Date: Thu, 09 Apr 92 11:09:20 -0700
From: Jeff Johnson <jjohnson@hpljaj.hpl.hp.com>

On Monday, April 6, I took a day off work to go to Sacramento to
testify in favor of a State Assembly bill that would, if CNID is
approved in California, require free per-line blocking as an option for
phone customers.  The bill would make per-line blocking (with per-call
unblocking) the default; phone users who want per-call blocking would
have to request it.

The bill was written by Common Cause, and was sponsored by Assemblyman
Rusty Areias (D-South San Jose).  It was up for consideration in the
Transportation and Communications committee, on which Areias sits.

The experience of testifying can only be described as extremely
frustrating, even though the Assembly members who were present clearly
opposed the telco position on CNID.

Areias' staff asked me to arrive at 1:30, when the committee hearing
was scheduled to begin.  In fact, the hearing didn't start until 2:30,
so I went out and walked around Sacramento for an hour.  When it got
started, Areias wasn't there; he was in a meeting of the Agriculture
committee meeting, which he chairs.  Therefore, his bill kept being
postponed until he arrived.  His staff kept saying that he was "on his
way" and "would be there in fifteen minutes", but in fact he didn't
arrive until about 5:45 pm.  So from 2:30 to 5:45, I stood in the
crowded hallway outside the hearing room.

While I was waiting, I spoke with the lobbyist from Common Cause.  She
asked to see my statement.  I showed it to her, and she said it was way
too long.  I told her that Areias' staff had told me to draft a
3-minute statement, but she said: "Three minutes?  Way too long.
You'll have to cut it in half.  The committee has a lot of bills to get
through today."  She said, "You don't need to explain the bill; the
members know what it does.  Cut this paragraph, that one, and that one.
 And why aren't you calling the service 'Caller ID' like everyone else
does?"  I said that part of what CPSR has been arguing all along is
that "Caller ID" is a misleading name for the service.  She said: "But
this isn't the place to clarify that.  Call it 'Caller ID'."  Her tone
was *not* that of someone making friendly suggestions.  Sigh, so I
worked on cutting down the statement.

When the bill finally came up, Areias read his supporting statement,
then about seven people gave supporting testimony.  There was a lot of
commotion in the room:  Assembly members talking to their aides,
audience members going in and out, etc.  However, it did seem to me
that at least some of them were listening to my statement (original,
3-minute version attached below).  I had been advised by Areias' staff
that I would be asked questions about the technical feasibility of
per-line blocking with per-call unblocking, but in fact I wasn't asked
any questions.

When we finished, the Chair of the committee said with a mischievous
grin: "Well, I don't suppose there is any opposition to this one, so we
can just vote, huh?"  The telco lobbyists rushed forward to testify
against the bill.  They gave their usual arguments about how per-call
blocking "balances the privacy interests of caller and callee."
However, they were continually interrupted by several committee members
who were clearly opposed to CNID.  In a sense, that was nice to see:
the telco people were really put on the hot seat, and told in no
uncertain terms that they were "nuts".  They clearly weren't prepared
for that; they gave stammering, badly expressed replies.

However, in another sense, this scene was also very frustrating,
because it was clear that the Assembly members had *little or no
understanding* of the bill.  They were trashing the telcos for the
*wrong reasons*.  About half of them (or half of those who were paying
attention at all) clearly thought they were voting Yea or Nay on CNID
itself.  So we heard all of their favorite personal anecdotes about why
CNID is bad (or good).  The other half of them confused subscribing to
CNID (i.e., receiving numbers) with indicating a blocking option.  They
seemed to think that the telcos want everyone to have (and pay for) the
CNID service by default, and that the bill would make the default be
that you don't subscribe unless you ask to.  They were really outraged
by the thought that the phone companies would dare try to force CNID on
everyone to guarantee themselves income.  No *wonder* they were trashing the
 telcos!

Even more mind-boggling was the fact that the telco lobbyists did
nothing to dispel these misunderstandings.  They simply kept repeating
their canned arguments for per-call blocking, apparently oblivious to
the fact that the assembly members weren't at a level of understanding
where blocking options were even relevant.  After a number of
exchanges, it became clear (to me) that the telco lobbyists had only a
*slightly* better understanding of the issues, and the bill, than the
Assembly members did.  They didn't distinguish between subscribing to
CNID and indicating a blocking option either!  When I realized this,
the engineer in me wanted to jump up and say: "Here,  I see how you can
dispel their concerns.  I can argue for you better than you can; I can
give the *argument* you should be giving."  Instead, I just sat there,
shaking my head.

Throughout this whole time, Areias sat there, saying nothing to clarify
the situation.  Even the state legislative analyst whose job is to
explain the bill, who I *know* understands the bill (because I have
talked to him on several occasions about CNID), kept silent.

The Assembly members then voted.  Though some were voting for CNID
itself, some against it, some for forcing everyone to subscribe to
CNID, others to give people a choice, and perhaps some for what the
bill was actually about, all votes were of course expressed only as
"yea" or "nay" for the bill.  Not surprisingly, the committee was
split, so the bill died.  Areias said nothing, and left immediately to
return to his other committee.

I left, feeling a combination of disgust and bewilderment.   Not
because the bill died, but rather because rational debate and reasoned
argument had been overwhelmed by ignorance.  Because they didn't know
what they were voting for or against.  I would have *much* preferred a
well-reasoned rejection of the bill to what actually occured.  I now
think the advice that the Common Cause lobbyist had given me -- not to
bother explaining the bill -- was wrong; that was precisely what the
assembly members needed.  Of course, it is not clear that *my*
particular explanation would have helped them.

I'm afraid the experience has left me extremely cynical about CPSR's
approach to this issue.  We spend lots of time and effort making subtle
arguments about better alternative services, problems of per-call
blocking, what the defaults should be, etc.  The lawmakers, on the
other hand, are at the level of:  "Fer it/agin' it."   I spent a day
and a half preparing my testimony, but it seems to me that I would have
probably done better to give an argument like: "Caller ID:  booga,
booga, booga!  Bad, bad, bad!"  Or maybe, "Caller ID is the devil's
handiwork!"  I'm serious.

So much for our democratic process.

Sheesh!
JJ

P.S. About the only positive thing that came out of it was that I left
copies of my prepared statement (below) and of my prepared answers to
anticipated questions with Areias staff, and these were distributed to
committee members' staffs.

---------------------------------
Testimony of Jeff Johnson
Chair, Computer Professionals for Social Responsibility
Re: California State Assembly Bill 3425
April 6, 1992

My name is Jeff Johnson.  I'm testifying today as Chair of Computer
Professionals for Social Responsibility (CPSR) a public-interest
organization that examines the social impact of technology.
Professionally, I'm a researcher at Hewlett-Packard, and formerly at US
West, a Bell Operating Company.

CPSR has reviewed Assembly Bill 3425, sponsored by Assemblyman Areias,
and supports its passage.  It is quite appropriate for the legislature
to establish minimum privacy requirements for communications services
in California.  The existing blocking law, though ambiguously worded,
is usually interpreted to require per-call blocking of number
identification.  Given that interpretation, it does not adequately
protect Californians against unintentional disclosure of personal
information, and should be strengthened.

If Calling Number ID (CNID) is approved in California with only
per-call blocking, unintentional disclosure would be common, as callers
forget -- or do not know -- to block.  Let's consider some of the ways
in which disclosure errors would occur.

If CNID is approved, phone companies would be required to inform
customers about the service and how to block, but it is naive to assume
that all phone users would be adequately warned.  Bill inserts are
often ignored, or read only by the person who pays the bill.  New
residents won't know to block when they first arrive, especially those
who don't speak English; is their privacy unimportant?  Visitors may
not know to block, and so may unwittingly add their hosts to marketing
lists.  No matter how much phone companies spend trying to inform the
public about per-call blocking, it is obvious that even years from now,
a substantial proportion of Californians won't even know that CNID
exists, much less that blocking is available.  In short, with per-call
blocking, many Californians would have their numbers disclosed by the
phone companies without their knowledge or consent, due to inadequate warning.

Even customers who have been warned will make mistakes.  As a designer
of computer systems, I know that users sometimes forget important
steps, like forgetting to save one's work before turning off the
computer.  Consider how often you call movie theaters for show times.
Forget just once to block, and bang:  you're in a national video-movie
marketing database.

If CNID is allowed in California, the way to minimize the likelihood of
disclosure errors is to make the default be per-line blocking with
per-call unblocking.  This greatly reduces the chance that phone users
will have their numbers disclosed without their consent.  It also
reduces the burden on phone companies to assure that every phone user
understands the service.  And make no mistake about it, per-line
blocking with per-call unblocking is technically feasible.

In summary, while CPSR is on-record with the California PUC as
preferring alternatives to CNID such as name identification, number
identification becomes minimally acceptable if per-line blocking with
per-call unblocking is provided as a default, with provisions for
customers to specify that they want otherwise.  For this reason, CPSR
supports Assemblyman Areias' bill.

Thank you for the opportunity to testify.  I will now be happy to
answer questions on the technical feasibility of various proposals or
on whatever topics you like.
---------------------------------------
Answers to Likely Questions
Re: California State Assembly Bill 3425
April 6, 1992

Q:      What is your assessment of the technical feasibility of
per-line blocking with per-call unblocking as a default service, and of
having the unblocking code be different from the blocking code?

A:      Per-line blocking with per-call blocking is completely feasible
for telephone companies to implement.  So is having different codes for
blocking and unblocking.  They may not want to implement these things,
but they can easily do it  if required.  The infrastructure that
enables so-called CLASS services like CNID is provided by electronic
switching systems introduced in the eighties that conform to what is
called the Signalling System 7 (SS7) standard.  Phone operating
companies purchase switching systems from switch manufacturers.  There
are currently five such manufacturers:  AT&T, Northern Telecom, NEC,
Siemens, and Eriksson.  Of those, AT&T and Northern Telecom have
between them about 90% of the U.S. market, that is, almost all SS7
switching systems in use in the U.S. -- including Calif. --  were made
by either AT&T or Northern Telecom.  I'm telling you this because I
know that the AT&T and Northern Telecom switches make it easy for local
phone companies to set up whatever sort of blocking -- with whatever
codes -- is required.  The other three brands of switches probably do
the same, but I know the situation for the most widely-used switches.
If you're interested, I can describe in laymans' terms how the services
and codes are set up in these switches.

Q:      Please do.

A:      The switching machines have tables inside them -- software
tables -- that indicate what codes do what.  There is a main table that
indicates what the default services and codes are for everyone.  Then,
for each phone line handled by the switch, there is a table that
indicates how that line's services and codes differ from those defined
in the main table.  Finally, there are settings for each call that can
override the settings for a particular line.  So, to establish per-line
blocking with per-call unblocking as a default for all customers, the
main table would be set up to issue blocking codes for all calls on all
lines, and also to recognize whatever code is to be the unblocking
code.  Then, for customers who want something other than the default,
the table for that line would be set up to indicate the desired service
for that line, e.g., per-call blocking.  The tables are explicitly
designed by the switch manufacturers to allow local phone companies
great flexibility in defining services.  Since the tables are
implemented in software, changing the services and codes they define is
easy.  Defining per-line blocking as a default, and defining different
codes for blocking vs. unblocking, is therefore no problem.

Q:      Why is per-call blocking (i.e., having to press three keys to
block disclosure of one's number) such a burden on telephone users?

A:      Though phone companies often pretend that the relevant burden
is the physical one of pressing three keys, anyone who knows anything
about interactive system design knows that the important burden is the
mental one:  having to remember to punch the keys.  If someone gave you
a new car and said, "Oh, by the way, you have to remember to press *67
on the dashboard here each time you try to stop or else the brakes
won't work," you would consider that a heavy burden, and probably
refuse to drive the car.  The maker of the car might argue:  "Hey,
what's the problem?  It's only three keys."  The three keys aren't the
issue.  The issues are:  how easy is it to forget, and what are the
consequences of forgetting?  As I said in my testimony, people often
forget important steps.  This is especially likely when they have other
things on their mind; like when you return home and have houseguests
coming and need to order that take-out food quickly.  Bang:  you're in
Roundtable's national list of pizza lovers, as well as in the lists of
everyone Roundtable sells its list to.

Q:      You predicted that if only per-line blocking is offered,
telephone numbers will often be disclosed unintentionally because not
everyone will have been adequately warned.  Would you consider phone
companies liable for such errors?

A:      I am not a lawyer and so won't presume to prejudge legal
liability.  I do think, however, that phone companies should be
prepared for numerous "wrongful disclosure" and "failure to warn"
lawsuits if they offer only per-call blocking, and they should include
the cost of fighting such suits in their financial forecasts of the
per-call blocking scenario.  They should also consider that their
customer education program would have to be much more comprehensive --
and therefore expensive -- under a per-call blocking scenario than
under a per-line blocking scenario.

Q:      As a public-interest organization, is CPSR allowed to support
particular bills, i.e., to lobby?

A:      First, IRS regulations do not completely prevent 501(c)3
organizations such as CPSR from engaging in lobbying, they simply
restrict the amount of lobbying such organizations may do to a small
amount.  CPSR does very little lobbying -- we mainly provide testimony
when asked, as here -- and so are well within the IRS limits.  Second,
IRS limits on lobbying are expressed in terms of how much of an
organization's annual budget may be spent on lobbying activities.
After all, the IRS wants to make sure that tax-deductable contributions
aren't spent promoting particular legislation.  However, I am here as a
volunteer.  No CPSR money has been spent.  Therefore, my testimony may
not even count against CPSR's lobbying quota.

