7 ms·
It's not so much about expecting things to be private, but more of expecting things not to end up in databases, virtually forever. As programmers, we tend to v
by scrollaway 12y ago
It's not so much about expecting things to be private, but more of expecting things not to end up in databases, virtually forever.
As programmers, we tend to view the world in a very "black and white" fashion, similar to our security model: If it's public, there is no expectation of privacy.
But the world is not like this ... shouldn't be like this. While there should be both technical and legal mechanisms in place to prevent privacy abuses, it does not mean that, if they're not there, we should simply expect governments to do as they please, log everything, put us in all sorts of databases, make those databases accessible to all sorts of people, etc.
- Agnogenic 12y agoI have to agree. The issue is isn't about the USPS having access to the information on the outside of mail, but that they are cataloging it. When I send a letter, I am providing my information for a service to be fulfilled and not for it to be put in long-term storage. In some sense, I think this is comparable to email. If Google was giving out a record of the addresses I've sent email to and received email from, I would consider it an invasion of my privacy. Services need to be upfront about what is being done with our data--anything else is dishonest. When I know what is being done with my information, I can make an informed decision on what services to use.
- nfoz 12y ago> If Google was giving out a record of the addresses I've sent email to and received email from, I would consider it an invasion of my privacy. Yet you're sending this in cleartext over the internet...
- wavefunction 12y agoAnd yet they are doing that in full knowledge of what they are doing, which is not what is happening either with the post office or the expectation one has with Google. Different situations have different expectations and outcomes.
- XorNot 12y agoWe have had actual cases where someone mailed a bunch of anthrax through the US post. Given that the post can be used to send dangerous physical objects, it seems prudent to maintain medium-term records of what was sent and where to assist in the prevention of this type of terrorism. Because the alternative, is we wait till it happens, we have no record of where anything came from, and then we get to engage in a multi-month campaign of opening and inspecting mail arbitrarily considered "suspect" to try and prevent the next round.
- wooter 12y ago"type of terrorism" what makes your example "terrorism" and not "crime"?
- XorNot 12y agoWhy is any terrorism not just "crime"? Call it what you want: if we have no records of where mail was sent from and where it got delivered too over a medium-scale timeframe, we are seriously limiting any ability to deal wit this type of activity effectively. The term "letter bomb" is a real thing that was and still is really done.
- DanBC 12y agoDo terrorists put a correct return address on the mail and use a local mailbox? Or do they forge a return address and use a distant mailbox? Mail data didn't catch the unabomber - has it caught anyone?
- XorNot 12y agoYou have to travel to the mailbox to send something from it. That mailbox is generally surrounded by other people who might notice it's comings and goings, or unfamiliar characters posting large packages. Frankly, the idea that metadata can't help us solve crimes but it can obviously be used to nebulously control a population is an oxymoron. Either metadata is useful, or it isn't. If it's useful, then why isn't it useful or potentially useful for dealing with crime? You can't have it both ways.
- hueving 12y agoBut Google does keep precisely the information you described and will quickly give it to federal law enforcement as required by law without informing anyone (also required by law).
- Agnogenic 12y agoAgreed. But, the issue is transparency. Now that I know what is happening with my Google data, I can at least make a decision to not use that service. Both the data collection and the act of keeping that data collection a secret are issues.
- baddox 12y agoI disagree, because simply having a government database of this ostensibly public information is still not a threat, if the government will not take any actions which would be unacceptable regardless of whether the database existed. To use an extreme example, say there is an advocacy organization which the government decides they don't like. I'll use the ACLU just as an arbitrary familiar example. They could use their postal database to find all the people that have written letters to the ACLU, and target them in some way. It could be something extreme, like throwing them all in jail, or something more subtle, like using the IRS to audit them. Now, I think we can agree that this action is unacceptable. But the thing is, this action would still be unacceptable even if they didn't have or use the database. If they simply sent a government official to your house, asked you if you were involved with the ACLU, and arrested you if you said yes, that would clearly not be a violation of privacy, but it would clearly be unacceptable. So I don't really see the scanning of the outside faces of letters to be a privacy issue. But I reserve the right to be appalled at the things the government might do and probably already does with this information, independent of how they acquired the information.
- zAy0LfpBZLC8mAC 12y agoThe whole concept of a threat is all about uncertainty, so it's somewhat incoherent to talk about it in connection with a supposed certain future. An argument of the same structure would be: Simply playing russian roulette is still not a threat, if the chamber will be empty. Well, yeah, _if_ - it's a threat precisely because you do not know whether that condition will hold, and there is a reasonable chance that it won't; if you knew, it wouldn't be a threat but either a boring game or a method of killing yourself. Also, why would your argument not apply to medical information, say? If they collected your medical information from your doctor's office to then blackmail you with it, that is not a privacy issue because they could also have come to your door and asked you about your medical history, and that would also be unacceptable, wouldn't it? If I understand your argument right, it is roughly "if 'they' can do something bad in more than one way, then either of the ways is not bad because it would still be bad if the other way was used"? I'm not quite sure how to make sense of that. Also, one anecdote from history: The nazis when they invaded the Netherlands used existing census information on religious affiliation that was easily available on hollerith punch cards to find and deport the jews. Would you say that that information collection was not a threat for the jews who probably got killed faster and in larger numbers than if the data had not been available in that form? edit: One more example to consider: A dictator is not a threat, if he doesn't do anything that would still be inacceptable if it was done by a democracy. Well, a dictator is a threat exactly because you don't know that, and we tend to prefer democracy in order to protect ourselves from that risk of having a dictator of the bad kind.