CDA Trial Transcript
4/12/96 (afternoon)
AFTERNOON SESSION
THE CLERK: All rise.
JUDGE SLOVITER: Good afternoon.
JUDGE DALZELL: Good afternoon.
JUDGE SLOVITER: I hope you are virtually, if not actually
nourished.
MR. MORRIS: Good afternoon, your Honor, I'm John Morris,
counsel for the ALA plaintiffs and somewhat out of turn, we are
presenting Dr. Albert Vezza as our next witness, he -- next and
final witness in this portion of the proceeding.
ALBERT VEZZA, Plaintiffs' Witness, Sworn.
THE CLERK: Please state and spell your name.
THE WITNESS: My name is Albert Vezza, V-E-Z-Z-A.
MR. MORRIS: Plaintiffs would at this time move that Mr.
Vezza's declaration, signed and submitted to the Court on April
9th, be accepted as his direct testimony in this case.
MR. BARON: Your Honors, good afternoon. Jason R. Baron for
the Justice Department. We do not object to Mr. Vezza's
declaration being put into evidence, subject to the caveat that I
will be asking Mr. Vezza questions regarding his technical
expertise on PICs, as opposed to his general familiarity with
PICs, during the course of my
cross-examination and his declaration will be submitted subject
to his expertise as an expert on the areas that I question him
about.
JUDGE SLOVITER: Fine. Our rulings have been consistent on
this matter.
MR. MORRIS: He is now available for examination by
defendants and the Court.
CROSS-EXAMINATION
BY MR. BARON:
Q Good afternoon, Mr. Vezza.
A Good afternoon, Mr. Baron.
Q Would you please tell the Court what the W-3 Consortium is?
A The W-3C or W-3 Consortium is a consortium of about 130-135
companies worldwide that have joined the consortium and the goal
of the consortium is to further the protocols in the standards --
the de facto standards, if you would.
JUDGE SLOVITER: Speak up, please.
JUDGE BUCKWALTER: Yes.
THE WITNESS: The de facto standards of the Worldwide WEB.
BY MR. BARON:
Q And your declaration indicates that you are chairman of the
Worldwide WEB Consortium?
A That's correct.
Q The consortium is comprised of a number of companies, correct?
A That's correct.
Q And Microsoft is part of it?
A Microsoft is a member of the consortium.
Q Netscape?
A Netscape is a member of the consortium.
Q America On Line?
A America On Line is a member of the consortium.
Q CompuServe?
A CompuServe is a member of the consortium.
Q Prodigy?
A Prodigy is a member of the consortium.
Q Could you look at Exhibit 167 in the binder that I've provided
to you and I believe binders have been provided to the Court, as
well, of defendants' exhibits, starting at 167.
JUDGE BUCKWALTER: Did you say 167?
MR. BARON: Yes, it's a smaller black binder, your Honors.
JUDGE SLOVITER: Here it is, the black.
JUDGE BUCKWALTER: Sorry, I had it upside down.
Go ahead.
BY MR. BARON:
Q At the back of Exhibit 167, there's a listing titled members
of the W-3 Consortium, it's the last three pages of the exhibit,
do you see that listing?
A Yes.
Q And you testified at your deposition last Friday that this was
a "pretty good list in terms of accuracy", correct?
A I said that it was, as of that date it was a pretty good list.
We try and keep it up to date. I can't tell you today whether
this list is accurate or not.
Q Okay. Now, you would agree, would you not, that there are
benefits to companies in joining the consortium, correct?
A There is one benefit.
Q And what is that?
A You get to sit at the table and argue about what and how the
Worldwide WEB standards should evolve. But in point of fact, we
give everything away to non-members also.
Q In fact, at your deposition last Friday, you said that the
benefit was and I quote, that's Page 18, Line 22, I'll just read
it to you, Mr. Vezza, if it's okay. "You get to sit at the table
and discuss with all of the other members, the evolution of the
protocols of the Worldwide WEB", correct?
A That's correct.
Q There are conferences and advisory meetings and workshops on
the evolution of standards and protocols, correct?
A That's correct.
Q Some of the workshop topics have included security and payment
issues, correct?
A That's correct.
Q And PICs, correct?
A That's correct.
Q We'll get to PICs in a few minutes. You've had a number of
payment workshops, including one in which VISA, MasterCard and
Microsoft attended to discuss various ideas on payment mechanisms
for credit cards, correct?
A That's correct.
Q In fact, on that specific topic, you told me at your
deposition, that there are anticipated other ways of making
payment on the net, including CyberCash, DigitCash and debit
cards, correct?
A There will likely be such ways of making payment on the WEB.
Q And work on protocols to enable those to occur is ongoing,
correct?
A That's correct.
Q And the W-3 Consortium has frequently taken the lead and/or
has made recommendations for the way certain protocols have
evolved on the WEB, correct?
A Sometimes we lead, sometimes we follow.
Q And sometimes you make recommendations?
A And sometimes we make recommendations, that is correct.
Q The W-3 Consortium is currently working on furtherance of
HTTP, the HyperText Transport Protocol, correct?
A That's correct.
Q And W-3C is working on HTML, specifically how to put in
mathematical formulas into HTML documents, correct?
A That's correct.
Q And as I indicated before, W-3C has taken the lead on PICs,
correct?
A That is correct.
Q Just as a preliminary matter, you agree, Mr. Vezza, with the
proposition that standards are at the heart of the Internet,
correct?
A Yes, they -- without standards we wouldn't have any
operability.
Q With respect to the Internet, if you don't have
predictability, if you don't have a standard, then everything
fragments, correct?
A That is correct.
Q And as a general statement subject to exceptions, you would
agree that for the Internet, you have to have a protocol that
suppliers and vendors, platforms and communication providers
adhere to, so that there can be intercommunication on the Internet
--
A Yes.
Q -- correct?
A That's correct.
Q It's also true, is it not, that the U.S. Government was
heavily involved in the creation of the Internet, correct?
A Well, let's back up a minute. I'm not sure that that's
-- they were involved, but all the people involved were technical
people. The fact of the matter is the Government funded lots of
the work, early work on the ARPA net. But it was the technical
people in the university community and in the laboratories that
did all the work of ironing out all the standards, okay. The
Government supplied the money, it did not mandate, do it this way.
Q You're familiar with the Advanced Research Project Agency?
A I am very familiar with the Advanced Research Project Agency.
Q And the acronym for that is ARPA?
A That's correct.
Q And ARPA funded the development of the ARPA net through the
70s and 80s, correct?
A That is correct.
Q And the U.S. Government has funded the IETF, correct?
A They have provided some funding to that, I believe, yes.
Q And the U.S. Government has funded the creation of the
Internet Society, correct?
A That is correct, although lots of funding for the Internet
Society now comes from private sources.
Q And the U.S. Government through the National Science
Foundation, NASA and DOD continues to fund lots of work today,
correct, on the Internet?
A It funds work associated with the Internet.
Q Indeed, the U.S. Government also supplied seed money to the W-
3 Consortium itself, correct?
A That is correct.
Q ARPA gave 1.5 million for research to M.I.T., of which 50 to
$100,000 went to the consortium, correct?
A That's correct.
Q Now, your role at W-3C is that of chairman, right?
A Yes.
Q And what are your duties?
A My duties basically are to make sure that we behave in the
manner that doesn't get us into trouble with our members. Okay.
It's very easy to do that. We are -- we -- develop
pre-competitive standards software and we try and remain neutral
to the industrial community at large. And doing that requires
some sophistication, both technically and managerially and
executive-wise. And that's my role.
Q You said at your deposition that your responsibilities would
be for what "most people would call marketing", that is, your
responsible for going out and explaining to people what W-3C does,
correct?
A That's one of the responsibilities, yes.
Q You have a technical staff at W-3C that works on PICs, right?
A Yes, I do. Not only at W-3C, but at other companies, also,
have contributed heavily to PICs.
Q And Timothy Burners Lee (ph) runs the team that does all the
technical work, correct?
A That's correct.
Q You don't consider yourself one of the technical experts on
the minute details of PICs, correct?
A That is true.
Q And you don't purport to having designed any of the exact
structures, of labels or exact structures of the syntax for
servers that's used on PICs, correct?
A That is correct.
Q And you don't claim that you could -- you could design these
structures, correct?
A I don't claim that I -- I have better staff that can do it
better than I. I use them to do that.
Q In fact, you haven't designed any software related to PICs,
that's not your function, correct?
A That is not my function.
Q And you haven't designed or constructed any software for the
purpose of labeling documents on the Worldwide WEB or the
Internet, apart from the PICs project, correct?
A Say that again, I'm sorry.
Q You haven't designed or constructed any software for the
purpose of labeling documents on the Worldwide WEB or the
Internet, apart from the PICs project?
A That is correct.
Q Out of all the documents that appear on the WEB site for PICs,
involving technical specifications, you have authored only some
"presentations", correct?
A Yes.
Q Would Defendants' Exhibit 174 be an example of one of these
presentations. If you turn to that?
A Let me clarify that. This presentation, the graphics was done
by someone at Netscape. I crafted most of the language that went
with it, with some help from the technical staff.
Q You had some input into this?
A Had a lot of input into it.
Q Now, you informed me at your deposition that the idea for PICs
came up at an early advisory committee meeting of the W-3
Consortium around December 14, 1994, correct?
A That is correct.
Q And the idea was broached at that meeting about a rating
system like the movie rating system, correct?
A That is correct.
Q And thereafter, in July of 1995, there was an article about
pornography on the Internet and many of the W-3C members started
calling up and inquiring about the idea of putting labels on
content, correct?
A That's correct, but we were working on it before that, because
we had a different idea about labeling. I mean, the PICs labeling
scheme is a general scheme.
Q You told me that, "we've had it in the hopper as some
activity", that was the statement you made at the time, correct?
A That's correct.
Q And there was a workshop in August, '95 where members of the
W-3C set out in earnest to work out on developing the PIC standard
platform for labeling content?
A Slight correction to that. There were members of the W-3C
team there, but there were also a number of representatives from
companies who were not members. And the reason they were there
was because we felt that they were a necessary ingredient in this
activity. That is, the content -- many of the content providers.
Q And it's your testimony, Mr. Vezza, is it not, that the issue
of pornography on the Internet was and I quote, "one of the major
reasons for going forward", with the PICs project, correct?
A It was one of the major reasons why our companies thought we
should go forward with it, not the reasons why we thought we
should go forward with it.
Q And you were going forward with it whether or not the July,
'95 article came out, correct?
A That's correct.
Q But the article had the effect of speeding up the time frame,
correct, for going forward?
A That's true.
Q The July article gave you a strong reason for accelerating
PICs development, correct?
A It gave us a strong reason for -- it gave the companies a
strong reason for asking us to accelerate.
Q Given the article, W-3C member companies that it would be a
good idea for the industry to come together with a labeling
scheme, correct?
A I would say that's a fair statement.
Q I'd like to turn to Page 56 of your deposition, which I've
given you. Let me read a statement from your deposition about
what PICs is and ask you whether you agree with it. It's Page 56,
Line 6. I will read the text in the six lines. "PICs is a set of
protocols which" -- this is your testimony. "PICs is a set of
protocols which allow a label -- a service to describe its
labeling system -- that's one aspect of it -- to a client or to
the recipient of a communication. It is a method by which, if all
of the vendors adhere to the PIC standard, then all of the
browsers that they develop will be able to understand a labeling
system or rating system, if you will." Was that your testimony?
A It goes on further, but I believe that that's the intent of
it.
Q But that statement is accurate?
A Mm-hmm.
Q And if you turn to Page 58, Line 9, you were talking about the
mission of PICs in your deposition and let me quote a statement
and see if you agree with it. You said, "If you're going to label
something and you want a piece of software to take action on that
label, the software has to understand what the label means. And
that's the mission of PICs. Is it to allow the software at either
a proxy server, an on line server, Internet access provider or the
actual home computer to understand the label and take the actions
whether or not to allow the content to be viewed by the viewer or
to block it." Do you agree with that statement?
A Yes, but let me read the first part of it.
Q All right.
A I think that the sentence before it begins -- is also
important -- a couple sentences. I used the term browser
understanding and I used it in generic form. By that, I mean not
only the browser, but all the software beneath it that accesses
the Internet.
Q And why is that important?
A Well, because the -- the blocking may occur either in the
browser itself or in some software underneath the browser that's
the actual access port to the Internet.
Q If you could turn to Exhibit 174, this is the presentation I
referred to earlier. It is not my intention to have you walk
through the presentation this afternoon, Mr. Vezza, I just have a
couple of questions about it. If you would turn to the page that
says, ideally, at the top.
JUDGE BUCKWALTER: Ideally?
MR. BARON: "Ideally, children use the Internet with their
parents." It's six pages in.
JUDGE BUCKWALTER: What exhibit are we on?
MR. BARON: 174. Six pages into the exhibit.
JUDGE DALZELL: Defendants' Exhibit 174.
JUDGE SLOVITER: Yes. I'm just commenting that we didn't
have this before. This is new to us, right, these exhibits?
MR. BARON: That is correct, your Honor.
JUDGE DALZELL: Yes, we haven't referred to this before?
MR. BARON: That is correct, your Honor.
JUDGE SLOVITER: We just got it?
MR. BARON: Yes.
JUDGE SLOVITER: Okay, so we haven't been through it?
MR. BARON: That's right, but this is impeachment exhibits
for purpose of cross-examination.
JUDGE SLOVITER: I understand that. It's just that --
MR. BARON: We do intend to submit this particular exhibit
with other exhibits in our case.
JUDGE SLOVITER: I'm not questioning the exhibit, I'm just
saying that I don't know about my colleagues, but we're looking at
it fresh and therefore, don't have -- I don't have anything
intelligent to ask or question, but that may not matter. It may
not be even if I had it.
MR. BARON: Well, if the Court desired at the end of the
question, we could certainly walk through the exhibit or any other
exhibits.
THE COURT: Well, that's why -- that's the only reason I
asked. But, go ahead, you conduct your examination.
MR. BARON: All right, okay.
BY MR. BARON:
Q Could you explain to the Court what this page, which says
"Ideally, children use the Internet with their parents" and the
next page were purporting to represent?
A Purports to represent a child sitting with a parent making --
having access to the Internet. And the next page says but parents
may not always be present and children may run into an
inappropriate material.
Q You told me at your deposition that with respect to the second
page, we show -- well, with respect to these two pages, we show a
child using the Internet -- this is Page 89, Line 7 of the
deposition -- "We show a child using the Internet with a parent
supervising and that's an ideal situation. Unfortunately, that
ideal situation is not always the case, very often a child is
using the Internet without a parent's supervision." Do you agree
with that statement?
A Where is that?
Q You made it in your deposition at Page 89, Line 7.
A Line 7. Yes.
Q Could you turn to a couple of pages after this in Exhibit 174.
There is a scenario, one --
A Where?
Q It's three -- three pages after the last page we referred to.
A Yes.
Q Could you tell the Court briefly what this scenario is getting
at?
A The scenario is getting at a parent using a system of Netscape
and create -- or Microsoft Explore and creating a profile for the
child, so that the child can only access materials that the parent
deemed is appropriate for the child to see.
Q If you look at the next page, could you explain what the box
that's checked block unrated sites means in the context of this
page?
A In the context of this page, the parent has decided that this
ten year old child is not -- does not have the sophistication or
the maturity to handle material that is -- might have some sexual
connotation, but that has not been rated yet and therefore, has
checked block unrated sites, so that the only material a child
will be able to have access to is that material which is
specifically rated.
Q In a PICs --
A According to the child's criteria.
Q -- in a PICs compatible format?
A In a PICs compatible format.
Q And that's one way PICs works. And PICs also works -- that's
correct, right? One way it block all --
A That's one way --
Q -- unrated sites?
A -- right. And another way would be not to check that and if
you look further on in this presentation for the 16 year old
child, the parent has chosen not to check block unrated sites,
because the parent has a feeling that the child is mature enough
to know the family's values and how to avoid that material.
Q Okay, that's all I'm going to have with that exhibit. Now,
PICs allows a content provider to place a label on the content of
a document, correct?
A That's correct.
Q And it also allows a third party to rate content, correct?
A A third party may rate content. He may give it to the content
provider or he may put it in a bureau.
Q When a content provider is rating his or her own content on
the Internet, that's referred to as self-rating, correct?
A That is correct.
Q Now, it's your view that where a PICs label is put in a
document is immaterial, correct?
A No, I went back and checked the specs after my deposition and
it appears in two places if it's put in the document. It's either
put in the HTTP header or it's put into MediTech, you're right,
Mr. Baron.
Q Let's look at Page 62 of your deposition, Line 7 and through
16. And I understand that what you've stated just now -- let me
give you a chance to read --
A Line?
Q -- that passage - from Line 7 to 16. Let me read it for the
Court. "Answer" -- well, a question, "Could you describe the
technical steps of where a label would be put?" "Answer: Where a
label is put is immaterial. All that has to happen is that when a
client reaches for a particular activity, a particular URL and
says, give me that content, that the label arrive at the client
before the content or that the -- that's not even necessary,
depending upon how the folks who build browsers decide to do it.
All that's required is that the label arrive at the Browser and
the Browser examined the label and determined whether to show that
URL -- the contents of that URL." Now, aside from the tag or
where it's put in the header and that is the rest of that passage,
correct?
A Well, the issue here is whether or not it's attached to the
document that's coming or whether there's a label bureau that
supplies the label by itself. In other words, you reach for the
document and you say, well, wait a minute, go off the Label
Bureau, get the labels for that document and bring him back. And
then get the document if the label is okay.' So, that's what I
was trying to drive at when I made those comments.
Q Okay. Now, with respect to is there content creators in the
first way, the self-writing method. The method of labeling may
vary from place to place. For example, Company A might
automatically prompt you to label a page when you start creating
it and place the label somewhere, is that correct, that's one
methodology?
A Say that -- I'm sorry, I'm missing --
Q Well, you -- you testified that there were different methods
of labeling for PICs compatible labels. And one method, that is
in your deposition, one would be where you might automatically be
prompted to label a page when you start creating it and in a
different methodology, a different institution might require the
labeling based on a set of HTML creating tools, is that correct?
A Are we talking about -- I'm missing something here -- are we
talking about creating labels when you're creating a document or
are we talking about when labels come down to be read by a browser
or the client software, to be interpreted as to whether or not to
show that software? I don't understand the question.
Q Why don't you look at Page 63, Lines 16 through 23 of your
deposition? Tell the Court what you were getting at there?
A What I was getting at there, in this instance here, the
question was "How" -- let's look at the question. I -- the way I
understood the question was how would labels be created? And what
I responded to here is that it's up to the people who create the
software, who allow you to create documents for the Worldwide WEB
or the Internet, as to how they will prompt you to create the
label.
Q Okay.
A And that's all I'm saying. And it could happen, I mean --
Q In a variety of way?
A -- in a variety of ways.
Q Right.
A And that each of the designs would be fine.
Q Okay. PICs compatible self-labeling is not in widespread use
today, correct?
A That is correct.
Q In fact, you don't personally know of anyone who has actually
self-labeled their own site with PICs, as of today, correct?
A That is correct.
Q How many different ratings systems are possible in PICs?
A Many rating systems are possible, but I don't think that that
will happen in the market place.
Q PICs doesn't rate sites on a single number or value, correct?
A That is correct. In PICs, you are able to distinguish between
literature which might have some explicit content in it or just
pornography. In fact, PICs is general enough so that you could
use the Dewey Decimal System, you could use the Library of
Congress system. You can use any kind of system you'd like to
provide a label. And that's its flexibility and its power.
Q There are innumerable numbers or values possible under PICs,
correct?
A That is correct.
Q The categories in PICs can be quite complex, correct?
A The categories in PICs are meant to differentiate subtleties
in all of the material that's on line.
Q Can you look at Defendants' Exhibit 175. It's a document that
has a title, Rating Services and Rating Systems and their machine
readable descriptions.
A Yes.
Q Do you recall my showing you this exhibit at your deposition?
A Yes.
Q And you had seen earlier versions of this document, correct?
A I have seen earlier versions of this document, that's correct.
Q Can you turn to Appendices -- well, look at A, B and C, but
start with A? It's --
JUDGE DALZELL: Appendix A to Exhibit 175?
MR. BARON: Yes, it's Page 7 of 12, at the lower
left-hand corner.
BY MR. BARON:
Q This is a hypothetical rating service called Good Clean Fun,
correct?
A That is correct.
Q It is presented to be in PICs compatible format, correct?
A That's correct.
Q And the document says this is, "one of the simplest possible
rating systems, using a single-category, minimum recommended age,
we present the machine description for a fictional service that
uses this rating system." Do you see where it says that?
A Yes.
Q And that would be one of the simplest possible rating systems,
correct?
A It's a simple rating system, that's correct.
Q You told me at your deposition that with respect to Appendix
B, that you had a passing acquaintance with the RSAC rating
service?
A Yes, I've talked to the folks there about it. I talked to our
own staff about it.
Q Could you explain for the Court what -- how this rating
service works in terms of being in the PICs compatible format?
A That the example here is in PICs compatible format, okay.
Now, I'm --
Q The number of categories.
A -- do you want me to go through line by line?
Q No, but there are a number of categories here described as
values that you didn't put in PICs, right?
A Oh, sure. There's -- well, let's look down, there's a lot of
-- one of the labels is the amount of violence.
Q Well, why don't we go down to the bottom of the page where
there is nudity and sex?
A Okay.
Q And there are label descriptions in categories, none or values
0, 1, 2, 3, correct?
A Yes.
Q And 4 on the next page?
A That's correct.
Q And they're -- different standards apply to the labels?
A And the different standards apply.
Q Okay.
A None is zero, revealing attire is number one. Partial nudity
is number two. Frontal nudity is number three. Provocative
frontal nudity is number four.
Q And you told me that with respect to Appendix C, the same
self-writing system you weren't familiar with that one?
A I'm not as familiar with it, no, I'm not.
Q But looking at it, it does look like it's more detailed than
Appendix B and A, correct? In terms of the number of values
assigned to the labels?
A Yes, in terms of the number of values, that's true.
Q With respect to -- that's all I'm going to do with that
exhibit. With respect to all of these ratings, PICs will allow a
content rater or a third party rater to rate a whole site,
correct?
A That is correct.
Q Or individual pages within a site, correct.
Q Or files.
A Or files.
A You can ave --
JUDGE SLOVITER: What was that word?
MR. BARON: Granularity.
MR. BARON: Maybe you can explain it to Judge Sloviter.
JUDGE SLOVITER: Nobody else can -- everybody else can stop
listening.
MR. BARON: And to the rest of us?
THE WITNESS: We talk about granularity when we say if you're
going to rate a site or a generic label for the whole site, then
you should apply the label to that site, which is the most
conservative label for the whole site. If you want to then go
down deeper and label pieces of that site, we say that's more
granular labeling and you can label a variety of pieces of that
site. In fact, you can label all -- every place where you can
have an entry point you can label.
JUDGE DALZELL: I mean, you could ultimately label every
single word, so says Dr. Olsen.
THE WITNESS: No, when we say label, what we mean is label
where you can get at it by a URL or some entry point.
MR. BARON: Only a select number of third party rating
systems --
JUDGE SLOVITER: Judge Sloviter thanks you.
MR. BARON: Thank you, your Honor.
BY MR. BARON:
Q Only a select number of third party rating systems out there
-- there are only a select number at present, correct?
A Yes, there are, you know, half a dozen or thereabouts that I'm
aware of, there may be more.
Q That you're personally aware of?
A That I'm personally aware of.
Q And you don't know how many sites have been rated out on the
Internet, correct?
A I do not know at this time how many sites have been rated on
the Internet.
Q It isn't very realistic that third party rating systems can
rate all of the sites on the Internet, correct?
A Depends upon really the business model the third party rating
service is going to employ. Some rating services, I anticipate,
will employ people who will go out and do ratings. And they will
probably be interested in a certain segment of the Internet which
they will rate very carefully. There will be other parties who
will make contracts with people and say, I will let you -- here is
my criteria for rating, rate yourself. And we'll sign a contract
with you and we'll do spot checks and if you -- if the -- if you
break my contract, I'll yank my rating from you. Or I may even
take you into court and sue you. There may be some penalties
associated with that. If you take that kind of a business model,
many sites can be rated. Those that -- by a third party rating
service. And there are many business models that one could
employ. And I don't know which ones are going to flourish and
which ones are going to be good. But certainly, a lot of sites
can be rated fairly rapidly under certain business models.
Q At your deposition last Friday, I asked you a question, "How
realistic is the statement" -- this is at Page 128, Line 16 --
"How realistic is the statement that, for example, the
hypothetical rating service, Good Clean Fun, rates all the
material on the Internet, how realistic is that?" "Answer: Not
realistic."
A That's Good Clean Fun. That's a specific rating system. We
didn't talk anything about a business model. Somewhere in this
deposition I also pointed out to you that how many sites could be
rated would depend upon a business model that a third party rating
service would employ.
Q You agree with Mr. Bradner, who has previously testified in
this action, that the Internet is doubling every nine months,
correct?
A I believe I stated that it's possible, but that I couldn't --
I couldn't verify it.
Q And that the Worldwide WEB is experiencing phenomenal growth?
A That's correct.
Q The W-3 Consortium doesn't keep track of how many sites are
being rated by third parties, correct?
A At the current time, we do not.
Q I asked you at your deposition, if you have 22 million URLs on
the Internet covered by Alta Vista today and a rating service has
rated 5,000 sites and you turned the blocking mechanism to the
toggle where it's blocking all unrated sites, then you have
effectively shut down 99 percent of the Internet, do you recall
that question?
A I recall that question.
Q And you answered, "That's the user's choice though, that's me
as the user, my choice to do that for my children or my choice to
let them see it."
A That is correct. The parent can decide for the child whether
or not he wants to block unrated sites. The other aspect of that
is that many of the sites that children would be interested in,
would likely be rated very quickly. The third aspect of it is
there's lots of stuff out there that children don't care about. I
have tons of stuff on the WEB site at my laboratory at M.I.T. that
would be of zero interest to children under the age of about 14 or
15. Furthermore, I could go to any of the ratings services and
say, give me a label, this is what's on it. They'd probably give
me a label for the site and we'd be done with it. I don't think
it's a big problem. I mean, I think this is a start-up problem.
Q And you understand --
A It's not the -- it's a start up problem, it's not a problem in
the long run.
Q Well, you don't dispute the notion that if there are 5,000
sites rated today and they are 22 million URLs and you're blocking
all unrelated sites in that option, you're blocking 99 percent of
the Internet, correct?
A I don't dispute that.
Q Okay, let's talk about self-rating schemes. Put aside PICs
for the moment. You would agree that it could be arranged to have
a tag or a label put on a variety of Internet applications,
correct?
A Yes.
Q There's some generic tag that's possible for UseNet, correct?
A PICs is one of those?
Q The same for UseNet postings?
A Yes.
A And a standard tag for browsers to pick up for documents on
the WEB could be created, correct?
A PICs is one of those tags.
Q And there are other tagging schemes that are possible?
A That's correct.
Q And the same for FTP, correct?
A That's correct.
Q Could be arranged?
A Yes.
Q And same for IRCs, correct?
A Yes.
Q In fact, if the attacking scheme was a simple four character
set consisting of a label or a tag. For the Worldwide WEB, let's
stay on that. The Microsofts and NetScapes of the world could
reconfigure their browsers to pick up the four character tag in
short order, correct?
A Not -- from my perspective, that's not a rational solution, it
gives you --
Q I didn't ask that.
A Okay.
Q I asked whether they could, as a technical matter --
A Was a technical matter.
Q -- was pick up the tag in short order.
A Yes.
Q And your answer is?
A Yes.
Q In some cases, it could be hours or days, correct.
A Well, I don't know. It would be depend upon where and how
they would have to modify their browser and making the change is
usually the simplest part. Testing it is usually -- takes all the
time.
Q I just have a few more questions, Mr. Vezza. If I didn't know
the address or the URL for a platform for Internet contents
selections home page, would you agree that one way to search for
the PICs home page would be to type in the word PICs?
A That's one way of doing it.
Q And you could use any search engine to that?
A You could use a variety of search engines to do that.
Q You've used various search engines like Yahoo and Alta Vista
in the past, correct?
A I've used mainly Alta Vista, but I've used Yahoo,yes.
Q You also told me at your deposition you've used WEB Crawler
once or twice, correct?
A I did not. I am not familiar with it.
Q Well, you -- you --
A Where is that?
Q At Page --
A I said I know what a WEB Crawler is, I have not used that
particular --
Q I think the -- it's Page 212, Line 14 -- Line 13, I asked,
"Question: You're familiar with WEB Crawler, though?" Your
answer: "Not particularly, no, I may have used it once or twice,
but I use hundreds of things." Do you see where I said that?
A Where is that?
Q Page 212, Line 13, 14, 15.
A Line 12 says not that one.
Q Well, I --
A Not the one I use, I use Alta Vista.
Q Line 13, "Question: You're familiar with WEB Crawler though?
A I said not particularly, no.
Q No. And then, I may have used it once or twice, but I use
hundreds of things?
A Well, I may have. I honestly don't know whether I have or
not.
Q Would you turn to Exhibit 179?
A Yes.
Q Do you recall my showing you this exhibit at your deposition?
A I sure do.
Q Do you see the Line 48 or where it's numbered 048 down -- this
-- my representation to the Court is that this was a down loaded
page, using WEB Crawler under a search-PICs. Do you see where
Platform for Internet Content selection is listed?
A Yes, I do.
Q And there are a number of other listings on this page?
A There are a number of other listings on this page.
Q That's all I'll say on that. Last question, Mr. Vezza. At
that time I deposed you last Friday, you did not recognize the
name Donna Hoffman, correct?
A That is correct.
Q Or Scott Bradner, correct?
A That is correct.
Q You asked me, could you tell me who they are, correct?
A Yes.
MR. BARON: That's it, that's all my questions.
MR. MORRIS: Just a few questions, your Honor.
JUDGE SLOVITER: Thank you.
REDIRECT EXAMINATION
BY MR. MORRIS:
Q Mr. Vezza, on the question of your expertise, you're very
familiar with the PICs specification, is that correct?
A Yes.
Q And to use Mr. Baron's terms, there perhaps are some minute
details that you may not be familiar with, is that correct?
A That is correct.
Q But, with the specification documents, both the labeling
specification documents and the rating service specification
document, you are very familiar with those?
A I'm -- I'm familiar with the -- very familiar with those.
I've been through them the last few days.
Q Let me go back. Mr. Bradner -- I mean, Mr. Baron asked you
about a four character string and you indicated that that was not
a rational approach. Was there something that you had wanted to
add?
A Yes, the problem with that is there is very little granularity
to be -- for someone to express the fact of what this material is.
I'll give you the example and which was raised by the jurists
here, that of showing something to do with AIDS and an erect
penis. You could, in fact, in PICs say this is educational.
Okay. This is educational, this is what the educational activity
is. And a parent may be able to set up or would be able to set up
a browser to discern that, so that in fact, a 15 year old would be
able to down load that information because it would specifically
say this is educational. Or in the case of the National
Geographic, all of those things. I was struck by the fact that
you kept asking the previous witness, what would you tell your
parent -- what would you tell those people how to deal with it?
Well, I mean if we have to -- if everybody has to go to an
authority to try and get a decision about how to deal with
material they want to put on the Net, we're going to have a lot of
authorities making decisions like that. Why don't we let other
folks bring in a whole group of people that can help make those
decisions. And that's what the PICs labeling system is about. It
is to allow a lot of granularity and allow people who know how to
do that to actually create the labels appropriately.
Q And the PICs specifications aren't limited to just labeling
sexually oriented material, is that correct?
A That's correct. The PICs specification are a general labeling
scheme. We looked at it and we said what we need is something
that will deal with the Dewey Decimal System, deal with the
Library of Congress. Or deal with any other kinds of labeling
schemes that we might think of in the future. And that's what
it's all about. It's going to be used for a variety of reasons.
Whether or not it's used for rating or not, it's got a life of its
own already.
Q And so, if you hypothesize that there might be a desire to
label content by its sexual -- sexual levels. And then there
might also be a desire to label content -- a different content
according to whether I dealt with racial issues or Nazism or
things like that, PICs would be able to label -- to be used in all
those situations.
A PICs would be able to be used in all those situations. We
have -- we have behind us the Internet community that the industry
that's putting software out there to use PICs, we have evangelized
this around the world. I just came back from the Far East. Jim
Miller just came back from Europe. We;re talking to all of those
folks, they're all interested in the idea of labeling. They are
all interested in using this system.
It has another advantage in that a rating system here in the
United States could rate foreign sites according to the U.S.
values. And foreign sites could rate U.S. sites according to
their values. If you -- if you go to some European countries, they
think that our violence is abhorrent. Okay, they just --they say
how can you have so much violence in your pictures and what not.
So, I think a value is local and that has to be taken into account
and PICs allows for that value to be local. Q Now, Mr. Baron
asked you a number of questions about the block unrated sites
option that the PICs specifications allow. Would this option
interfere with an adult's ability --
JUDGE SLOVITER: I would suggest you might get a little
closer or talk a little louder. Not too close, because it's --
MR. MORRIS: Thank you, your Honor.
BY MR. MORRIS:
Q Would this option, the block unrated sites option within the
PICs specification, would it interfere with an adult's ability to
access content on the Internet?
A No, I mean, the adult could change it to not to block unrated
sites and so they could access anything they wanted.
Q So, an adult would have access to the full range of content on
the Internet?
A That's correct.
Q And would the block unrated sites option mean that minors
could never, ever get access to unrated sites?
A No, in fact, if you go through the scenario, one of the -- one
of the things in the scenario, the child says I need --gets to an
unrated site, the child says I need this for my homework. The
mother comes back and gives the child permission. So, in fact,
with the child's -- with the parent's supervision, a child could
get at it.
Q And under the block unrated sites option, would a parent be
able to sit down with her children and access the Internet
together?
A Yes.
Q And in that method, they'd be able to --
A To control the situation.
Q And get access to the whole range of content on the Internet?
A That's correct.
Q Now, but with the -- if you -- if a parent were to turn the
block unrated sites option on to activate that option and to block
unrated sites, would the parent be able to protect their children
from indecent or patently offensive sites that didn't have a
label?
A That's correct, that's exactly what that option is for.
Q And this protection would protect United States citizens from
foreign sites.
A That's correct.
Q Foreign content.
Q And that's under the PICs standard, a child could be protected
form indecency and patently offensive material without any
requirement that labeling be done mandatorily across the Internet?
A That is correct.
MR. MORRIS: No further questions, your Honor.
MR. BARON: No further questions at this time.
JUDGE BUCKWALTER: Here is a question. You said PICs are
tabs?
THE WITNESS: I'm sorry?
JUDGE BUCKWALTER: Are PICs --
THE WITNESS: Are PICs tabs?
JUDGE BUCKWALTER: No, are PICs tags, I'm sorry.
THE WITNESS: Oh, no. PICs are -- PIC stands for Platform
for Internet Content Selection.
THE COURT: Right.
THE WITNESS: It's a standard by which a rating service can
express how they will rate things. And it's also a standard for
how labels should appear.
JUDGE BUCKWALTER: Yes.
THE WITNESS: Okay.
JUDGE BUCKWALTER: And what are tags, then?
THE WITNESS: Oh --
JUDGE BUCKWALTER: I thought you said --
THE WITNESS: -- that was a detail that Mr. Baron asked me
about in my deposition as to whether or not a PICs label could
appear in a Meta tag of an HTML document. And I, it was late and
I didn't remember and I didn't know and I said, no, I don't think
so. But in fact, he was right.
JUDGE BUCKWALTER: Okay. All right, I --
THE WITNESS: That it can appear there.
JUDGE BUCKWALTER: -- I didn't quite get that exchange and I
wanted to be sure I understood it. The other thing you mentioned
was about PICs will be, you said words to the effect, successful
if all vendors adhere. What's the chance of that happening?
THE WITNESS: Extremely high.
JUDGE BUCKWALTER: Mm-hmm, okay.
THE WITNESS: Netscape is going to have PICs. Microsoft has
already announced that they are releasing PICs compatible explorer
in June. I can't believe that Netscape would be far behind. In
fact, Netscape may even beat them to the punch. Certainly
Spyglass will. Spyglass -- both Netscape, Microsoft and Spyglass
helped us with these specifications. Most of the ratings services
that I know that are out there, area already, Surfwatch, are
changing to a PICs compatible spec rating scheme. So, I think the
chances are very high that --
JUDGE BUCKWALTER: Okay.
THE WITNESS: -- it will all happen.
JUDGE BUCKWALTER: Thank you.
JUDGE DALZELL: I mean, the reason is because there are
enormously powerful market forces that are driving that, are there
not?
THE WITNESS: That's correct.
JUDGE DALZELL: And you know that from your meetings of the
W-3C, right?
THE WITNESS: I know that from the meetings of the
W-3C and I know that from the meetings of the people who keep
urging us to quickly get PICs out and get it into the net.
JUDGE DALZELL: So, that they can say to parents like me, for
example, who have small children, don't worry.
THE WITNESS: That's correct.
JUDGE DALZELL: Come on line with us, your kids won't see
what's in Mr. Coppolino's book.
THE WITNESS: We hope so. I mean, AOL is there, Prodigy is
there.
JUDGE SLOVITER: We'll be forever now on this.
THE WITNESS: They're all there.
JUDGE DALZELL: Well, looking at Defendants' Exhibit 167 and
the members of the W-3 Consortium, am I correct that there is no
Government member of that? That is to say, no United States
Government member?
THE WITNESS: That's correct. They have a lot of input into
the process, but there's not really -- they're not really members.
And I have to tell you the reason. The reason is because we give
everything away. In our contracts with our members talks about
how we deal with the intellectual property that we developed. And
the Government doesn't know how to deal with the way we want to
give stuff away. And when you have several members that want to
be. We have several Government agencies, can we be a member. I
said if you can sign the standard contract, you can be a member.
But they don't want to sign the standard contract, they want to
give us a bunch of FARs (ph), which those FARs countermand the
standard contract intellectual property clauses.
JUDGE DALZELL: FARs?
THE WITNESS: Federal Acquisition Regulations.
JUDGE DALZELL: Okay. So, I am correct then and I asked Mr.
Bradner about this when he was here. You now know who Mr. Bradner
is?
THE WITNESS: I know that he testified.
JUDGE DALZELL: Okay, good. That in fact, the Worldwide WEB
is a creature completely divorced in its creation from any
government, isn't that right?
THE WITNESS: Almost true. Almost true. The fact of the
matter is Tim Burners Lee did it when he was a staff member at
Cerne (ph), which is a particle physics laboratory in Geneva,
which is funded by the European Governments.
JUDGE DALZELL: By the European governments?
THE WITNESS: European governments, that's correct.
JUDGE DALZELL: But Cerne is funded by it, but it's not an
agency of any government?
THE WITNESS: It's not an agency of a government. In fact,
there's a lot of United States participation in Cerne activities.
One of our faculty members from M.I.T. is a big player in the
Cerne activity.
JUDGE DALZELL: All right, so --
THE WITNESS: And spends a lot of time there.
JUDGE DALZELL: So, in truth, it would not be fair to say
that with respect to the Worldwide Web that Government has
"nurtured it".
THE WITNESS: I think --
JUDGE DALZELL: As in the sense of a mother nurturing a
child.
THE WITNESS: That is true of the Worldwide WEB.
JUDGE DALZELL: OKay. Now, we have been told with respect to
IP address blocking, okay. That address blocking requires that
the user's computer have stored on it's hard disc a list of all IP
addresses which should be blocked. And then it goes on to say that
the more you have on there, the more space you're using on the
disk. Ergo, the user will soon run out of disk space to block,
true or false?
THE WITNESS: I don't know what was being said by IP
blocking. Certainly PICs does not work that way.
JUDGE DALZELL: So, PICs would not use up all the disk space?
THE WITNESS: It certainly wouldn't.
JUDGE DALZELL: Why not?
THE WITNESS: Because the label could come with the document
or it could be down loaded from a service. It may be stored on a
disk. I mean, we admit that that can happen, but it's not
necessary that happened that way.
JUDGE DALZELL: When do you estimate, as practical matter,
the market will adopt PICs in a ubiquitous way.
THE WITNESS: I think that when Microsoft announces in June
or they've already announced, but when they release in June, they
will release an Explorer III with PICs compatible blocking
software. I think Netscape will be shortly behind that. And at
that point, it's been adopted.
JUDGE DALZELL: All right.
JUDGE SLOVITER: I'd like to understand how it works from the
-- again -- or mae more specifically, when you say -- when you say
that the parent can say block all unrated sites. And then in the
colloquy that took place, the analysis was that all patently
indecent sites would -- patently offensive and indecent sites
would be blocked. But they would be blocked, would they not,
along with everything else. And is there any way -- I mean, I'm
not sure how one would pick out from the rating, sites that are
patently offensive. Is that what you suggested?
THE WITNESS: No, I didn't suggest that.
JUDGE SLOVITER: I mean, is that what you were?
THE WITNESS: The question went to what happens if you check
block unrated sites? What would happen then is you would block
sites that did not come with a label or did not have a label to
associate with him. YOu would only allow information into
computers that had a label.
JUDGE SLOVITER: Okay, but all I'm suggesting is there is no
way in which that you know of or through PICS,m in which you can
identify marginal material?
THE WITNESS: Marginal material is always in the eye of the
beholder. And it's the person who's doing the rating that's going
to decide what kind of a rating they're going to place on that
material.
JUDGE SLOVITER: All I'm saying is when you block unrated
material, you block it all?
THE WITNESS: That's correct.
JUDGE SLOVITER: That's what I wasn't clear on.
THE WITNESS: Block all material that's unrated.
JUDGE SLOVITER: That's right. All unrated material.
Whether it is for children or --
THE WITNESS: It could be pictures of ducks.
JUDGE SLOVITER: That's right.
THE WITNESS: You would be blocking them, too.
JUDGE SLOVITER: No, that was my point. That's just the
point, there was in the colloquy some possible ambiguity that what
you would be blocking was only material that was unrated that
could be considered questionable?
THE WITNESS: Yeah.
JUDGE SLOVITER: And I gather that that's not what you're
testifying.
THE WITNESS: That's not what I'm testifying.
JUDGE SLOVITER: It's everything, no matter how innocuous,
would be unavailable to -- once that button or once that tag was
pushed.
THE WITNESS: That's correct. However, if it were a child
and the child wanted to see that material, he could ask his
parent.
JUDGE SLOVITER: Oh, no. I understand that.
THE WITNESS: Okay.
JUDGE SLOVITER: That's not the -- it was the technicality of
what was being asked. Okay. Does counsel have anymore?
MR. BARON: Just a couple questions, your Honor?
JUDGE SLOVITER: Sure.
RECROSS-EXAMINATION
BY MR. BARON:
Q Mr. Vezza, Judge Dalzell referred to enormously powerful
market forces that are driving adoption of PICs. If all the
browsers in the world support PICs, then self-labeling with Pics
would work, will it not?
A That's a complicated question and requires a complex answer if
you'll allow me. Self labeling, everybody labeling their own way
is going to be chaos and I think everybody will recognize that.
But self-labeling, according to some, a few, half a dozen, dozen
rating mechanisms will work.
Q And just to be clear, you were talking about how there is this
drive towards adoption of PICS and PICs compatible architecture.
That adoption of PICS compatible architecture, in terms of the
market place, is a different issue and poses different problems
than the actual rating of sites using PICs compatible
architecture, correct?
A Repeat that?
Q First there's the adoption of the architecture.
A Right.
Q But the question of rating sites on the Internet, on the
Worldwide WEB, that's a different issue?
A That's a different issue and it's going on and it will -- the
minute there are browsers out there, there will be lots of folks
that will be getting into the game. And they're different
business models will cause sites to be rated. Some sites will be
rated slowly by one business model, but more carefully. Other
sites will be rated very rapidly, but perhaps no as carefully.
MR. BARON: That's all I have.
JUDGE SLOVITER: This is a relatively unique and novel
experiment in information, isn't it?
THE WITNESS: Yes.
JUDGE SLOVITER: And with the exception of the movie rating
system which has --
THE WITNESS: It's a one rating system.
JUDGE SLOVITER: -- which doesn't do anything at all like
this because so simple in the sense to this. There's really
nothing out there, is there, that you can use as an analog.
THE WITNESS: That's correct. In fact, we looked at the
moving rating system carefully and said, gee, it will not capture
the kinds of material that we'd like to have ratings posted to.
It will not capture the essence of literature. It will not
capture the essence of art, it will not capture the essence of
education. It just won't capture those things and this is one of
the reasons why we have come up with a relatively sophisticated
system.
JUDGE SLOVITER: But to work, isn't it so that there has to
be a commitment to continue this indefinitely because there is
material that will be coming on to the Internet and Worldwide WEB
indefinitely.
THE WITNESS: That's true, but we only have -- I don't know
any other way around it.
JUDGE DALZELL: And your point, I think, well, to me or Judge
Sloviter was in a lot of sites, this issue is simply beside the
point?
THE WITNESS: That's correct.
JUDGE DALZELL: If you're in the area of microbiology, I
mean, if you were to raise this subject, I assume they'd look at
you as though you're speaking Ordu or Sanskrit?
THE WITNESS: Right, that's correct.
JUDGE DALZELL: It's totally irrelevant to them?
THE WITNESS: It's irrelevant. I mean, you know, you could
-- I deal in microbiology. I have, you know, 500 gigabytes of
microbiology information here and you'll probably get a label that
says microbiology and everybody will forget about it.
JUDGE DALZELL: And much of those 22 million sites are of
that nature?
THE WITNESS: Is that kind of information, that's correct.
That's exactly -- that's exactly correct. I mean, I have tons of
stuff at M.I.T. under my control that nobody would care about. I
mean, it's just not -- it's not relevant -- it's a rating and you
just slap a label on it and it says this is what this is and be
done with it.
JUDGE SLOVITER: Well, really maybe all material coming out
of the science labs at the University of Pennsylvania, for
example?
JUDGE DALZELL: So, that would have a label that would just
be say academic.
THE WITNESS: Well, you have to be a little careful about
that. You noticed there were some EDU sites.
JUDGE DALZELL: I sure did.
THE WITNESS: Okay, so I'm not going to -- I'm not going to
bite on that one that easily.
JUDGE SLOVITER: What you're saying is that for practical
purposes was the exception of possibilities that specific material
may begin to draw and get closer to the line, that the -- well,
you're, I guess answering my question is, that the amount of
material that needs the kind of attention that would be required
to be satisfactory is more limited than all of the material on it,
but nonetheless, there would have to be some care.
THE WITNESS: There is some gradation that has to take place
in terms of that marginal material from -- not marginal to
marginal to patently offensive.
JUDGE SLOVITER: Okay, thank you. Anymore?
JUDGE DALZELL: Anymore questions based on our questions?
JUDGE SLOVITER: No, thank you.
(Witness excused.)
JUDGE SLOVITER: Do you want to break now?
JUDGE DALZELL: Yes, it's a good time to take a break.
JUDGE SLOVITER: All right, we will break now before the next
witness. And I think counsel should be prepared to stay late. If
those -- some of you have trains, I don't know how long the next
witness will be. But if some of you -- oh, do you want to talk to
them?
JUDGE DALZELL: Yes, may I just see counsel briefly?
(Court in recess at 3:05 o'clock p.m.)
(The following occurred at 3:25 p.m.)
DEPUTY CLERK HIGGINS: Court is now in session. Please be
seated.
JUDGE DALZELL: Okay, does the Government have Dr. Olsen for
us?
MR. BARON: At this time, your Honors, we call Dr. Dan Olsen,
Jr. to the stand.
JUDGE SLOVITER: What's the timing? Until what time is he
available till, quarter after 5:00?
MR. BARON: Yes.
JUDGE SLOVITER: Remember, it's Friday afternoon, so you have
to --
MR. BARON: He's taking a taxi to the airport and 6:30 is his
plane.
JUDGE DALZELL: He'll be in good shape.
JUDGE SLOVITER: Well, you still have to be able to get
there, physically, taxi or not.
DAN OLSEN, Sworn.
DEPUTY CLERK HIGGINS: Please be seated. Please state and
spell your name.
THE WITNESS: Dan Olsen, O-L-S-E-N.
MR. BARON: At this time, your Honors, we offer Dr. Olsen's
declaration into evidence.
JUDGE SLOVITER: Is there any objection?
MR. ENNIS: Your Honors, there is an objection. We do not in
general object to Dr. Olsen's qualifications, and we understand
the Court has been generous in receiving testimony. But based on
his specific admissions during his deposition, there are five
discrete areas in which we do contest his expertise, and I would
appreciate an opportunity for a brief voir dire limited to those
deposition questions.
JUDGE DALZELL: Sure.
MR. ENNIS: First, your Honors, we contest the expertise of
Dr. Olsen regarding Surfwatch, Cyber Patrol, Net Nanny and similar
end-user blocking software.
VOIR DIRE EXAMINATION
BY MR. ENNIS:
Q Dr. Olsen, is it true that you have never installed or run
Surfwatch or Cyber Patrol or Net Nanny?
A I did not use the specific applications. My understanding of
what was done there was based on analysis of the required
algorithms.
Q Is it basically fair to say that you have read their Web
pages?
A I have read the Web pages and considered carefully what the
underlying algorithms necessary to accomplish the goals are.
Q And have you primarily read those Web pages and learned what
you've learned about them since you were retained by the
Department of Justice in this litigation?
A With regard to specific products I have considered those pages
in the past but not extensively. However, I have done some work
on exactly what the algorithms necessary to do that would be.
Q Have you read the Web pages since you were retained in mid
March of this year?
A Yes.
Q We also -- you've never conducted any independent studies of
Surfwatch, Cyber Patrol or Net Nanny, have you?
A I have not.
Q And not personally run them yourself?
A I have not.
MR. ENNIS: The second area is parental control technologies
employed by America Online, Compuserve, Prodigy and Microsoft
Network, including whether the tagging system Dr. Olsen proposes
would work in those closed communities.
BY MR. ENNIS:
Q Dr. Olsen, is it fair to say other than cursory notice of AOL
announcements recently, you have no knowledge of AOL's parental
technologies, and you have not subscribed to or used Prodigy,
Compuserve or Microsoft Network?
A That is correct.
Q Is it also fair to say that at least as of your deposition
Tuesday you did not know whether the tagging system you propose in
your declaration would work within AOL, Compuserve, Prodigy or
Microsoft Network within their closed communities?
A I do not know whether that tagging system would work.
However, I do understand some characteristics of what they do that
is relevant to my declaration.
MR. ENNIS: The third area in which we contest Dr. Olsen's
expertise is with respect to either direct or third party
verification of credit cards, including the technology of
verification, the costs of verification and the types of
transactions that would be verified.
BY MR. ENNIS:
Q Again, Dr. Olsen, with respect to that subject matter, is it
fair to say that you have just simply read the Web pages of
MasterCard and Visa since being retained in this litigation?
A I would have no technical knowledge of what MasterCard and
Visa do.
Q Is it fair that you've just read their Web pages?
A That is correct.
Q And since being retained --
A That is correct.
Q -- you've had no discussions with personnel at MasterCard or
Visa?
A I have not.
Q And I believe you just said you have no expertise in the
technology of credit card verification.
A The technology of what MasterCard and Visa themselves do, I
have looked some at verification.
Q You have no idea what they would charge to verify for
commercial transactions?
A I cannot say what they would charge.
Q And I take it you don't even know whether they would verify
for a noncommercial transaction, that is, for a speaker who wanted
to provide his speech for free?
A I would have no useful knowledge.
Q Now, with respect to third party verification systems, is it
also fair to say that you have only looked at the Web pages of
those third party verification systems since being retained in
this litigation?
A Clarify what you mean by a third party verification system.
Q Other systems that will themselves go to MasterCard or Visa
and obtain verification of credit card information, adult ID check
systems --
A Oh, adult ID as to what they -- I have read their Web pages
and I have studied some of their technical documents.
Q And that's again since being retained in this litigation?
A That is correct.
Q Have you interviewed anyone at those third party verification
systems?
A No.
Q Have you interviewed any speakers who have used those third
party verification systems?
A I have not.
Q Have you ever used those systems yourself?
A I have not.
Q Do you have any idea how many people are registered with those
systems?
A I do not.
Q Could be one, could be 1,000, could be 500, you have no idea?
A I have no idea.
Q Is it fair to say that you don't know how the "referral
technology" between a third party verification system and a
speaker would work, and you're foggy about how the third party
obtains verification; is that fair?
A No. Referral technology, please clarify that.
Q Well, would you look, please, at pages 247 to 249 of your
deposition transcript which we took last Tuesday. I'm using the
phrase you used there.
Have you located those pages?
A 247?
Q Right. And you were asked the question, "Okay, what happens
when John comes to my site?
"Answer: Okay. My understanding is that somehow, and I
would have to check the details, is that initially he not having a
password would be referred to adult check for which he would pay
whatever it was and then be referred back. How that referral
technology works, I can't say right now. I'd have to look at the
detail."
And then you say further down, "Question: Okay. Now, how
does that work?
"Answer: As I remember the technology there, IC Verify
provides -- I believe it is a PC-based system, but they may have
mentioned Unix (ph), I'm foggy there, whereby a phone line is
connected to your machine or multiple phone lines, and then
software makes the request. A call gets made to MasterCard,
whoever they're connected to or the bank, however they do that,
and a charge gets processed."
Do you recall that question and that answer?
A Yes, I do.
Q Does that -- since you used the phrase "referral technology,"
what does it mean in that context?
A Referral -- excuse me. Referral technology as I mentioned it
there on page 248 has to do with when someone arrives at a site
that would like to be verified, say by Adult Check or Validate,
they are transferred back to the Adult Check site. They do that
by means of code. They have the original people who want to be
blocked, they have to put that code into their HTML pages. Now,
the details of exactly what the syntax is without again looking at
the spec I could not quote to you, but the substantive nature of
exactly how that mechanism works, I do understand that clearly.
MR. ENNIS: All right. Now, the fourth area in which we
challenge Dr. Olsen's expertise is with respect to PICs
technology, or how cumbersome it would be to use PICs technology.
BY MR. ENNIS:
Q Dr. Olsen, is it fair to say that your knowledge of PICs has
been gained exclusively after being retained by the Department of
Justice in this litigation?
A I have read the specifications since being retained by the
Department of Justice. Most of my opinions on PICs have to do
with computer science technology in general.
Q Again, you read the downloads from the Web pages?
A Yes.
Q Since being retained in this litigation?
A That is true.
Q Is it fair to say you have no idea and have no expertise in
how cumbersome PICs would be to use?
A I have never seen PICs used.
Q Is the answer to the question then yes?
A In terms of experimental evidence, I have no idea what it
would take.
Q Would you please look at page 133 of your deposition
transcript?
A Number again, please?
Q 133. And do you see the question, "Question: How about for
PICs technology?
"Answer: For PICs technology, it is more cumbersome than
that. How cumbersome, I would be unwilling to characterize
because I think that's beyond my expertise in PICs other than that
I read the materials and it is a more cumbersome mechanism for
rating."
Do you see that question and answer?
A Yes.
Q Now, do you believe today you are qualified to testify as an
expert in how cumbersome it is to use PICs technology?
A I would not characterize myself as being able to say how
individuals or how human beings would function in trying to create
such specifications. I am perfectly comfortable with identifying
to you exactly what you would have to do in the technology, and
what has to be specified.
MR. ENNIS: A final area in which we challenge Dr. Olsen's
expertise, your Honors, is in the day-to-day operations of
libraries, including how libraries classify materials and what
libraries do.
BY MR. ENNIS:
Q Dr. Olsen, you've had no experience running a library, have
you?
A That is correct.
Q And is it fair to say that you do not have knowledge of the
day-to-day operations of an existing library?
A That is correct.
Q And that you do not have knowledge of what librarians do and
how librarians classify materials?
A That is correct.
MR. ENNIS: That concludes my voir dire, your Honors.
JUDGE SLOVITER: We will note the plaintiffs' reservations.
We will hear Dr. Olsen and the points will go to the weight of his
testimony.
MR. ENNIS: I appreciate the courtesy, your Honor. Thank
you.
CROSS-EXAMINATION
BY MR. ENNIS:
Q Dr. Olsen, for simplicity and speed, I'm going to talk first
about a category of information which I'll call for convenience
tagging speech. Your declaration reflects the opinion that it is
simple for a speaker or a content provider to add a four-character
string to the address or the title of their speech; is that
correct?
A That is correct.
Q And the purpose of adding that four-character string is to
communicate information about the content or nature of the speech
that is so tagged or labeled; is that correct?
A It could be used for that, yes.
Q The four-character string that you propose in your declaration
as an example is a dash followed by the capital letter L followed
by the numerals 18; is that correct?
A That is correct.
Q What information would that string of characters communicate
to listeners or speakers who knew nothing other than those four
characters?
A That would communicate nothing. It was not designed for that
purpose, to communicate with people. It was designed to
communicate with software.
Q And what would it communicate to software that knew nothing
other than those four characters?
A If the software had not been configured to catch it, it would
mean absolutely nothing.
Q Well, suppose it had been configured to catch it, what would
it mean to the software?
A It would depend on what the software was programmed to do. If
the software was programmed to ignore it, the program would ignore
it.
JUDGE BUCKWALTER: Dr. Olsen, could you stay back a little
bit? I'm losing some of your answers because of your puffing into
the microphone. Thank you.
BY MR. ENNIS:
Q What do you mean the phrase dash capital L --
JUDGE BUCKWALTER: What was that -- I'm sorry, I did miss the
answer to his last question, if the software --
MR. ENNIS: I'm sorry, your Honor.
THE WITNESS: The question, I believe, was if I put a -L18
and some software catches it, he said what would that mean to the
software. My answer was it would mean whatever that software was
programmed to mean.
If, for example, we were talking about a Web browser and that
Web browser was programmed to look for -L18 and this is a minor
flag was turned on in the browser, then it would refuse to ask for
that information. It's merely a code.
BY MR. ENNIS:
Q That's assuming the browser would understand -L18 to mean not
appropriate for people under 18; is that what you're saying?
A All I used, I approached this from a technical point of view
that said could I identify material that's not appropriate for
people under 18. It was my understanding the Act was concerned
with that. And I suggested one mechanism by which that could be
encoded.
Q But the four characters, -L18, could as easily mean okay for
people under 18.
A Absolutely. It depends on the convention --
Q It's a convention.
A Absolutely.
Q Does the letter L there mean legal?
A It means less than.
Q It means less than. Why is it capitalized?
A I just felt like it. I needed an example.
Q Does the -L --
JUDGE DALZELL: So it wouldn't look like a one.
THE WITNESS: That is actually true.
BY MR. ENNIS:
Q Does the -L18 string mean inappropriate for persons under 18
for any reason whatsoever?
A When I approached this problem, I approached it in response to
Mr. Bradner's testimony, which was that it was not possible to
tag. My understanding of the law was that it was important to
identify what was not appropriate for under 18-year-olds. I
fabricated a tag that could possibly represent that. That could
be programmed to be anything, that's not important.
Q So it could mean inappropriate because of violent content?
A I made no judgment about why you would want to say not for
less than L18.
Q Pardon me?
A I made no judgment as to why you would want to say less than
-- not less than --
Q So there's nothing in those characters which would indicate to
a speaker who might use them that it means inappropriate because
of sexually oriented content?
A It would not. You would need something like PICs to give more
detail.
Q All right. In other words, it would be necessary that there
develop a common understanding of what -L18 means and a consensus
among speakers worldwide and among listeners worldwide of what
-L18 means.
A You would need a consensus between speakers and the software
that was being used by listeners, yes, you would.
Q Now how long do you think it would take -- well, first, is it
fair to say you've created this -L18 concept in the last two
weeks?
A Yes.
Q How long -- and you haven't submitted it to any Internet
standards community?
A I have not. I created it purely as an example of what was
possible.
Q Do you know of any comparable convention, for example, Kidcode
(ph) has ever been submitted to any Internet standards community
organization?
A Yes. Mr. Vezza just testified that PICs had been.
Q Do you know what Kidcode is?
A No, I don't, I'm sorry.
Q Now, how long do you think it would take for the consensus you
acknowledge would have to spring up to develop around the use of
the four character -L18 string?
A We're focusing carefully on L18. Mr. Vezza has just testified
that there are market forces that would make such things available
very quickly. He testified under PICs. PICs is comparable to
what I wanted to do.
Q Well, the reason I'm asking that question is we're talking
about compliance with an Act that's in effect.
A Mm-hmm.
Q And your proposal is to use -L18. How long do you think would
it take to emerge the consensus that worldwide speakers would know
what that means and begin to use it?
A It would depend on the people who are producing Web browser
software, other client software. To the extent that they adopted
it, it could move rapidly. To the extent they didn't, it may move
much more slowly.
Q In other words, it's uncertain.
A Absolutely.
Q Do you agree that technologically and conceptually there is no
difference between using the string -L18 or using the string XXX,
triple X?
A The only difference would be, and I purposely selected L18 for
this, and that is that currently in most of the Internet commerce
I'm acquainted with, L18 does not appear. XXX does have some
societal meaning, but other than that technologically there is no
difference.
Q Technologically and conceptually.
A No.
Q And I take it technologically and conceptually there would be
no difference between using as the string the letters SEX?
A There would be a problem with SEX. The problem that would
occur there is when you select a word that has a common English
meaning, you start to interfere with the normal use of that word
in its normal meaning. So, for example, if you were screening on
the word SEX, you would screen out things from some organization
that was discussing why one should not have sex.
Q I understand that --
A So -- when you incur -- when you bring in a word that has a
common everyday meaning, then you start to get fuzzy about what
you are and are not screening for.
Q Unless the common, everyday meaning is pretty clear like X-
rated, triple X?
A Yes.
Q Well, let's stick with triple X for now.
A Yeah, that one works because it's sort of a code in our
society that's been established by makers of various kinds of
movies.
Q Do you agree that the, as you described it, simple act of
adding a three or four-character string to an address or title of
your speech which says XXX or which says -L18 would not by itself
insure that minors would not have access to that speech?
A Without filtering software on the receiving end, no, it would
not be sufficient.
Q In fact, isn't it fair to say that in order for the speaker to
have assurance that the speaker's speech is not displayed in a
manner that would be available to a minor, it's not enough that
the speaker tag the speaker's speech
-L18 or something comparable if all they did was tag and there
was no other cooperative technology, that wouldn't be enough,
correct?
A That depends on what kind of assurance you want. If you want
the kind of assurance that's necessary in a monetary transaction
where you're actually exchanging something of significant value,
that would not be enough. If you want something less than that,
which is to prevent a large number of minors from accessing a
piece of material, and as Mr. Vezza testified, you would have
almost all the browsers looking for PICs or some other labeling.
Q I'm not talking about if you have the browsers looking. As of
now I'm just talking about the simple act of tagging, self rating
my own speech, labeling it XXX or -L18, if that by itself will not
insure that my speech so tagged would not be available to minors;
correct?
A And my testimony is it depends to the extent to which browsers
catch it. If there are no browsers catching it, you have nothing.
If most of the browsers in the world catch it, you can have a high
degree of assurance.
Q Well, that's talking about what browsers might do, and we are
going to get to that. Just take this in logical order. The mere
act of tagging or labeling or self rating speech by itself is not
sufficient to insure that minors will not have access to that
speech. Correct?
A That is true.
Q So if I self rate my speech triple X, that by itself is not
going to insure that minors do not have access to it.
A I will agree with you on that with the caveat that
communication does not occur without both a speaker and a
listener. And to have ignored the listener is to have ignored
something important.
Q All right. Now, as you just started to indicate in your
answer, the tagging is not enough, you're going to have to have
some capacity to block the speech once it's been tagged, to filter
or block it; is that correct?
A That is correct.
Q And in order to do that, you're going to need to have software
that's capable of recognizing the tag and then if it's set that
way to block speech with that particular tag, correct?
A That is correct.
Q And that software could be used not to block but to find an
access speech that's so tagged, correct?
A That is also correct.
Q All right. The blocking software could be either at the
speaker's end of the communicative chain or somewhere down the
communication pipeline; is that correct?
A That is correct.
Q Now, for convenience and simplicity here, I'm going to try to
break this down. I'm going to speak first about speaker blocking
and ask you questions about that, and then I will ask some
questions about user or client blocking, blocking at the user or
client end. Is that convention understood?
A Yes.
Q The main point we have here is we're not trying to block all
speech that's tagged -L18 or triple X or whatever, correct? We
want that speech to be available to adults --
A Yes, that is correct.
Q -- but not available to minors.
A That is correct.
Q So it's not just a simple matter of having a software program
that automatically blocks all speech with that label.
A That would not be appropriate.
Q So in addition to the software, we are going to need to have
some means of verifying the ages of the listeners who are
requesting access to that speech; correct?
A If we are to verify that it doesn't go to a minor, yes, we
will need that.
Q Yes. And that will mean that with respect to blocking at the
speaker end, the speaker would have to have some way of verifying
the age of the people attempting to access that speech; correct?
A That is correct.
Q And if we're blocking at the client or user end, there would
have to be some way at that end to verify the age of the person
attempting to access the speech.
A That is correct.
Q All right. Now, first let's talk about user end blocking.
With respect to user or client blocking, do you agree that if the
speaker is relying on the client's software to do the blocking,
existing client's software and present technology is insufficient
to provide assurance to speakers that the client software is able
to block and is actually blocking?
A One more time, please?
Q Do you agree that insofar as the speaker is relying on user
blocking software, that existing technology and present technology
is insufficient to provide any assurance to the speaker that the
client's software is actually able to block and is actually
blocking?
A With current technology the only assurance you might have
would be a statistical one in the sense that if I understood X
percent were blocking, then I would have an assurance that some X
percent was not getting through.
Q Well, in particular, if you were to speak today, there is no
client browser software that would verify to you, the speaker,
that that client browser software was filtering, correct?
A That is correct.
Q There's no such browser on the shelf purchasable today.
A Does not exist.
Q So if you as the speaker tag your speech -L18 or triple X and
then you were relying on blocking at the user end to make sure
minors didn't access the speech you had so tagged, there's no
technology today that could give you assurance that your speech
would not be available to minors.
A Again I'd qualify that with assurance. For example, if you
did tag it with XXX, it's already been testified that Surfwatch,
for example, is watching for XXX. To the extent that Surfwatch is
deployed, you have that amount of assurance that it's not getting
through, but no more than that.
Q Well, let's be clear here. You could have some assurance that
anybody who is using Surfwatch should get blocked. But what I'm
asking is you could have no assurance as the speaker that the
particular home that's getting that speech and using Surfwatch or
any other -- any browser is actually running that program and is
actually blocking.
A That is true. There is no technology that will give you a
handshake that says yes, indeed, I am doing this service for you.
Your only assurance is statistical.
Q Let's turn for a moment to blocking at the speaker end. At
the speaker end, if the speaker is going to do the blocking, the
speaker is going to have to have software or have a contract with
a company that has software that can do the blocking at the
speaker end; correct?
A That is correct.
Q Is it fair to say that not all speakers have that software
today?
A Whether or not they have it, I couldn't say. Whether or not
they could obtain it readily off the shelf, there is software that
one could obtain that could assist in doing that. Whether or no
all speakers right now have access to that on their site, I don't
know what they have installed on their site.
Q Would most of the software you're talking about involve use of
a CGI script?
A Yes, that would be one way of doing it.
Q Do you have any knowledge whether the Web pages that are made
available by America Online, Compuserve, Prodigy, Microsoft, to
speakers who contract with them for those Web pages have the
capacity to provide CGI script technology?
A It is my understanding that they do not currently allow that
for reasons of security.
Q Now, we've talked a little about the software, but we began by
indicating the software is not enough either. You have to have a
separate age verification system in order to screen minors from
adults. Now, again for simplicity, let's try and break it down
and I'll ask you first what could speakers directly do themselves
if they were going to be doing the blocking and they were going to
be doing the age verification to determine who's adults and who's
minors. First about what speakers could do directly themselves,
and then I'll ask you what speakers could do through third party
age verification systems. Okay?
A We're assuming the speaker has control over the Web server?
The software is actually providing the service on the Web?
Q Yes.
A Okay.
Q And now the speaker has to determine age.
A Exactly.
Q Are there two basic ways of determining age? One is through
some sort of an ID, adult ID, or adult access code, and the other
is through credit card verification, is that fair?
A Those are the ones that I'm aware of.
Q Let's talk first about the use by the speaker of an adult ID
or access code system. If a speaker wanted to make his or her
speech available to the entire adult world, is it fair to say that
there is no existing ID or access code listing of which people in
the world are adults and which are not?
A There is no on line listing that I know of that would identify
all of the adults in the world.
Q Is there any off line listing you're aware of?
A Not that I'm aware of.
Q Now, putting aside for the moment adult -
A Excuse me.
Q Sure.
A I'm sorry. The only possibility I could think of is possibly
the Social Security Administration.
Q Okay. That's not -- that information is not available to
common speakers, correct?
A Yeah, but it is there, and I'm sure they know how old we are.
(Laughter.)
Q But what we're talking about here is speakers who are trying
to comply with the Communications Decency Act and to verify age.
There's no such listing available to those speakers.
A No.
Q Now, putting aside for the moment the use of credit cards as a
proxy for adulthood, we're really now trying to find out actual
adulthood, is there any system of which you're aware, any adult ID
registration system, any adult access code system that a speaker
could use to determine whether the listeners attempting to access
that speaker's speech are adults or not?
A Existing today I know of no adult code mechanism. That does
not mean one could be readily built, and one of the things we've
seen about the Internet is technology develops very quickly when
there is a need. But I do not know of one that exists at this
moment.
Q And if one sprang up, would it take some sort of an authority
to have control over that?
A It would depend to what extent you wanted to insure adulthood.
There are various ways that it could spring up whereby libraries
or other public institutions could verify adulthood and then all
share them in a central database. That is possible. Whether they
would not, I could not say.
Q Would you look, please, at page 224 and 225 and 226 of your
deposition?
A Okay.
Q Do you recall being asked the following question. "How do you
envision those would work?
"Answer: I guess part of the question is, it would depend on
who issued them. Some authority is going to have to issue them.
I don't know who that might be."
Do you see that?
A Yes.
Q And then further down you say "Not necessarily a Governmental
authority," do you see that?
A What line, excuse me?
Q Line 20.
A 20, okay.
Q And then line 11 on page 225, "Yeah, I don't know of one. If
that becomes a necessary part of adult access on the Internet, I
would immediately postulate the market would create such
authorities quite quickly." That's what you just testified to,
correct?
A Exactly.
Q And then you go on to page 226 to say, line 7 -- line 8. "As
far as some other way of obtaining the code, I know of no such
service." Correct?
A That is correct, other than the possibility of Verisign (ph)
and I could not testify as to the details of what they're
proposing.
Q Okay. Now, do you agree that there is nothing in current
technology that would enable a speaker to go to a news group, a
chat room, a list serve, and get back an adult access code for
everyone who was listening?
A I do not know of any such technology.
Q So again there's no way a speaker today could get assurance
through technology that only adults were in the news group, chat
room or list serve to which that adult wished to speak?
A If you want a positive verification, I know of no way. It
would only have to be statistical as I mentioned before.
Q All right. Now, in addition to the adult ID system, which
there's none in place today, another possible way for the speaker
to verify age is through the proxy of a credit card, correct?
A Yes. I would be reluctant to say there is no other possible
way. That is the only other one I know of.
Q Now, you've already testified that you don't personally know
what it would cost you as a speaker to verify credit cards, and
you haven't checked that cost with credit card companies.
A I have not.
Q And suppose you wanted to make your speech available for free.
Do you know whether credit card companies would verify those
noncommercial transactions at all?
A I have no knowledge. I would suspect they would not.
Q Let's turn to speaker verification of age through third party
systems. This time the speaker is not doing it directly. Your
declaration suggests that there are springing up some third party
systems that might do age verification --
A Such as Adult Check or Validate.
Q Adult Check or Validate. And again, what you know about those
is what you learned from their Web pages recently?
A Yes. I read their technical specs.
Q I assume there's no -- you're not aware of any adult ID list
or adult access code that's available to those third party
verification systems that's not directly available to you as the
speaker; correct?
A From my reading of it what they're doing is they're building
their own databases.
Q Based largely on the use of credit cards?
A Yes.
Q And again, since you didn't profess to know about this, you
don't know what those third party verification systems would do
with credit card information in order to verify it.
A No. I have no idea what they actually do with --
Q All right. Well, let's move on then. I'd like to shift
ground now and talk about another subject for a while, PICs, PICs
technology. You would agree, would you not, that PICs, a browser
with PICs technology, could be set to reject all unflagged speech?
A Yes.
JUDGE BUCKWALTER: All what?
MR. ENNIS: All untagged or unflagged speech. All speech
that did not carry a label that was a PICs compatible label.
JUDGE BUCKWALTER: Yes, I understand. I didn't hear, I'm
sorry.
BY MR. ENNIS:
Q Correct?
A That is correct.
Q Or you would agree, would you not, that if a parent wanted to,
a parent could set a PICs browser to prevent that parent's
children from accessing any speech at all unless A, the speech was
PICs flagged, and B, the speech was PICs rated as appropriate for
minors by a third party rating service the parent trusts.
Correct?
A With a minor caveat which I hope isn't a quibble, and that is
that there are no PICs browsers right now.
Q Well, I understand.
A But assuming there was, setting that aside, yes, a parent
could do that. A parent could set such a browser and one can
easily be built to exclude everything except what was rated
appropriately.
Q By a third party rater that the parent trusted.
A That is correct. It --
Q The parent doesn't have to trust --
A May I finish?
Q Sure, I'm sorry.
A The net result of that would be to create a kid's ghetto which
is, you would lock them out of most of the Internet, but you could
do that.
Q Well, it might be a kid's ghetto or it might not, depending on
how many third parties had rated how many sites as appropriate for
children, correct?
A That is correct. Part of my declaration is is that that would
be a problem.
Q All right. But a parent for example could say, I'm going to
set my PICs browser so that when I'm home, I'll have access to the
whole Internet, myself. But when I'm not home, I'm going to set
it so my child will not have access to anything on the Internet
unless it's A, PICs flagged, and B, rated in a PICs-compatible way
as appropriate for my child by the Boy Scouts of America, the
Christian Coalition, any third party rating service the parent
trusts.
A That is correct.
Q And the parent doesn't have to trust the rating attached to
the speech by the speaker. The parent may have no idea whether
Joe Jones, the speaker who says this material is appropriate for
kids, is rating accurately or not, right?
A It is true that by going through a third party they don't have
to trust the speaker, they can trust the third party if they
desire.
Q All right. Now, if a parent has set their PICs browser in
that way, to block all access by their children to material from
the Internet unless it's appropriately rated by a third party the
parent trusts, in that circumstance then isn't it fair to say that
the children would be protected from any inappropriate material
regardless of whether speakers flagged their speech or not, and
regardless of how they tagged their speech. Correct?
A It is true they would be protected from inappropriate
material. They would also be protected from most of the Web.
Q From any part of the Web that hadn't been rated, approved by a
third party rating system.
A Which is most of the Web.
Q But the point here, the critical point I believe we agree
upon, is that if a parent is concerned enough to deprive their
child of access to the whole Web, to set their PICs browser in
that way, then at that point there is no need for Government or
anyone else t compel the speaker to label their speech at all.
A If parents want to deprive their children of most of the Web,
they could do that. That is true.
Q Of course a parent doesn't have to set PICs technology that
way. A parent could set PICs technology to allow everything to
come to their child unless that material had been rated as
inappropriate for children by a third party rating service the
parent trusts, correct?
A That is true. They could also do that if the speakers had
accepted the responsibility and they could trust unrated speech,
then they'd have all the Web.
Q Well now, your proposal, unlike PICs, relies exclusively on
trusting the speaker to label properly, responsibly, is that fair?
A So does the PICs self-tagging scheme.
Q Yes, but --
A I would not want to characterize that as my proposal or PICs.
There are -- a primary feature of the PICs system is self rating,
and I am entirely in favor of that and supportive of that.
Q All right. The PICs also strong features and encourages third
party rating, correct?
A Right.
Q And your proposal doesn't rely on third party rating systems
at all.
A No. My declaration specifically points out problems with
third party systems.
Q Well, let me talk about some of the problems about relying
exclusively on the speaker to rate. How is a parent to know
whether potentially 40 million speakers who have rated their
speech around the world have acted responsibly?
A The same way I know whether or not my bank's advertisement is
true, because there are laws that say they cannot lie.
Q Ah. Well, speaking of laws, does your proposal make any
assumptions about whether foreign speakers would responsibly label
their speech -L18 or whatever convention is adopted?
A In doing my analysis of the problem I did not consider foreign
speech because I was not aware that Congress could legislate
foreign speech.
Q So with respect to any speech that for any reason is not
labeled appropriately in the -L18 code or whatever other
convention comes up, under your proposal that speech would reach
the home?
A That is true.
Q All speech that's foreign posted where people don't care about
United States laws, that would reach the United States also --
A Unless some other measure was taken, that is true.
Q We seem to have shifted ground here. We're now talking a
little more about your particular -L18 proposal, so let's stay
there as long as that's where we're at.
A Okay.
Q Again, a difference between your proposal and PICs is that
PICs can be imbedded in the client browser or in the actual
computer operating system, correct?
A You're talking about client side blocking?
Q Yes.
A It wouldn't really matter whether you imbedded in the -- in
terms of -L18 or in terms of PICs, they are the same in that
regard. You can imbed them in the operating system, you can embed
them in the browser. There are technical preferences as to why
you go one or the other, but it doesn't really matter.
Q Your proposal envisions that the software for recognizing the
-L18 tag and blocking it would be imbedded in the browser,
correct?
A That is one place. You could also imbed it exactly the place
Surfwatch embeds its keyword checking.
Q But your primary proposal, your recommendation is that it be
in the browser?
A My position, my technical position is I don't care. It's
irrelevant.
Q Well, doesn't your declaration say that if it's imbedded in
the operating system, that can complicate the operating system,
and that is not a good idea?
A It might. It might. But that's a technical decision. It
could work. It would not be a problem.
Q Oh, I thought your declaration said that it would be a
problem.
A If I had to make an implementation choice, that's probably
where I wouldn't put it. But it would work.
Q You'd put it in the browser.
A It would work in the operating system. It doesn't matter.
Q All right. Insofar as your proposal is imbedded in the
browser, isn't it fair to say that even if a parent went to the
trouble to buy a browser some point down the line in the future
that has the software to read your -L18 code and has that
installed, turns it on, that a minor in that home could easily go
out onto the Internet and for free download a quite different
browser that's not configured to recognize and block -L18,
correct?
A They could do that. They could download one that did not
recognize PICs. They could download one that didn't do any of
that. Yes, enterprising children could do that.
Q And if a child were to use that way of getting around your
proposal, then --
A And the PICs proposal.
Q Well, not if the PICs proposal is embedded in the operating
system.
A Well, then embed mine in the operating system, too. It
doesn't --
Q Well, all I'm trying to get at is you quibble with PICs about
whether it should be in the operating system.
A Well, let me state that for the purposes of this law,
quibbling about in the operating system or in the browser is
irrelevant. Both technologies can go both places. It doesn't
matter. And another set of enterprising teenagers could un-
install it from the operating system.
Q Have you ever tried to do that?
A To un-install stuff?
Q No, to un-install Surfwatch or Cyber Patrol or --
A No.
Q -- Net Nanny from an operating system?
A No, I haven't. However, you could reinstall the operating
system and it would effectively take it away. And I know a large
number of teenagers that can do that.
JUDGE DALZELL: That can't or can?
THE WITNESS: Can.
JUDGE BUCKWALTER: Can.
JUDGE SLOVITER: Have you seen them? I mean have you
observed that? Is that from your own --
THE WITNESS: I would not characterize them as a majority.
In fact, I would characterize them as a relatively small majority,
but almost every high school has four or five kids who really love
their computers, and yeah, they could do that.
JUDGE SLOVITER: Have you seen -- speaking of your own
knowledge, have you seen them do that?
THE WITNESS: To see them specifically install an operating
system?
JUDGE SLOVITER: Or take one out, yeah.
THE WITNESS: Yes. It's a common problem we have actually in
our labs at BYU. When they don't like the operating system we put
in there, they take it out and put their own in. It causes no end
of grief.
JUDGE SLOVITER: But are they kids or are they over 18?
THE WITNESS: These are freshmen just out of high school,
some of them --
JUDGE SLOVITER: But they're over 18?
THE WITNESS: It depends.
JUDGE BUCKWALTER: Maybe.
THE WITNESS: I guess I would characterize to you --I would
characterize to you that their knowledge is not significantly
higher when they enter our program than they had in high school in
that regard. For some kinds of machines it actually is relatively
easy.
BY MR. ENNIS:
Q Dr. Olsen, you testified that you haven't done this yourself.
Are you aware that some of the end user software that is on the
market today is specifically designed so that if a child attempts
to do exactly what you just said they could do, that the whole
system shuts down?
A If they attempt to disable it?
Q Yes.
A To the extent that they directly try to disable it, yeah, I
believe that's probably the way it works, but that's not the only
way to get at the problem. You can reinstall the operating
system.
Q You agree, do you not, that a significant number of the host
servers that are connected to the Internet are located outside of
the United States, possibly 40 percent?
A I couldn't characterize the percentage, but it's a nontrivial
number. It's a very significant percentage, yes.
Q And growing?
A And growing.
Q Let me turn to a group of miscellaneous questions for a
moment. They don't fall neatly under any category. Would you
agree that because of the ways that news groups propagate through
other news groups, that even a speaker who posted to a tagged news
group, to a news group that was tagged in its heading -L18 or XXX,
would have no assurance that his or her speech would not be
available to minors?
A Same problem we discussed before.
Q So your answer is yes.
A Your only assurance is to the extent that filtering browsers
have been deployed.
Q And in order to be safe, a speaker who wanted to post a
message to such a news group would have to either establish their
own news server, buy, own and operate their own news server, or
independently verify that everyone -- and independently verify
that everyone accessing it is an adult, that's one way, correct?
A Characterize safe for me, please?
Q In order to be safe, a person who wanted to post to a news
group --
A No, the issue is, are we talking about safe in that safe from
prosecution under the CDA?
Q They wanted to comply with the Communications Decency Act, and
they wanted to post a message which they think it's appropriate to
post, certainly to adults, but might be considered indecent for
minors.
MR. BARON: Objection. Calls for a legal conclusion.
JUDGE SLOVITER: Let's go back to the -- by this point I've
lost the question.
JUDGE BUCKWALTER: Well, I would permit it though, wouldn't
you?
JUDGE SLOVITER: Yeah, is --
JUDGE DALZELL: Is the question intelligible to you?
THE WITNESS: The issue I'm talking about is the issue of
safe.
JUDGE BUCKWALTER: Right, yes.
THE WITNESS: And if you wanted, without drawing a legal
conclusion --
JUDGE DALZELL: No, a layperson's, a layperson's --
JUDGE SLOVITER: Let's let him answer when it's rephrased so
we can all understand it.
THE WITNESS: No, it's -- if I can clarify it, it will be
fine.
MR. ENNIS: Let me rephrase the question without the word
safe in it.
THE WITNESS: I think I can clarify it, if it's fine.
JUDGE SLOVITER: Well, why don't you let him ask his question
and then you can answer --
THE WITNESS: Well, the issue is --
JUDGE SLOVITER: No, no, let him ask his question, okay?
THE WITNESS: Oh, I'm sorry.
BY MR. ENNIS:
Q Suppose I'm a speaker and I want to post a message to a news
group, and I want to be sure that my particular message is not
going to be available to anyone under 18 who has access to that
news group. I can't do that. I can't be sure that people in
existing news groups are under 18, correct?
A If the assurance is zero minors will receive this, no.
Q Now, one thing I could do is I could buy, own and operate my
own news group, correct?
A That is one way.
Q But then I would have to independently verify the age of
everyone who had access to it.
A And that would give you absolute assurance, yes.
Q Right. Another way I could do it is by posting a reference in
the news group to my speech, but posting the speech somewhere
else, for example, on a Web site that I own and control where I
also independently verify age.
A That is true.
Q Short of that though there is no way I can just speak to a
news group and have assurance that my speech will not be available
to minors.
A Again, the issue of assurance. If you're talking about
absolute perfection, no. If you're talking about some reasonable
statistical view that 90 percent of the minors cannot receive my
speech, then there are other possibilities. But if you're talking
about a contractual monetary assurance, no.
Q Well, let me ask it this way. Is it fair to say that as of
today with respect to all or the vast majority of news groups in
the country, a speaker posting a patently offensive message to a
news group has no way of insuring that that message will not be
available to persons under 18?
A Under the assumption that browsers are not filtering, that is
true.
Q Are you aware of the Internet Yellow Pages and other
directories?
A I'm aware of a number of such things, yes.
Q Are you aware of any of those directories that at the present
time list speech according to what's appropriate for people under
18 and speech which is only appropriate for people over 18?
A I have not seen such a category, no.
Q Now, speaking not as a legal matter, I'm not asking for a
legal conclusion under the Communications Decency Act, but if you
were a speaker and you thought about posting a centerfold from
Playboy Magazine, would you personally think that that image might
be indecent or patently offensive for persons under 18?
A You're asking me for a conclusion about the decency of a
particular piece of --
Q No, I'm asking for your personal opinion, would you personally
think that that image might be indecent or patently offensive for
persons under 18?
A So you're asking me for a conclusion as to patently offensive?
Q You personally.
MR. BARON: Objection. It's beyond the scope.
JUDGE SLOVITER: What? Excuse me?
MR. BARON: It's beyond the scope of his direct examination.
JUDGE SLOVITER: Okay. A lot of the witnesses have been so
asked, you know, from both sides, but go ahead. It is beyond the
--
JUDGE DALZELL: But so what?
MR. BARON: We're presenting Dr. Olsen as a technical expert.
JUDGE BUCKWALTER: We understand, but --
MR. BARON: Not as a speech witness.
JUDGE BUCKWALTER: We'll let him answer that question.
JUDGE SLOVITER: We're going to let him answer it.
THE WITNESS: Okay. If we consider the local community which
consists of Dan, Dan would be offended.
BY MR. ENNIS:
Q And how about the seven dirty words?
A Dan would be offended.
JUDGE BUCKWALTER: Who's Dan?
JUDGE SLOVITER: Who's Dan?
THE WITNESS: That's me, I'm sorry. That's me. (Laughter.)
JUDGE DALZELL: Oh, he's the community. He is an expert on
what would offend him.
THE WITNESS: That's a relatively small community, but it's
the one I know best. (Laughter.)
MR. ENNIS: May I take one moment, please?
JUDGE SLOVITER: Yes.
(Pause in proceedings.)
JUDGE SLOVITER: I should have waited for you to ask him.
JUDGE BUCKWALTER: I thought it was an acronym. (Laughter.)
JUDGE DALZELL: I can't imagine why you would think that.
MR. ENNIS: Your Honors, we have attempted to divide cross-
examination between us, and I know Mr. Hansen has some questions
in areas that I have lightly touched upon but didn't explore in
depth, so I think I have no further questions at this time.
MR. HANSEN: Good afternoon, your Honors. My name is
Christopher Hansen, one of the lawyers representing the ACLU
plaintiffs. Mr. Ennis is modest. He asked most the questions I
wanted to ask, so I have relatively few to ask the witness. I'd
like --
JUDGE SLOVITER: The witness won't cry about that.
CROSS-EXAMINATION
BY MR. HANSEN:
Q Dr. Olsen, I'd like to bring this from the abstract down to
the concrete. I'd like you to assume a hypothetical situation for
me, if you would. I'd like you to assume first that I am a
nonprofit organization, noncommercial organization, that I run a
Web site, that I would like to get the information on my Web site
to the maximum number of people I possibly can, and I currently
offer the information for free to everybody, and that this Act has
gone into effect as of 6:00 p.m. this evening and I don't want to
go to jail. I don't want to even risk going to jail.
I'd like to now talk about what it is I might have to do in
order to make sure that the Government doesn't start investigating
me at 6:15 this evening. Now, first what I have to do is I have
to tag my speech under your proposal, correct?
MR. BARON: Objection. I just want to clarify that this is a
hypothetical, that as the Court is well aware the CDA is not being
enforced pending the decision of this Court and --
JUDGE DALZELL: That hasn't eluded our attention.
(Laughter.)
MR. BARON: Thank you, your Honor.
BY MR. HANSEN:
Q The first thing I would have to do is tag my speech under your
proposal, correct?
A Correct.
Q Now, I want you to assume that the name of my Web site is the
Electronic Frontier Foundation, and I want you to assume I have
14,000 separate files in my Web site, some of which might be at
risk under this statute and some of which might not. The first
thing I'm going to have to do is go to each of the 14,000 files
and decide whether to tag each one indecent or not indecent,
correct?
A Whether or not you would actually have to look at all of them
is actually problematic. I assume that you've categorized them in
some way and you could recognize quickly that there are large
categories you don't need to look at. But I will grant you that
there is some labor involved in doing the tagging. I would not
characterize that that label means you looked at every one of
those 14,000 pages.
Q 14,000 files. A file can be a lot more than one page,
correct?
A Yes. For clarity, when I refer to a page it means a file, and
I will use them interchangeably. I'm sorry.
Q Now, I want you to assume that some of my files are in fact
multiple pages and within that file some of the material is
clearly decent by anybody's standard, and some is potentially
indecent under somebody's standard. Then I have to not only go to
my 14,000 files, but I have to go below my 14,000 files and decide
what parts of those files to tag and what parts not to tag,
correct?
A Not necessarily. It depends on how much of a compelling
interest you have in making sure that all of those actually go to
children. You could, if you wanted, and assuming tagging browsers
were deployed, you could immediately tag your entire site as being
inappropriate for minors and then over time work you way down
exposing pieces as you felt it was important to get them out to
minors.
JUDGE SLOVITER: Excuse me. Suppose it's a large museum and
they have -- I mean I don't know if that's within the scope of
what you're asking, Mr. Hansen.
MR. HANSEN: It is, your Honor.
JUDGE SLOVITER: Suppose it was a large museum, a really big
one with a lot of different pieces in the museum from different
cultures all over. Does that make it -- that's the only way I can
sort of try to think about this.
THE WITNESS: Right. The issue here --
JUDGE SLOVITER: And I didn't mean to stop you in your --
THE WITNESS: No. Please forgive me for trying to
characterize what the counsel is driving at. There is an issue of
what it would take to get started, in other words, this transition
period when the Act goes into effect. And then there is the issue
of what would I have to do on an ongoing basis over time. So if I
have a large electronic museum, of which at this point there are
relatively few, electronic museums.
JUDGE SLOVITER: All right, but it could be another kind of
museum. It could be a museum of paintings which one might --
THE WITNESS: Your Honor, the reason I characterize a
difference is because if you had a large museum of paintings which
was not in electronic form and you wanted to put them on the Web,
there is a process by which you would have to go through to put
them on the Web. So as you put them on, I presume you would tag
them. And you already are going through a process of paging
through each one of them, so you only incurred a small additional
cost of evaluating them. That's different than I already have a
huge electronic library and I have to go back and sort it all out.
JUDGE DALZELL: Which you would agree is a big task, or
potentially a big task?
THE WITNESS: Yeah. If you didn't have a characterization
that would help you narrow that down, it could be a big task.
Many libraries have characterizations that can help them, but I
couldn't, as Mr. Ennis has pointed out --
JUDGE DALZELL: You mention that in your declaration, the
Library of Congress system.
THE WITNESS: There are ways to use classifications to get
closer. Whether they would work in all cases I couldn't say.
JUDGE SLOVITER: Well, I'm sorry, I interrupted you, but I
was trying to make it concrete for me, not necessarily for him.
BY MR. HANSEN:
Q And if I'm trying to avoid going to jail by 6:15, you suggest
one of my options is I tag my whole site -L18, thus preventing
minors from getting access even to those portions of my site that
I know are perfectly acceptable to minors. That's one of the
short-term solutions I have; is that right?
A If you were going to do it by 6:15 tonight, I cannot conceive
of another solution. There's not time to do anything.
JUDGE DALZELL: Except label the whole thing -L18.
JUDGE BUCKWALTER: That's what he said, yeah.
THE WITNESS: Exactly, exactly, yes.
BY MR. HANSEN:
Q Well, you proposed one other alternative in your deposition
which is I could shut my entire site down for a month or two and
reconfigure the whole thing, correct?
A Yes. However, as we've talked about, just labeling the entire
site would be a little less Draconian and is easily possible.
Q But would prevent all minors from accessing all of my
material, even that which is unquestionably decent?
A That is true.
Q Now, there are some kinds of files that you can't tag because
tagging would destroy the file itself; correct?
A No, that is not correct.
Q So you disagree with Mr. Bradner's position on that?
A Mr. Bradner -- I do not disagree directly with Mr. Bradner's
position. Having read through what Mr. Bradner said, Mr. Bradner
said you could not change the contents of those files because it
would destroy the format. In that regard Mr. Bradner is correct.
However, the proposal which I put in which was in direct
response to Mr. Bradner's declaration was that you could change
the name and that would not damage the file. I specifically, in
designing the -L18 counterproposal, if you will, I specifically
looked at that because there are things that you can't change the
contents, but by changing the name you have tagged.
Q Well, I'd like you to look at your deposition, if you would,
at page 256. I'm sorry, it's 255. My Xerox is unclear.
A Which line?
Q 12. The question you were asked was, "I understood Mr.
Bradner to suggest there were some files you couldn't embed a tag
in, that you would simply destroy the integrity of the file, is
that accurate?"
And your answer was "Yes."
"Question: And what kinds of files are those?
"Answer: A GIF, G-I-F, image."
Was that your testimony?
A Yes, except the proposal that is stated in the declaration
does not change the file. It changes the name.
Q And if the -- okay, I understand.
A So I did not directly contradict Mr. Bradner in that regard.
He is absolutely accurate. You can't go in the file and mess with
some of them.
JUDGE DALZELL: And a GIF again is what?
THE WITNESS: It stands for graphical interchange format.
The example would be the same if it was JPEG (ph) which is another
graphics format, MPEG (ph) which is a movie format. Mr. Bradner
testified about some data files that he had. It would probably
not be appropriate to damage the data. That's why I specifically
went for the -L18 proposal, because you could change the name
without actually having to modify the data and damage the
integrity of what he's doing.
BY MR. HANSEN:
Q But the only way that would work is if you tagged the entire
file with the same tag. And so if it were one of those files we
were talking about before, part of which was unquestionably decent
and part of which was questionable, you'd still have to tag the
whole thing indecent for your proposal to work, correct?
A No. If we take for example --
Q Correct?
A Well, let me -- I'll get there.
Q All right.
A If you take the GIF image, for example, that is true. You
would have a hard time unless you did something to break the image
apart to put the tags in because there is no mechanism for
embedding. If it was HTML, you could embed a tag inside as Mr.
Vezza has already testified.
Q Now, I want you to assume hypothetically that the name of my
Web site is CDT, Center for Democracy and Technology, and that I
add on to my Web site the equivalent of hundreds of typed pages of
material a week. Every time I did that from now until eternity
under your proposal I would have to read all of those -- and
assume I'm not the author of all of those pages. Under your
proposal every time I did that I would have to read it and decide
whether to tag it or no and decide whether to tag subparts of it
or not; is that correct?
A If there's going to be a label associated with it, somebody
will have to read it. It will either be you or a label bureau or
a parent, but somebody will have to make the judgment.
Q But under your proposal it's me that has to make the judgment,
right?
A That is correct. Assuming you are the speaker.
Q And you said at your deposition that if I want to make the
right kind of judgment under the law, I might well have to hire
and consult a lawyer in making these judgments, correct?
A Or you could be conservative about it and just classify it for
adults.
Q Now, even -- I want you to next assume that my Web site is
called Critical Path AIDS Project, and on my Web site I have
hundreds of links to other sites. Under your proposal, how does
that work?
A Okay. If you were using -L18, the links to the other sites,
there were actually several things we talked about in the
deposition. One was that the other site is friendly with you, and
they are going to tag their stuff in some regular way in which
case a small program could be written that would go through and
fix your links in the same regular way.
Q So I'd have to get on the phone or get on E-mail with all of
these hundred