> Democracy works because we assume that communication can be done freely and anonymously.

The real name policy on Google+ doesn't change anything about that. People can still communicate anonymously, just not via Google+. Democracy does not require anonymity on every communication platform imaginable. There also isn't anonymous internet access, and there aren't anonymous domain names (or only few).

Democracy depends on the State allowing anonymity, not on every communication provider implementing it. Google thinks that it's system profits from accountability, and they have the right to enforce it on their platform.

If it's a single provider doing it, without anything close to a dominant market share (the current situation), I agree. It would become more problematic if it became de facto the standard communication platform, though, or even a large share of it. For example, if Facebook messages replace email for 90% of the population, you will find it very difficult to avoid using Facebook messages.

I don't see it as much different than the State, really. Companies with smallish market share are like local governments: relatively harmless, because you can just avoid the ones you don't like. Bigger governments and companies get increasingly hard to avoid; it's easier to move out of Pittsburgh than to move out of the U.S.A., and it's easier to avoid Google Plus than to avoid the big-4 telecoms.

If a G+ or Facebook achieve 90% of communication volume, it means that anonymity isn't an important factor for most people and you could in fact argue that the high market share is due to the fact that there are less problems with spam and abuse because there is accountability. If a significant amount of people desire anonymous communication, there will be a communication platform for them.

Also you can avoid G+ and the big-4 telecoms much easier than you can avoid living in a state. The need to live is much greater that need to have internet access, no matter how you slice it.

You can avoid living in a particular state more easily than you can avoid internet access, I'd argue. In 2011, the hardship of going without internet is much higher than the hardship of moving from California to Nevada, for example.

I do agree that national-level laws are more coercive, in part because many countries are just very large, and in part because immigration laws make it hard to move.

"Democracy works because we assume that communication can be done freely and anonymously. If these assumptions don't hold true for major communication platforms such as Facebook, Twitter or Google+ we do have a significant problem."

Face to face communication has always been available freely and anonymously. I'm not quite certain that any other form of communication has ever been either freely available or remotely anonymous for anyone remotely like an average person.

What communications platform has it ever held true for?

1.) Telegraph -- cost money, not universally accessible, open, message assured to be read by at least 2 3rd parties.

2.) Mail -- costs money, need a registered address, handled by a quasi govt agency, can be read by 3rd party.

3.) Land line Telephone -- costs money, not universally accessible, requires registered address, every action is logged including recipient and duration. Calls can be listened to by 3rd party.

4.) Cell Phones -- costs money, not universally accessible, burner can be had with no address or name. Calls can be listened to by 3rd party. Location tracking available.

5.) Home Internet -- costs money, not universally accessible, requires registered address, everything routed through ISP who keeps user logs, network that can intercept.

It's unclear to me why the service layer attracts so much attention, when the underlying network layer currently makes it largely impossible to have what you want for the average person. Having anonymity at the service layer is almost confusing when the underlying framework guarantees it is largely not real. There are all sorts of options already available at the service layer that can give all sorts of false appearances, many fewer at the network layer to make those false appearances turn real. Twitter and Facebook and G+ are trivial issues in comparison to getting around the fact that backbone and access points are a network owned by 3rd party intermediaries (just as with all the previous communications platforms above).

While all the above points are true, there is a big difference between picking up the phone to call someone in 1950 and posting an update on a social networking site in 2011.

I would submit that there are two factors that make the difference: 1) laws regulating the service in question and 2) ease of data aggregation / mining.

In both cases there are barriers to interception & analysis of communication that effectively provide or used to provide a certain level of expected privacy and anonymity. Of course no communication is 100% secure and of course anything can be intercepted, but how hard is that to do?

In 1950 if you wanted to get pictures of someone's house, learn the names of their known associates, find out what books they own, what magazines they read, what opinions they have, their work history, their goals, their aspirations, which of their relatives they are closest to, etc. etc. -- your work would have been cut out for you. It would involve physical trips to places like courthouses, time and effort in taking pictures, developing film, driving around, possibly snooping where you shouldn't, asking lots of people lots of questions, and so on.

Now all that is out there with a few mouse clicks and not only is that the case, it is rapidly becoming the social norm that it should be out there and if it isn't, as Mr. Schmidt says, then you must have something to hide.

The genie of cheap, accessible, centralized data storage and data mining is out of the bottle. At least with email there is some sort of legal precedent to treat communications as privileged. With social networks and the like, users give up their rights to the commercial entity that provides the service. And as others have pointed out, rolling your own services is not an attractive alternative.

Agreed, the network layer can be intercepted, but the network layer either has to be hacked into or somebody in law enforcement has to leave a data trail regarding the interception. At the service layer, it is becoming assumed that you give up all rights to your data, and this is something whose repercussions have yet to be fully felt.

I think this misses the wider point. Google are basically saying: "we can't guarantee your anonymity so we won't pretend we can".

By enforcing a real name policy they are sending a message to those that need real anonymity: use a system designed from the ground up to give you privacy.

Anonymity and pseudonymity are two orthogonal concepts. Google could easily offer pseudonymity without making any sort of offer of anonymity. Arguing against pseudonyms with "We can't guarantee your anonymity" is nothing but a straw man.

I don't think the everyman views them as separate concepts.

People honestly believe that if you're using a pen name, then you're anonymous to those who you haven't told your real name to. Note even the terminology used here "Real Name", as in anything that isn't your government ID name is not actually a name. Pseudonyms in formal language are 'names', but not so in informal language.

Then caveat emptor.

"You are not required to use your Real Name to use the Service provided by Google. However Google makes no assurance that your true identity cannot or will not be disclosed even if you make use of a Pseudonym to use the Service."

... or words to that effect in the TOU would address the matter just fine for most people. For those who truly require military/intel grade anonymity, there are alternatives. For those who just want to engage with the general public without revealing their real name to all and sundry (absent hacking or subpoenas), the above should be sufficient.

If Google wish to go the extra mile and provide assurable anonymity, that's another matter.

Otherwise, this is a case of the perfect being the enemy of the good.