Given that you have to give your SSN for all kinds of miscellaneous stuff where you don't particularly trust the recipient, these days (credit apps, etc), I don't think anyone has the illusion that their SSN is private, do they? Still?

Privacy is effectively dead, and the only substitute we have is secrecy, which can be destroyed by any party to the secret. This is the world we live in, now, and I can't see any way to reverse course, no matter what the law says.

Yes, but remember, I'm talking about the situation in Stallman's world. No one would use non-open software, and no one would send their personal information.

As far as privacy being dead, "dead" is a misleading term here. Privacy is a concept, and as such has no vital signs. It can be alive again as easily as a comic book character. The transition from the time before universal suffrage is just as big as the change from the current situation to privacy as a right.

If you ask ordinary people about these topics, you'll find that many, if not most people out there have some sort of expectation about privacy. I think that these are rights waiting to be discovered. Just as my downloading a GPL'ed app or marking the checkbox under the Terms of Service means I agree to a specific contract, I think that there should be certain implicit contract going the other direction. Just as no one should be able to, say, sell kitchen utensils that blow up and kill people.

The SSN as a secret is in trouble. However, there are more robust protocols than the secret number.

In that case, both my blathering and Stallman's would be irrelevant, on this subject.

I think that universal suffrage is not in the same category as privacy, because nothing about technological advance really dictated it. Privacy and copyright are more like slavery (not making a moral statement here!), in that after certain technological advances, it's just not possible to sustain them and continue competing economically. That's what I mean when I say that privacy is dead: without some unforeseen technological problem (or solution, if you're so inclined) that makes it possible to control other people's computers with your software against their will, it's only a matter of time until societal mores catch up and recognize the de facto situation.

Actually, what you're talking about is Digital Rights Management. IF DRM worked, it could be used in the other direction, and your private information could be distributed in the form of software agents. Big if here, granted. However, I suspect that something like Vernor Vinge's "Trusted Computing Environment" will eventually exist. It's too useful not to.

There is another mechanism that already can "control other people's computers with your software against their will" using legal means. Namely the GPL.

Something like the GPL combined with simple protocols can protect private information, say, like a number. Using one-time proxy-numbers, I can already shop with my credit card, but without revealing my account number. Craigslist also has a similar mechanism to protect emails. If this were built into email clients, then it would be easy for users to register, and always know exactly who sold their email address to spammers.

Better yet, why not build registration protocols into browsers in the same way RSS reading and SSL session key negotiation is?

This sort of thing will not completely prevent the aggregation and inference of private data. But combine these mechanisms with legal protections of privacy, and it will make it very hard for corporations to use the data they aggregate illegally.

only a matter of time until societal mores catch up and recognize the de facto situation

There was a time when the US was very cavalier about copyright, and we just up and published our own editions of books without paying royalties. Societal mores were fine with this. (Dickens had to go on tour and do readings of _A Christmas Carol_ to make money from it in the US!) Somehow, the situation shifted the other direction in this case.

I agree that it's a big 'if', because I don't think that DRM will actually ever work reasonably well, and I don't think usefulness is a huge boost -- there are too many confounding factors, even if the encryption itself isn't cracked.

I'm not sure what to quote of the middle of your post. :) First off, "legal" is pretty fluid; I can find a jurisdiction that allows me to do whatever I want with personal information. Second, once someone has a copy of personal information in an unencrypted format, the source becomes very difficult to identify (unless everyone gets a customized copy, as you point out for email; I used a scheme like this to track who was selling my email to spammers from 1998 for a few years, but it didn't do me any good to find out, and it was a bit of trouble to keep up with).

But essentially, privacy and copyright are the same fight, and if you can come up with an effective enforcement tool for either one, the other one is solved as well.

Somehow, the situation shifted the other direction in this case.

I'm not suggesting that there's any social inevitability, here, except as a result of a technological one.

The GPL seems to work regardless. It works because it is respected by and large by the programmer community, and because most of the relevant places respect the relevant laws.

Second, once someone has a copy of personal information in an unencrypted format, the source becomes very difficult to identify (unless everyone gets a customized copy, as you point out for email;

A "Trusted Computing Environment" would ensure that only the trusted environment would ever get a copy.

I used a scheme like this to track who was selling my email to spammers from 1998 for a few years, but it didn't do me any good to find out, and it was a bit of trouble to keep up with).

If it was implemented as a protocol and integrated in the browser, the user would only have to see "Received at address send to X" and the rest could happen automatically.

I'm not suggesting that there's any social inevitability, here, except as a result of a technological one.

But printing kept on getting more efficient throughout the Industrial Revolution. The cost of transcription was a relative non-issue. The technology trended in a direction that would indicate more copying. Yet it came to pass that there was less copying of books and that it wasn't as acceptable. Why?