Search Engines and Privacy Rights on the Web Last week, the Justice Department asked a federal court to order Google to comply with a subpoena issued last year, seeking all search queries within one specific week, and 1 million random URLs. Google is fighting the subpoena. The government hopes to use the data to support its case in a court battle over the 1998 Child Online Protection Act. The act is intended to protect minors from exposure to online material deemed harmful to them, primarily pornography.
NPR logo

Search Engines and Privacy Rights on the Web

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Search Engines and Privacy Rights on the Web

Search Engines and Privacy Rights on the Web

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


This is TALK OF THE NATION, I'm Neal Conan in Washington. Last week, the Justice Department asked a federal court to order Google to comply with a subpoena issued last year seeking all search queries within one specific week and a million random website addresses. Google is fighting the subpoena, in part, because they say that handing over the data would compromise trade secrets. Three other search engine companies, Yahoo, MSN and America Online complied with similar demands.

Government agencies often request personal information from internet companies to gather evidence in criminal investigation, but this time, there is no criminal investigation. Instead, the government hopes to use the data to support its case in a court battle over the 1998 Child Online Protection Act, or COPA, a law that's intended to protect minors from exposure to online pornography.

The subpoenas raise at least two important questions. First, how much data do internet companies really have on their users, and second, should the government be allowed to demand broad swaths of data like this from internet companies? Later in the program, it's the Talk of the Nation opinion page.

Proposals for a new internet fast lane that would only be available to websites willing to pay a premium; would it tilt the field? But first, internet privacy. If you have a question about the subpoenas in this case, about how much information online companies gather about you or what you can do about it, give us a phone call. Our number here in Washington is 800-989-8255, 800-989-TALK. Email us, the address is Joining us now is NPR Correspondent Laura Sydell, she's with us from NPR's office in San Francisco. Good to have you on the program.

LAURA SYDELL reporting:

Hi, Neal.

CONAN: What do we know about what type and how much information the government's already received from AOL, Yahoo and MSN?

SYDELL: We don't know much, actually. The only thing that these companies have said is that they've handed nothing over that is what they would consider private or personal, so nobody's name. The government doesn't know where these things are coming from. AOL claims that what they handed over actually were things that the government could have gotten themselves just by going online and looking around. Less is known about Yahoo and MSN and what they gave.

CONAN: Now, how does the government say this data will help it in its case for COPA, the Child Online Protection Act?

SYDELL: Well, what the Act does, essentially, is, it says that, sites which are pornography sites, will have an active responsibility to keep minors out. So, they have to have, you know, they have to have somebody give a credit card, or something that identifies them as an adult.

CONAN: Some sort of filter.

SYDELL: Well, not a filter. A filter is another kind of thing, and this is about filters. Normally, a lot of parents use filters. They have filters on their computers. The government is saying that those filters are not really that effective. And that, therefore, you actually have to have these websites do something themselves. And that's where this law has been challenged. Because the courts have looked at it suspiciously saying, well, isn't there another way to do this that isn't a burden on free speech, isn't a burden on these businesses?

So, in this instance, the government is hoping that, by getting all of this information, and they can see, say somebody typed in a search query that seems innocent, did it bring back a pornography site? So, if a child went in and typed, I don't know what phrase, any phrase, any given phrase, Disney and girls or something, did it bring back a pornography site? And was the filter able to keep it out? If the filters weren't able to keep it out, the government's saying, this is why we need this much stricter law. So, that's what they're trying to get from Google by looking at all these search queries.

CONAN: But if they're getting no private information, how would they know which search query came from somebody who was 12 years old and which came from a 35-year-old?

SYDELL: They wouldn't. I think what they're trying to see is what various search terms bring up. So, they're looking to see, okay, there were this number of search terms, which seemed completely innocent.

CONAN: I see, I see,so...

SYDELL: What did they bring back?

CONAN: Yeah, it used to be that would bring up a, you type .com, instead of org, you'd get a pornography, doesn't do that anymore, but...

SYDELL: Right, exactly.

CONAN: That might be an example of it.

SYDELL: Exactly.

CONAN: So, as we're going through this now, why is Google of the three, resisting the subpoena?

SYDELL: Well they say, actually, what they're saying is it's proprietary. Google as you know is the hottest search engine, and they have a lot of rivals. And a lot of people would like to understand their algorithms, how it is they're able to do the kind of search they do, bring back the kind of results they do. And Google is saying, by giving the government this information, somebody could, basically, back-engineer, could look at what came up, if somebody who is a rival gets a hold of this, and know more about what Google does, and possibly try to recreate it.

And so they're saying, we don't want anybody to get that information. Now, what's happened, interestingly, is that, civil libertarians have been giving kudos to Google for this, although that wasn't really their defense. But that's probably also a good thing for Google, because Google has always been this company that doesn't like to do business the way anybody else does business. And they like to say that we're a company that will do no evil, I believe is the motto.

Or, don't do evil. And so, this gives them even more good publicity by saying, we're standing up to Big Brother, here. So, this is a really good thing. And, I guess, the other thing for them as well is when they went public, people really were thinking, they will start to change. So, this shows, no, we're not changing, we're still standing up. So, this is actually very good publicity for Google.

CONAN: On the other hand, it also discloses just how much information Google and these other search engine companies do gather on their customers.

SYDELL: It does, indeed. And, I think, in fact, I spoke with a young man who uses Google Gmail and uses the search engine all the time, and Google has an email service, and with it, when you sign up, they actually say, we're going to keep track of what's in your email, and use it to send you targeted advertising. And, this young man was saying to me that he knew that, and then, all of a sudden when he heard about this, it made him think, huh, God, they really do have a lot of information on me. What if somebody got that?

CONAN: And if, you know, if they could get your personal queries, I mean, it's almost like reading your mind.

SYDELL: Well, that is the thing, and that's actually, what some civil libertarians are saying, why the government shouldn't be able to get this. Because, really, these are communications. These are communications that you might not want people to know, you might not want to tell your wife, or, who knows, you know what I mean. Or your kids or your mother, who knows. But it's actually a very personal thing to know about somebody, what it is they've been searching. And it's not private.

CONAN: Laura Sydell, thanks very much.

SYDELL: You're welcome.

CONAN: NPR's Laura Sydell joined us from the NPR office in San Francisco. Let's turn now to Tom Lee, a Professor of Law at Brigham Young University. He used to be a lawyer with the Justice Department, involved in defending COPA, and he joins us now from his office in Salt Lake City, Utah. Nice to have you on the program today.

Professor TOM LEE, (Professor of Law, Brigham Young University): Good to be here, thank you.

CONAN: Given you involvement in the case, I think you probably hear Laura's description of how the government hoped to use this data, does that sound right?

Prof. LEE: I think she's got it about right, Neal. One clarification, or one sort of further point to understand here is that, this case has bounced up and down in the court system a couple of times. And this is Congress' second or third attempt to deal with what, I think everyone agrees is a very serious problem, which is access of minors to sexually explicit material, commercial pornography on the internet.

And, the government's in the position that it's in right now because the Supreme Court sent this case back down to the District Court on the basis of its conclusion that there was a need for further evidentiary development. Specifically, on this question that Laura mentioned, which is, there a less restrictive alternative to the age-screening mechanism that COPA sets up? And, in particular, to compare that age-screening mechanism to filtering.

And, that's what the government's trying to accomplish, here, as I understand it. These are subpoenas designed to assemble the raw data that the government's experts would then take to evaluate the effectiveness of filters, and then to take that next step that the Supreme Court mandated, which is to compare the filters, effectiveness of the filters, with the effectiveness of the age screen or age-verification mechanisms that COPA sets up.

CONAN: And does this strike you as a reasonable way to gather this kind of information, or might have there been other ways to go about it?

Prof. LEE: Well, the government made an initial attempt to go about it in a different way. And it was to evaluate and to criticize the effectiveness of filters at the general level, and at the general level, by submitting expert testimony and other grounds for questioning the effectiveness of filters. Filters are widely known to be both under and over inclusive. They under-block, and they over-block. They miss certain kinds of sexually explicit material.

CONAN: And they block stuff that's, when looking at it, is actually quite innocent.

Prof. LEE: Yeah, exactly, because filters by their very nature use words. They tend to use phrases and words, and they can't very effectively screen by way of just focusing on the images. And the government made all those points in a litigation in the first time around, and made those arguments to the United States Supreme Court.

The Supreme Court said, that's not good enough, that under the First Amendment doctrine, as the court articulated it in the majority opinion in the case, you had to go back and you had to assemble a statistically valid showing, an evidentiary showing, comparing and contrasting the effectiveness of the age verification mechanism with the filtering mechanism. That's what the government's doing now. It's responding to a Supreme Court majority opinion, and it's assembling raw data to give to its expert witnesses. The subpoena focuses on two categories of information. One, a random sample of URLs or websites that are available from Google's database, and second, a random sample, of, or a one week sample, I guess, a time based sample, of search queries. What sorts of searches do people make when they go on Google, and to get, like the discussion suggested earlier, not just Google, but other search engines, to then take that raw data, and allow the expert witnesses in the case to evaluate the effectiveness of filtering software, figure out whether, if you were to search for a certain URL, a certain website, or undertake a certain search string, what would you get to and what could you prevent minors from accessing if you used the filtering software?

CONAN: Okay, Tom Lee, stay with us. We're going to take a short break, and when we come back, we'll get calls in from listeners. 800-989-8255, if you'd like to join us, or you can send us e-mail, That's our new e-mail address, if you'd like to join us that way. I'm Neal Conan, we'll be back after the break. Talking about Internet privacy, and how much personal information Internet search engines gather about us.

I'm Neal Conan. This is TALK OF THE NATION from NPR News.

(Soundbite of music)

CONAN: This is TALK OF THE NATION. I'm Neal Conan, in Washington. We're talking about Internet privacy and Google's decision not to comply with the government subpoena to provide information about a week's worth of searches, and a million URL sites. Three other big search engine companies did comply with those subpoenas. If there are questions about how much personal data Internet companies keep about you each time you surf the Web, give us a call, 800-989-8255. Our e-mail address is

We're talking right now with Tom Lee, a professor of law at Brigham Young University; formerly worked on this case at the Justice Department.

Let's get a caller on the line. This is Scott. Scott's calling from Greenville, North Carolina.

SCOTT (Caller): Thank you, Neal. Curious as to Professor Lee's interpretation of the Fourth Amendment implications of this kind of broad subpoena for all kinds of people's information, since it's a given that search engine companies have a great deal of information about their users, is this sort of broad subpoena not over broad under the Fourth Amendment unreasonable search?

CONAN: I believe that was one of the arguments, Tom Lee, that Google did make.

Prof. LEE: Google has raised two concerns. One of them is a privacy concern, which, I think, Scott's question gets at, and the other has to do with trade secret and other...

CONAN: Yeah, we talked about that with Laura Sydell earlier. Let's get to this other point.

Prof. LEE: Yeah, let me talk about the privacy interest here. As I understand it, Scott, the subpoena that has been issued, here, would not reveal any information specific to any individual. In other words, what the government is requesting would be lists of URLs and search terms stripped of any information that would tie that back to any individual internet user.

So, yeah, it's true Google may elsewhere, through its Gmail and other services, compile user-specific, individual-specific information that may raise all sorts of privacy concerns, that may or may not be allayed by the fact it was mentioned earlier, which is when you sign up for Gmail, whether you know it or not, you may be voluntarily giving up some of your privacy rights by agreeing to allow them to give information away.

But, as I understand it, here, the information simply wouldn't disclose any of that particular user-specific information. We're just talking about the raw data of URLs and the raw data of search query.

CONAN: And I can anticipate the next question from a listener, which would be either thin end of the wedge or a slippery slope, that this is just the beginning, the first bite of the apple, that it's a bad precedent.

Prof. LEE: And that certainly is always a concern, Neal, but I think that, you know judicial, decisions area made one case at a time, and the case that's before the court right now, I don't think it's very far down any sort of slippery slope. I think this is a very narrow, limited, carefully crafted subpoena that is designed to get at a very serious problem, and designed to respond, again, as I mentioned earlier, to a Supreme Court decision that says government may not defend this statute merely on the basis of broad generalizations about its concerns about internet filters.

It's got to compile an evidentiary submission that evaluates the effectiveness of those filters, and that compares them to the age verification requirements of COPA. That's what it's doing, and I think this is pretty narrow. And I understand there's a slippery slope out there. I just don't think we're there yet.

CONAN: Okay, Scott, thanks very much for the call.

SCOTT: Thank you.

CONAN: Here's an e-mail we got from Liz Klinkenbeard(ph), who writes that she works in media relations for Yahoo and wanted to make sure that we had a copy of Yahoo's statement on this story, and Yahoo's commitments to user privacy.

Quote, "We are rigorous defenders of our user's privacy. We did not provide any personal information in response to the Justice Department's subpoena. In our opinion, this is not a privacy issue. We complied on a limited basis, and did not provide any personally identifiable information."

Now that, again, from a press person for Yahoo.

And let's get another caller on the line and this will be Chele(ph). Chele's calling from Winston Salem in North Carolina.

CHELE (Caller): Yes, I was wondering, why is it the government cannot just sit down in front of their computers and perform Google searches to get this information, or better yet, write a software program that can go through Google's page like anybody else, and get the information on their own? Is that not a possibility?

CONAN: Tom Lee?

Prof. LEE: I think that the answer to that question, Chele, just has to do with the way the law of evidence works. The criticism, again, that the government got, and that resulted in this remand from the Supreme Court was you haven't compiled an evidentiary submission to the court to defend the statute. What you've done is to make generalizations about the effectiveness of filtering software, and to say we don't think filtering works. We think it's over and under inclusive.

The court said that's not good enough. There are serious First Amendment interests at stake here, and, you know, specifically we may, under COPA, block adult access to First Amendment protected material. And, in order to take account of those First Amendment interests that the court found important, the court sent this back, the Supreme Court sent this back to the District Court.

The way the rules of evidence work, it's not enough to speculate about how users may access certain sexually explicit or pornographic material on the Internet. We got to know what actually happens. We got to know what happens in the real world, and the way most people find access to information on the World Wide Web today is by using search engines.

And the government, therefore, has contacted these search engines, and attempting to assemble this evidence, this raw data that it will use, then, to ask its expert witnesses to look at and to evaluate whether these filters really work, or how well they work.

CHELE: Well, it seems to me that to be fairly irrelevant if all they need is the data for what particular phrases turn up what particular Web pages. It seems like they don't need my particular searches, or other users' searches. They could actually perform the searches with the Google search engine like we all do.

Prof. LEE: Again, they could, Chele, but the relevant question of evidence before the court is, what do people really do? What does the typical Internet user do in using the Google search engine, or the Yahoo, or the MSN, to try to access this material? And if the government's expert simply said here are the searches we came up with. Here are the searches that we dreamed up that we think people might engage in.

The obvious response from the ACLU would be, and frankly this would be a response that I think would get some traction in light of the Supreme Court opinion, no, that's just your own speculation. That's what you did last time around, that's what wasn't good enough, according to the Supreme Court, to support this statute. We got to know, actually, what really happens in the real world so we need to have some evidence, and that's what this is about.

CHELE: Do we have to have the actual information that we use, which I believe should be protected? Thanks, I'm going to end my call.

CONAN: Okay, thanks very much, Chele.

Prof. LEE: And again, just to clarify, Neal, I think the reason the rest of the search engines have responded here, and the reason that the Yahoo e-mail that you read says we don't think this implicates privacy interests is, I think this has been broadly misunderstood.

The gut reaction to this is, gosh, this is going to disclose privacy protected individual specific information. That's not what this is getting at. The government doesn't want to know who performed a search. The government just wants to know what the typical query looks like.

CONAN: All right, let's see if we can get one more question for you before we have to let you go. Alex is with us. Alex is calling from Reston, in Virginia.

ALEX (Caller): Hello.


ALEX: I have a comment that the administration has a much better technique they can use to prevent minors from getting access to porn, if they would only follow the recommendations of the task force on naming domain names on the Internet.

CONAN: Mm hmm.

ALEX: That they recommended a XXX, top-level domain name, where then porn sites could be pushed into that XXX domain name. Filters could easily block it.

It would basically be a self-segregation, and that was the recommendation from the task force, and the Bush Administration has blocked it for the last couple years. It would simplify this whole problem, and avoid searching and communications of adult U.S. citizens or...

CONAN: Tom Lee, what about a red light district on the Internet, XXX?

Prof. LEE: You know, there are, Neal, as Alex's comment indicates, a number of ways of dealing with the broader general problem that we're talking about here. And I'm not frankly intimately familiar with XXX domain that's been mentioned. I think it's worth exploring and thinking about a number of different solutions. The question here, in this case, though, is the particular solution that Congress came up with and whether it's constitutional under the First Amendment.

I happen to think that it's an eminently reasonable way of dealing with this problem. To say there is commercial pornographic information out there, and I think most reasonable people understand that there's an impact on children that represents a very serious problem. It isn't the same as the impact on adults and what this statute attempts to do is to provide an age verification requirement that will screen out children, and prevent them from being exposed to this material.

CONAN: And, Alex, I suspect if you created the XXX, somebody has to decide what's pornography, and what isn't, which raises a whole another bunch of issues. But anyway, thanks very much for the call.

ALEX: Sure.

CONAN: And, Tom Lee, we wanted to thank you for your time today.

Prof. LEE: Thank you, Neal.

CONAN: Tom Lee, a professor of law, now at Brigham Young University. He used to work for the Justice Department, and helped formulate some of the arguments in favor of COPA, the Child Online Protection Act. The tale of Google, and the government subpoena brings to light the volume of information Internet users unwittingly reveal about themselves when they go online. Xeni Jardin is a tech culture journalist, and a regular contributor to NPR's DAY TO DAY, one of our sister programs. She joins us from the studios of NPR West. Nice to speak with you again.

Ms. XENI JARDIN (Tech Culture Journalist, Contributor to NPR's DAY TO DAY): Hi, Neal.

CONAN: These search engine companies, Google online, Google, AOL, Yahoo, they learn an awful lot about you.

Ms. JARDIN: You know, they do. There was a terrific article in Wired News, I think, a few days ago, by Ryan Single, about all of the information that is collected and how to foil search engine snoops, as the title went, if you are concerned about the privacy issues. But, sort of, the basics are this. If you've never signed up for an account, say a Gmail, Google mail account or you've never logged into a search engine's website, chances are, they don't know your name and exactly who you are.

But these search engines connect what you search for through a little piece of code called a cookie, which has a little identifying number that identifies you uniquely, and using that cookie, the search engines can keep track of all the searches you make with your browser, and, and could, potentially, identify where you are by way of an IP address, and that's like a little numeric address that shows the computer that you use to connect to the Internet.

CONAN: It deposits this cookie in your computer?

Ms. JARDIN: Yeah, you can actually see the cookies that are stored for your computer by opening up your Web browser and looking at the list of cookies there, and one thing that people who are concerned about privacy might do is manually remove the cookies periodically, or you can tick off different preferences to manage those cookies. You may not know this, but for instance, Yahoo, I believe, sets a cookie that lasts, I think, for like, six months or something. It expires sometime in 2006. If I were to visit the Google site today, it would set a cookie that expires in 2036, so that's 30 years from now. That's a lot of searching.

CONAN: There's going to be an awful lot of requests for, where is Alex Chadwick? Anyway.

Ms. JARDIN: I'd like to know.

CONAN: Everybody would like to know, mostly his producer. Google's email service, called Gmail, even parses the message that its members receive to provide targeted ads.

Ms. JARDIN: Yeah.

CONAN: And Amazon, also, is another site, not one involved in this, but they go back and they try to figure out, you know, what else you would like to buy. If you bought these three science fiction books, maybe you'd like this other one.

Ms. JARDIN: You know, and, and, so the nice thing about cookies is that they allow service providers to provide personalized, tailored services back to you. For instance, the Google Zeitgeist. It's kind of a cool feature that shows what people are searching for on a given day, or say, localized searches, and just give you stuff that's going to be more useful to you.

But the scary thing, I mean, think about it, before, before we used Google, before we used the Internet, you had your phone company, you had the postal service, and you had stores, physical places where you would go and give people money and buy goods.

Imagine if the phone company, the postal service, and all the stores where you bought stuff all kept 30 years' worth of records on all the transactions that you made, all the phone calls, even the contents of the phone calls, and everything that you wrote in your letter to someone you loved somewhere across the world, and tried to serve you ads back against that. When you start thinking about how consolidated that flow of communication and personal information is, that's a little creepy.

CONAN: We're talking about Internet privacy, and subpoenas issued to search engine companies. You're listening to TALK OF THE NATION from NPR News.

And let's bring another voice into the conversation. Joining us now is John Palfrey, a professor and executive director of the Berkman Center for Internet and Society at Harvard Law School, who joins us to talk about what some people believe is a worrying aspect of this case, in that there's no law to protect private companies from handing, from having to hand over databases when the government issues subpoenas. He joins us from the studio at Harvard in Cambridge, Massachusetts. Nice of you to join us today.

Professor JOHN PALFREY (Professor and Executive Director of the Berkman Center for Internet and Society, Harvard Law School, Cambridge, Massachusetts): Thanks so much for having me, Neal.

CONAN: So, it's, it's the subpoena for all of this data about private citizens, even though it doesn't, in this case, it does not name individuals. That's what alarms people, isn't it?

Prof. PALFREY: It is. I think the, the key question is whether or not it's an appropriate thing for the government to be asking a large, powerful, private company, or a series of private companies, to turn over information about their customers in this manner. I guess I agree with most of your previous people on the call, Professor Lee and others, who point to the fact that this isn't about getting some individual piece of data about some individual internet user in order to solve a specific case.

So, there are very clear instances where the government needs to catch a bad guy who's done something wrong, or they want to prevent some terrorist act or something terrible, where they go to a big company and they say, turn over the information about this person based on this IP address. This is not that case. I think the problem here...

CONAN: Yeah, but when they do that, do they come with a search warrant? Have they gone to a judge and say, we need this information, and here's probable cause?

Prof. PALFREY: That's exactly right. So, in that context, there is, in the United States, very good safeguards set into the law that have a process whereby civil liberties are protected. This is a slightly different set of issues, right. This is to vindicate what the government thinks is its interest in trying to get a certain policy prescription passed.

CONAN: Mm-hmm. But if they get approval and go ahead with this, might there be other kinds of big, broad searches the government might try?

Prof. PALFREY: Right. So, your caller, Scott, was raising the Fourth Amendment issue and the privacy issue as a thin end of a wedge, and I think that that's one of the big concerns, and honestly, I think that is the bottom line why Google, maybe in addition to the trade secret concern, decided to draw the line here. Enough is enough.

CONAN: All right. Let's get another caller on the line. Nan's with us. Nan is in Oshkosh, Wisconsin.

NAN (Caller): Yes. I'm a management of information services professional. I've guarded network access, and surfed the internet and built databases, all that sort of thing, for about 30 years now. Basically, the government is going at this from the wrong angle. The actual search engine doesn't really have much to do where you get. It has to do with how easy domain access is, which is an internet function, which goes through the regulatory agencies for the internet, or what's on your computer.

On your computer, you can put firewalls on, you can restrict access through your network, you can put parental controls, there are all sorts of ways to fix this. Number one is browser choice. The only porn sites I ever ended up at were when I was using Microsoft's Internet Explorer, which didn't have decent filters at the time. I've changed to a different browser, and haven't gotten one since.

CONAN: Even if you wanted to get to one?

NAN: Well, you can get there if you want to, but that's the function of a parental control the parents could put on their Internet site. I had my children surfing the internet where I could watch them, so that wasn't the problem. A lot of it has to do with the laziness of not supervising the computer usage, either by an automated program, such as Net Nanny or some other parental control, or not setting up the firewall properly. And this can all be done when your computer is set up at the place your purchased it, if you don't have the knowledge yourself.

CONAN: We just have a few seconds before the break, but what if somebody misspells something by mistake, your 12-year-old?

NAN: If somebody misspells something very often, you could end up at the entrance to a porn site, but if you don't click on it, you won't get into it, and usually, the parental control will take care of that, and shut it down immediately.

CONAN: Okay, Nan, thanks very much for that.

NAN: Thank you.

CONAN: We appreciate it. We'll have more on this after the break and also the TALK OF THE NATION Opinion Page. I'm Neal Conan. It's the TALK OF THE NATION from NPR News.

(Advertising Break)

(Soundbite of Music)

This is TALK OF THE NATION. I'm Neal Conan in Washington, and here are the headlines from some of the other stories we're following here today at NPR News.

The Ford Motor Company plans to cut up to 30,000 jobs, and stop production at 14 facilities in the United States, Mexico and Canada. The nation's number two automaker is trying to stem huge losses in its North American operations. Union chiefs say workers are being made to pay for management's failures.

At a hearing today on Capitol Hill, senators discussed mine safety. Fourteen people have been killed in two separate mining accidents in West Virginia in the past three weeks. Details on those stories, and, of course, much more, later today on ALL THINGS CONSIDERED from NPR News.

Tomorrow on TALK OF THE NATION, we'll be listening to arguments for and against the White House's policy on eavesdropping on American citizens in context of the war on terrorism. Join us for that discussion tomorrow on TALK OF THE NATION.

Today, we're discussing internet privacy. Our guests are Xeni Jardin, a tech culture journalist and a regular guest on NPR's DAY TO DAY. Also with us is John Palfrey, professor and executive director at the Berkman Center for Internet & Society at Harvard Law School. If you'd like to join us, 800-989-8255, or zap us an email, And let's get another listener on the line. This is Todd, Todd calling from Salt Lake City, in Utah.

TODD (Caller): Yeah, hi, Neal.


TODD: I'd just like to make the comment, I don't believe there is a right to privacy when you use the internet, or if you use any type of, specific type browser or company. I believe that you're volunteering your information to give to a third party, and I believe that that could be subject to either the government's intervention, or there could be another company that may want to get that proprietary information, and that could be more of a technical response, but I'd just like to say, I don't believe there is an inherent right to privacy on the internet, since it's used as a public domain.

CONAN: Well, Professor Palfrey, I go to you first on that.

Prof. PALFREY: Thanks, Neal. Todd, so, you're partly right. In the context of a private company, there is no specific right, in the law, against that company doing something with those data. But it's actually not true with respect to the government, so the Fourth Amendment, the U.S. Constitution, absolutely applies, and you do have a right to privacy on the Internet against the government's intervention here, and that's one of the two arguments that Google has advanced.

CONAN: Mm-hmm. And Xeni Jardin, people sure use the Internet as if they think it's private.

Ms. JARDIN: They do. Nobody likes to read fine print, but one thing that we might be inspired to do this week is to take a look at the privacy policies and terms of use of the search engines that we visit. Google's privacy policy states pretty clearly what they will and will not do with your personal information, but part of the concern here is that when the federal government is the entity asking for data from that company, all bets may be off.

CONAN: Okay. Todd?

TODD: I just kinda look at it as if I wanted to conduct business on a street corner in front of a private business, that I know that I'm subjecting myself to whatever powers could be to gain that information, and I look at the internet, since it's a worldwide domain, the same type of situation applies. And I believe people who assert their right to a privacy of that are, are taking the Fourth Amendment out of context.

CONAN: Okay. Well, Todd, thanks very much. Appreciate that.

TODD: Thank you.

CONAN: Professor Palfrey, online companies, obviously, are keeping all this information, partly for their own use, and maybe, sometimes, also to sell it to others. Do you see that, simply keeping that information, is that an area of concern?

Prof. PALFREY: This is an area of concern to a lot of people. In fact, there's a number of scholars and lawmakers who have thought about regulating search engines and other internet service providers with respect to how long they can keep information.

The big tension here, of course, is that law enforcement would generally like these companies to keep the information as long as possible, and then be able to go back to that record, but I think that on the other side, privacy advocates would say, look, so much of our information is going online right now, and because there is the ability both to monitor what you do online, and then to search it afterwards in this big pool, but there has to be some limit.

CONAN: Mm-hmm. We got some advice earlier from Nan in terms of how to prevent your child from reaching online pornography, but on another point, Xeni Jardin, what can individual users do to maintain anonymity?

Ms. JARDIN: Well, Nan had a terrific point. And another thing that I would point out here is, look, how many times have you been downloading your email, and you get all kinds of sneaky invitations to, to sort of trick you into going to dodgy adult sites that way?

It's not just by search engines. One thing that privacy advocates will often point to, alternate browsers like FireFox, instead of Internet Explorer, give you greater privacy preference controls, and greater controls over how those website cookies are maintained. And for people who want to take the extra step, and who are a little more concerned, there are services like, a piece of software called Tor, T-O-R. You can go to, which is, basically, a system of anonamizing software. So that wherever you go on the internet, whatever you do, you are virtually anonymous. There are private companies, where you can pay a fee for anonamizing services, like, but sometimes, you know, they've been subject to criticism over poor performance issues, it slows things down. There are a lot of options out there, though.

Prof. PALFREY: Neal?

CONAN: Go ahead, professor...

Prof. PALFREY: On one of Xeni's points, which is great. Which is, I think there's a class of software that we ought to be much more worried about, which is the class of badware, or spyware, mailware, what lots of people call these nasty downloadable applications. So very often, people will download a peer-to-peer client, or they'll download a nice screensaver of a little kitty, and with it will come all these other programs that actually do collect personal information, and do bad things with it.

And, in some ways, that's the real immediate concern. There are concerns in the sort of gray area of how much the search engines and others do. But there is this other class of software that's much more dangerous, immediately.

CONAN: You can learn what search engines want to know about you, and ways you can protect your privacy online if you want to go to our website, if you trust it, Thank you both, we appreciate it.

Ms. JARDIN: Thank you, Neal.

Prof. PALFREY: Thanks a lot, Neal.

CONAN: Xeni Jardin is a tech culture journalist, and usually heard on NPR's DAY TO DAY. John Palfrey, professor and executive Director at the Berkman Center for Internet and Society at Harvard Law School. And coming up next is TALK OF THE NATION'S Opinion Page.

Copyright © 2006 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.