GCP with a decentralized network of seeders? You strictly cannot replace products like YouTube, Facebook, Netflix or Google with decentralized counterpoints. The demand and feasibility is not there.
dsacco 10y ago
>>At which point you're in the minority because the set of people who affirmatively want to give up their privacy is not larger than the set who either don't want to or don't care. The set of people who affirmatively want to give up their privacy is in the majority if most people don't care about privacy but want privacy reducing features. Sure, in a perfect world they'd take both, but we're talking about which is more important to them when they have to make a decision. You're arguing my second point, about market viability of a privacy-enhancing product. My first point, which supports my second point, is that most people give up their privacy because they understand the compromise and don't value privacy more than convenience-enhancing, privacy-reducing features. I'm not sure if I'm conveying my points well, let me try to rephrase. I don't think my postulated demographic is in the majority if they 1) desire competitive features which reduce privacy, 2) are willing to responsibly reduce their privacy for those features and if the market alternatives (decentralized or otherwise) increase their privacy without competitive feature parity. Basically, if the majority of people are ambivalent about privacy, understand the privacy compromise and still desire privacy-reducing convenience in features, then the alternatives will never become mainstream. This is my hypothesis, and I am pushing this hypothesis by defending the idea that most people give up privacy because they understand but don't value it, not because they are naive or ignorant and value it. The reason I am putting this hypothesis forward is because I think it's important that the HN crowd consider this possibility. It's easy to believe people give up their privacy because they don't have alternatives, or because they don't know any better, but that perspective comes across as condescending to people who fully understand the compromise they agree to, and there is a strong argument that this is the majority of the population. If that's true, you can't expect privacy-enhancing products to become mainstream if privacy is the only competitive feature, or if the privacy precludes competitive features offered by other companies. The overall conclusion, based on these premises, is that many privacy-valuing technies misunderstand the privacy values of the majority of the population. They believe that the aforementioned privacy-enhancing products can be competitive and replace the incumbents if only they could get people to understand the privacy breaches they're suffering. My grand unifying theory here begins with the hypothesis that most users actually understand the privacy compromise, and it's really the vocal HN zeitgeist that misunderstands the awareness of most users in this arena. By extension, they also misunderstand the market oppurtunity, mainstream potential or even feasibility of open alternatives that can change or replace the incumbents. At this juncture I'm not sure how else to convey my point. I feel as though we are probably talking past each other.

type0 10y ago

> not because they are ignorant, but because they are fully aware and accepting of what they're giving up. Do they know what they are giving up?!