Re:Should We Allow Bulk Searching of Cloud Archives?
by Amorymeltzer on Saturday January 16, 2016 @03:12PM
(#10916640)
>Fourth, this kind of voluntary search by the likes of Google is not unprecedented. Google scans outgoing Gmail messages to see if there are attachments whose hashes — that is, digital fingerprints — match those of known images of child pornography. When this first became widely known to the public, the company explained: “Sadly all Internet companies have to deal with child sexual abuse … It’s why Google actively removes illegal imagery from our services—including search and Gmail—and immediately reports abuse … This evidence is regularly used to convict criminals.”
>To be sure, child pornography filtering — and reporting — may be a place to draw a clear line. Not only is child pornography near-universally reviled and banned, but the matching algorithm for previously-identified images boasts no false positives, and, perhaps most important, possession of the file is not only clear evidence of the crime, but quite typically the crime itself.
Re:Should We Allow Bulk Searching of Cloud Archives?
by userbinator on Saturday January 16, 2016 @05:45PM
(#10917228)
I accidentally stumbled across this while searching for some obscure IC part numbers. The wording is something like "we have removed results of suspected child abuse" (emphasis mine --- I remember that part the clearest), and that scared me for several reasons; the implication that just being "suspected" is enough to censor, whether this was logged and now my searches are being scrutinised even more, and the reminder of effectively how much power they have over what people can find on the Internet.
I'd use DuckDuckGo if it weren't for the fact that its index is far smaller than Google's, although with the latter's increasingly aggressive pruning of results, maybe that won't remain the case for long...
Re:Should We Allow Bulk Searching of Cloud Archives?
by otakucode on Sunday January 17, 2016 @07:01PM
(#10921722)
Groups in Europe a few years ago were trying to push for video card manufacturers to build the hash database into their drivers, so that no image could be displayed which was not first checked. Never underestimate the amount of resources people are willing to dedicate to "stopping child abuse" as long as it doesn't require doing things actually shown to stop child abuse - like education, empowerment, and providing resources for the relocation of children leaving dangerous environments. None of that stuff gets headlines, or makes it look like a cop is doing a good job like going online, or opens the door to later extension to whatever sort of content groups feel like controlling.
Re:Should We Allow Bulk Searching of Cloud Archives?
by ikeboy on Saturday January 16, 2016 @03:25PM
(#10916677)
>matching algorithm for previously-identified images boasts no false positives
Hm. Wonder what algorithm they use, and if it's one that has collisions, and if so, can someone distribute images that alert Google's system but have no illegal content.
Re:Should We Allow Bulk Searching of Cloud Archives?
by ikeboy on Saturday January 16, 2016 @05:17PM
(#10917124)
Hm, so attempting to break it would require direct access to tests, which means you'd need to be a "qualified organization". Don't see any way of doing this that doesn't end in jail time.
Re:Should We Allow Bulk Searching of Cloud Archives?
by mindslight on Saturday January 16, 2016 @03:31PM
(#10916702)
Fighting use of such capabilities will always be a losing battle, because there will always be a situation that makes the majority clamor for greater powers. The only way to preserve security is to prevent the capability in the first place. This means building technology that preserves personal privacy and autonomy, rather than naive implementations that are easy to monetize.
All of us are culpable, to the degree we contribute cleartext to the butt and encourage others to do so as well, for creating these massive data silos that are gifts to (micro)politicians everywhere.
Re:Should We Allow Bulk Searching of Cloud Archives?
by mikegerwitz on Saturday January 16, 2016 @11:56PM
(#10918342)
I posted this comment on Schneier's blog, but I then saw this thread. So here it is:
This type of risk is one that you must accept if you rely on a Service as a Software Substitute[0].
By relying on a remote service for your data storage and computing, you give up both freedom and control---even if the data are encrypted; we've all seen what information can be gleaned from metadata[1]. In this case, perhaps where a document was distributed: if Eve knows that Alice viewed document X, and Bob has a document Y that is shown (through metadata) to have been sent from Alice, and that document has certain properties that match, then it might be enough to convince a court that Bob contains document X, and get a warrant.
> Companies treat child pornography and abuse in a special category. But the precedent remains, and it could be tempting to extend at least the exact-match kind of search to the Paris terrorist example
Terrorism and child pornography are two examples that are often used to justify broad searches, chilling free speech and violating privacy.[2] These types of privacy/security "trade offs" start down a very slippery slope.
> As so much of everyone’s private communications and work migrates into the hands of a few massive private companies, the net-wide search will become too tempting to leave alone. Exactly what makes it tempting is what makes it troubling
You should reject those services and do your own computing![0] These remote services are not a substitute for locally installed software---a distinction that many do not consider or realize.
Not everyone has the technical capability to handle certain problems on their own; hosting your own e-mail server is a good example. In such cases, users should, at the very least, choose a service that commits to their privacy, to the degree that LavaBit did.
For social media: consider decentralized alternatives, like GNU Social[3], GNU MediaGoblin[4], and many others. Using those services will also encourage your friends and family to ask questions about why you made that choice, and consider switching themselves, since they'll have others they know using those services. Even if you cannot host your own, you can shop around for hosts that you feel that you can trust, and still participate in the same network as your friends and family; there's no lock-in to one service like Twitter or Facebook. Wholesale searches are also made rather difficult by distributed networks.
Forfeiture of privacy and freedom should not be the default, as it is encouraged today.
>To be sure, child pornography filtering — and reporting — may be a place to draw a clear line. Not only is child pornography near-universally reviled and banned, but the matching algorithm for previously-identified images boasts no false positives, and, perhaps most important, possession of the file is not only clear evidence of the crime, but quite typically the crime itself.
I had no idea Gmail did this.
I'd use DuckDuckGo if it weren't for the fact that its index is far smaller than Google's, although with the latter's increasingly aggressive pruning of results, maybe that won't remain the case for long...
It uses Google, but Google won't get your ip.
Hm. Wonder what algorithm they use, and if it's one that has collisions, and if so, can someone distribute images that alert Google's system but have no illegal content.
https://www.microsoft.com/en-us/photodna
https://en.wikipedia.org/wiki/PhotoDNA
All of us are culpable, to the degree we contribute cleartext to the butt and encourage others to do so as well, for creating these massive data silos that are gifts to (micro)politicians everywhere.
This type of risk is one that you must accept if you rely on a Service as a Software Substitute[0].
By relying on a remote service for your data storage and computing, you give up both freedom and control---even if the data are encrypted; we've all seen what information can be gleaned from metadata[1]. In this case, perhaps where a document was distributed: if Eve knows that Alice viewed document X, and Bob has a document Y that is shown (through metadata) to have been sent from Alice, and that document has certain properties that match, then it might be enough to convince a court that Bob contains document X, and get a warrant.
> Companies treat child pornography and abuse in a special category. But the precedent remains, and it could be tempting to extend at least the exact-match kind of search to the Paris terrorist example
Terrorism and child pornography are two examples that are often used to justify broad searches, chilling free speech and violating privacy.[2] These types of privacy/security "trade offs" start down a very slippery slope.
> As so much of everyone’s private communications and work migrates into the hands of a few massive private companies, the net-wide search will become too tempting to leave alone. Exactly what makes it tempting is what makes it troubling
You should reject those services and do your own computing![0] These remote services are not a substitute for locally installed software---a distinction that many do not consider or realize.
Not everyone has the technical capability to handle certain problems on their own; hosting your own e-mail server is a good example. In such cases, users should, at the very least, choose a service that commits to their privacy, to the degree that LavaBit did.
For social media: consider decentralized alternatives, like GNU Social[3], GNU MediaGoblin[4], and many others. Using those services will also encourage your friends and family to ask questions about why you made that choice, and consider switching themselves, since they'll have others they know using those services. Even if you cannot host your own, you can shop around for hosts that you feel that you can trust, and still participate in the same network as your friends and family; there's no lock-in to one service like Twitter or Facebook. Wholesale searches are also made rather difficult by distributed networks.
Forfeiture of privacy and freedom should not be the default, as it is encouraged today.
[0]: https://www.gnu.org/philosophy/who-does-that-server-really-s... [1]: https://www.eff.org/deeplinks/2013/06/why-metadata-matters [2]: https://www.eff.org/deeplinks/2015/08/mandatory-reporting-us... [3]: http://gnu.io/social/ [4]: http://mediagoblin.org/