5 ms·

It's absolutely insane what they are doing (EU). Should libraries remove newspapers with my name if I don't like what's written there about me? It's pure censor

by comboy 9y ago

It's absolutely insane what they are doing (EU). Should libraries remove newspapers with my name if I don't like what's written there about me? It's pure censorship. It all stems from the fact that society hasn't yet adjusted to the reality that not everything on the Internet is true. People share between themselves lots of false beliefs. The Internet is just another medium. The problem is that if I put up a billboard that says John is a thief, they can find me and sue for defamation. If I do a website and use SEO, then it's not so easy, even though it serves similar purpose. But that is how the world works now. We are able to send information to many people, anonymously, often in a targeted way. That's how politics work, that's how ads work, that's how HN works. The problem is not the information, the problem is trust. It's not easy to solve, but it's doable.

abritinthebay 9y ago
You say it’s doable. How?
- comboy 9y ago
  I'm glad you asked :) I want to publish a website with thoughts so far and looking for more brains. What I have in mind is a global, multilayered web of trust. It would be a game changer. Of course, I don't mean it as a startup, it must be decentralized. I don't care who creates it, I just really want to have it. I'm sure you know about the small world effect[1]. Example of layer: coding skill. I assign weights to friends I know. Now if I want to know if X is a good coder I check what friends of my friends think about it and multiply it by weights I have assigned to those friends (recursive). But it's much more than just having reasonable opinion about coding skill of somebody who you want to hire or cooperate with. The crucial thing is that this rating is always relative. It depends on your place in the network. Any global rating can be and will be cheated. Also, it's perfectly fine that for different group of people different things have different values. It enables trustworthy reviews, trustworthy project dependencies, tenants, online stores, doctors, mechanics etc. Could probably help with scams and spam. If successful it has power to change politics inside out. I'm sorry I don't have a write up yet and I know that with an analytic mind you probably can find many problems with the potential implementation. Let's be clear, it's hard. But so very worth it. There are many problems to solve. Apart from bootstrapping it (which could probably be done with help of some other services) The big one to me is privacy. If I assign 2/10 coding skill to one of my friends I don't want him to find out. I've already came up with some potential solutions. But more people, preferably with cryptography background and definitely smarter than me are needed, who are interested in this problem and are willing to spend time thinking about it. I'm going to try to collect all my notes and thoughts into a website and post it on HN. (If you were able to deduce enough from this short description, you like it and you're interested, drop me an empty e-mail at wot@comboy.pl, I'll send you a single e-mail when I have some place to point to) 1. https://en.wikipedia.org/wiki/Small-world_experiment https://en.wikipedia.org/wiki/Small-world_experiment
  - yiyus 9y ago
    > Example of layer: coding skill. I assign weights to friends I know. Now if I want to know if X is a good coder I check what friends of my friends think about it and multiply it by weights I have assigned to those friends (recursive). But it's much more than just having reasonable opinion about coding skill of somebody who you want to hire or cooperate with. So, basically, your multilayered web of trust is just a popularity contest. Sorry, but I do not think this is a good idea.
    - comboy 9y ago
      I think this is one of these "doesn't work" that come to analytical mind when presented in a short form and the reason this idea is not more popular. It would be, if the ranking would be global and not through the lens of your network. You can have the most popular coder A, he does lot's of video tutorials and blogs. But you have your finely cured network regarding this technology because you are pro yourself. You know that according to people you trust, A is just good. B on the other hand, who never worked for his popularity and whose skills are only known to few other pros who worked with him, is excellent. That is what you need web of trust for. That's what you likely use today. It's just without strictly specified protocol, using your brain as a storage and meetings with your friends or conferences as a communication channel. It can be improved and made more efficient.
      - lovich 9y ago
        That still doesn't get rid of the popularity contest aspect. You are making an assumption that humans can make perfectly rational decisions most of the time, much less all of the time. The unfortunate reality is that we all influenced by non rational factors as a result of how our brains evolved
        comboy 9y ago
        Yes, you can't escape that unless you use some AI agent. I don't say I have solution for that. Once again, using it to "rate coders" is just one of use cases, but I'll stick to it for the sake of this argument. So you need a DBA. There's one here that could work for you. You can check data online about him, but you don't know who wrote it. You ask your friends around. Some guy you barely knows says he's a fraud. One of your friends says he worked with him and he's good. You're best friend doesn't know him, but his friend, who he trusts, says he's great. This is how you build your opinion about people (I'm aware it's a simplification but it's not relevant here). The mechanics doesn't change. Just the technology. But it's a leap as big as switching from snail mail to e-mail. How much you trust some information depends on where it comes from. But it doesn't always come directly from the trusted source. So you use your trusted sources to learn about this information source. You also learn to lessen your trust in sources that turned out to lead you to some bad information. It's not always easy to keep it all in your head. Business people learn to master that skill because that's the algorithm that works for acquiring trustworthy information and information is money. However not everybody's live revolves about building network of great contacts and staying in touch with them polling for some useful data.
karmajunkie 9y ago
What if the problem is not that society hasn't adjusted to this new space we call the Internet, but that the people in charge of the internet (or rather, the things on it) haven't figured out yet that they still bear some obligation to the rest of the world? To use your example, whether I put up a website or a billboard with defamatory information is irrelevant. We have laws that protect us, which declare that defamatory speech is not protected. Why does it matter that the speech is online or not when considering the essence of the law? As you stated, if you put up a billboard they can find you and sue you for defamation. Why shouldn't the same be possible online? If I get an injunction against a billboard with defamatory speech the owner of the billboard is required to take it down. Why shouldn't an online provider be equally obligated?
- openasocket 9y ago
  Except, in this case, the article isn't defamatory. It correctly depicts actions of genuine public interest that happened. The only argument to remove them is that they are old. I'd also note that the website itself is still up, containing the article in its entirety.
  - contravariant 9y ago
    It's not entirely clear to me that there's no defamation going on. Does running an add linking to the article for anyone interested in that person's name count as defamation? And if so, why would it be different if google chose included it prominently amongst the search results? I'm not entirely convinced that this should count as defamation, but there is room for debate.
    - openasocket 9y ago
      Information has to be false for it to be defamation, it's in the definition. The news article they are trying to de-list does not contain false information.
      - karmajunkie 9y ago
2ef90u 9y ago
Tech wonk thinks world belongs to tech wonks -- "Only tech can save us from ourselves!" declared would be cult leader "Just Google it!". Film@11
M2Ys4U 9y ago
>It's absolutely insane what they are doing (EU). Should libraries remove newspapers with my name if I don't like what's written there about me? That would be insane if that was how the law works. It's not how the law works.
- djsumdog 9y ago
  But it is. Clearly in the article, both people wanted the record of their crime removed. This is censorship, pure and simple. This is why the EFF opposes the Right to be Forgotten. And the bigger questions: why the fuck is Google is soul source of record for things on the Internet. We need other viable search engines that are no subject to this kind of pressure.
  - andromeduck 9y ago
    I think the analogue would be that the library would have to remove it from the catalog and censor the librarians but still keep it on shelves because the original content isn't directly affected as the action is one step removed.
- qznc 9y ago
  Then german law is nearly insane. In Germany newspaper are not allowed to name people, especially if they are accused but not convicted. The exception is people "of public interest", which is obviously not a clear line. So, quite literally libraries would have to remove newspapers with your name if you don't like what's written there about you ... unless it is of public interest. I consider this law ok.
  - ginko 9y ago
    This seems entirely reasonable to me.

by comboy 9y ago

abritinthebay 9y ago

You say it’s doable. How?

comboy 9y ago
I'm glad you asked :) I want to publish a website with thoughts so far and looking for more brains. What I have in mind is a global, multilayered web of trust. It would be a game changer. Of course, I don't mean it as a startup, it must be decentralized. I don't care who creates it, I just really want to have it. I'm sure you know about the small world effect[1]. Example of layer: coding skill. I assign weights to friends I know. Now if I want to know if X is a good coder I check what friends of my friends think about it and multiply it by weights I have assigned to those friends (recursive). But it's much more than just having reasonable opinion about coding skill of somebody who you want to hire or cooperate with. The crucial thing is that this rating is always relative. It depends on your place in the network. Any global rating can be and will be cheated. Also, it's perfectly fine that for different group of people different things have different values. It enables trustworthy reviews, trustworthy project dependencies, tenants, online stores, doctors, mechanics etc. Could probably help with scams and spam. If successful it has power to change politics inside out. I'm sorry I don't have a write up yet and I know that with an analytic mind you probably can find many problems with the potential implementation. Let's be clear, it's hard. But so very worth it. There are many problems to solve. Apart from bootstrapping it (which could probably be done with help of some other services) The big one to me is privacy. If I assign 2/10 coding skill to one of my friends I don't want him to find out. I've already came up with some potential solutions. But more people, preferably with cryptography background and definitely smarter than me are needed, who are interested in this problem and are willing to spend time thinking about it. I'm going to try to collect all my notes and thoughts into a website and post it on HN. (If you were able to deduce enough from this short description, you like it and you're interested, drop me an empty e-mail at wot@comboy.pl, I'll send you a single e-mail when I have some place to point to) 1. https://en.wikipedia.org/wiki/Small-world_experiment https://en.wikipedia.org/wiki/Small-world_experiment
- yiyus 9y ago
  > Example of layer: coding skill. I assign weights to friends I know. Now if I want to know if X is a good coder I check what friends of my friends think about it and multiply it by weights I have assigned to those friends (recursive). But it's much more than just having reasonable opinion about coding skill of somebody who you want to hire or cooperate with. So, basically, your multilayered web of trust is just a popularity contest. Sorry, but I do not think this is a good idea.
  - comboy 9y ago
    I think this is one of these "doesn't work" that come to analytical mind when presented in a short form and the reason this idea is not more popular. It would be, if the ranking would be global and not through the lens of your network. You can have the most popular coder A, he does lot's of video tutorials and blogs. But you have your finely cured network regarding this technology because you are pro yourself. You know that according to people you trust, A is just good. B on the other hand, who never worked for his popularity and whose skills are only known to few other pros who worked with him, is excellent. That is what you need web of trust for. That's what you likely use today. It's just without strictly specified protocol, using your brain as a storage and meetings with your friends or conferences as a communication channel. It can be improved and made more efficient.
    - lovich 9y ago
      That still doesn't get rid of the popularity contest aspect. You are making an assumption that humans can make perfectly rational decisions most of the time, much less all of the time. The unfortunate reality is that we all influenced by non rational factors as a result of how our brains evolved
      - comboy 9y ago
        Yes, you can't escape that unless you use some AI agent. I don't say I have solution for that. Once again, using it to "rate coders" is just one of use cases, but I'll stick to it for the sake of this argument. So you need a DBA. There's one here that could work for you. You can check data online about him, but you don't know who wrote it. You ask your friends around. Some guy you barely knows says he's a fraud. One of your friends says he worked with him and he's good. You're best friend doesn't know him, but his friend, who he trusts, says he's great. This is how you build your opinion about people (I'm aware it's a simplification but it's not relevant here). The mechanics doesn't change. Just the technology. But it's a leap as big as switching from snail mail to e-mail. How much you trust some information depends on where it comes from. But it doesn't always come directly from the trusted source. So you use your trusted sources to learn about this information source. You also learn to lessen your trust in sources that turned out to lead you to some bad information. It's not always easy to keep it all in your head. Business people learn to master that skill because that's the algorithm that works for acquiring trustworthy information and information is money. However not everybody's live revolves about building network of great contacts and staying in touch with them polling for some useful data.

karmajunkie 9y ago

What if the problem is not that society hasn't adjusted to this new space we call the Internet, but that the people in charge of the internet (or rather, the things on it) haven't figured out yet that they still bear some obligation to the rest of the world? To use your example, whether I put up a website or a billboard with defamatory information is irrelevant. We have laws that protect us, which declare that defamatory speech is not protected. Why does it matter that the speech is online or not when considering the essence of the law? As you stated, if you put up a billboard they can find you and sue you for defamation. Why shouldn't the same be possible online? If I get an injunction against a billboard with defamatory speech the owner of the billboard is required to take it down. Why shouldn't an online provider be equally obligated?

openasocket 9y ago
Except, in this case, the article isn't defamatory. It correctly depicts actions of genuine public interest that happened. The only argument to remove them is that they are old. I'd also note that the website itself is still up, containing the article in its entirety.
- contravariant 9y ago
  It's not entirely clear to me that there's no defamation going on. Does running an add linking to the article for anyone interested in that person's name count as defamation? And if so, why would it be different if google chose included it prominently amongst the search results? I'm not entirely convinced that this should count as defamation, but there is room for debate.
  - openasocket 9y ago
    Information has to be false for it to be defamation, it's in the definition. The news article they are trying to de-list does not contain false information.
    - karmajunkie 9y ago

2ef90u 9y ago

Tech wonk thinks world belongs to tech wonks -- "Only tech can save us from ourselves!" declared would be cult leader "Just Google it!". Film@11

M2Ys4U 9y ago

>It's absolutely insane what they are doing (EU). Should libraries remove newspapers with my name if I don't like what's written there about me? That would be insane if that was how the law works. It's not how the law works.

djsumdog 9y ago
But it is. Clearly in the article, both people wanted the record of their crime removed. This is censorship, pure and simple. This is why the EFF opposes the Right to be Forgotten. And the bigger questions: why the fuck is Google is soul source of record for things on the Internet. We need other viable search engines that are no subject to this kind of pressure.
- andromeduck 9y ago
  I think the analogue would be that the library would have to remove it from the catalog and censor the librarians but still keep it on shelves because the original content isn't directly affected as the action is one step removed.
qznc 9y ago
Then german law is nearly insane. In Germany newspaper are not allowed to name people, especially if they are accused but not convicted. The exception is people "of public interest", which is obviously not a clear line. So, quite literally libraries would have to remove newspapers with your name if you don't like what's written there about you ... unless it is of public interest. I consider this law ok.
- ginko 9y ago
  This seems entirely reasonable to me.