3 ms·

This is a problem with the internet in general. In the future, we will have an archive of everything everyone did on the internet and can use stylometry, Facebo

This is a problem with the internet in general. In the future, we will have an archive of everything everyone did on the internet and can use stylometry, Facebook likes and clicks, and temporal association to uncover individuals. The only way to opt-out is to just stop participating. All newer cryptography based systems for preserving privacy, like ZCash and Monero, are based around elliptic curve math that will one day be broken. As with 4chan or reddit, the belief that one is anonymous can lead one to engage in riskier or more controversial activity. This is at risk of being uncovered in the future, with potential for damage to the individual.

rothbardrand 9y ago
How soon is that day when elliptic curve cryptography will be broken? AFAIK RSA is still mostly unbroken and it has been decades. EC could last 50 years or 500 years, right? And if one were to, say, take some transparent ZEC, send it to an invisible location, then take it out to a transparent location and do the same again to a third transparent location (to then convert to bitcoin or whatever)... EC would have to be totally completely broken to figure out the private keys of all three transparent addresses (not just weak in the sense of $1B of GPUs can reverse find one key in 6 months). And still you don't know who those addresses belong to (and haven't broken the invisible part to know where the coins that went to the second and third address came from), right?
- louithethrid 9y ago
  Actually, i think the walled garden internet will collapse spectacularly, as any mono-culture has done before- beeing prone to viruses and attacks. Linux and Open Source might be the users last choice, but might be the last man standing one day- if for example sb would write a maschine learning virus able to detect and exploit new weaknesses in a dormant form.
- basketbasket 9y ago
  10 years, at most. Google is seeing to production-ready QC within the next 3 years for specific applications. Factoring elliptic curve point multiplications is log(n) with QC, so once the cat is out of the bag decrypting a large dataset should be trivial.

rothbardrand 9y ago

How soon is that day when elliptic curve cryptography will be broken? AFAIK RSA is still mostly unbroken and it has been decades. EC could last 50 years or 500 years, right? And if one were to, say, take some transparent ZEC, send it to an invisible location, then take it out to a transparent location and do the same again to a third transparent location (to then convert to bitcoin or whatever)... EC would have to be totally completely broken to figure out the private keys of all three transparent addresses (not just weak in the sense of $1B of GPUs can reverse find one key in 6 months). And still you don't know who those addresses belong to (and haven't broken the invisible part to know where the coins that went to the second and third address came from), right?

louithethrid 9y ago
Actually, i think the walled garden internet will collapse spectacularly, as any mono-culture has done before- beeing prone to viruses and attacks. Linux and Open Source might be the users last choice, but might be the last man standing one day- if for example sb would write a maschine learning virus able to detect and exploit new weaknesses in a dormant form.
basketbasket 9y ago
10 years, at most. Google is seeing to production-ready QC within the next 3 years for specific applications. Factoring elliptic curve point multiplications is log(n) with QC, so once the cat is out of the bag decrypting a large dataset should be trivial.