restrict reflect different normative choices about
the subjects targeted for filtering. Indeed, it is
developers first, and users second, who determine what gets filtered when such software is
implemented.
Although CIPA mandates the presence of filtering technology in schools and libraries receiving subsidized Internet access, it effectively delegates blocking discretion to the developers and
operators of that technology. The criteria
“obscene,” “child pornography,” and “harmful to
minors” are defined by CIPA and other existing
legislation, but strict adherence to these (vague)
legal definitions is beyond the capacity of filters
and inherently subject to the normative and technological choices made during the software
design process. Moreover, while CIPA permits
the disabling of filters for adults and, in some
instances, minors “for bona fide research or
other lawful purposes,” 29 it entrusts school and
library administrators with deactivating the filters,
giving them considerable power over access to
online content. Once FCC certification requirements have been met, it is these individuals who
shoulder the burden of ensuring access to constitutionally protected material. 30
In the single known U.S. attempt to install
filtering deeper into the network, the
Commonwealth of Pennsylvania in 2004 authorized the state attorney general (AG)’s office to
force ISPs to block Pennsylvania residents’
access to sites the AG’s office identified as child
pornography. A district court struck this regulation down on First Amendment grounds of overbreadth because the filters’ imprecision blocked
substantial lawful speech unrelated to child
pornography. 31 Since both possession and distribution of child pornography are criminal in the
United States, service providers do respond to
requests to remove it from their networks and
report it to the National Center for Missing and
Exploited Children when they encounter it.
Defamation
As in other national contexts, the potential for
legal liability for other civil violations, including
defamation and copyright, constrains the publishers of Internet content and certain service
providers in the United States and Canada.
These pressures can have a “chilling effect” on
lawful online content and conduct and can
threaten the anonymity of users. The content and
court adjudication of such laws is “state action,”
even when the lawsuits and threats are brought
by private individuals or entities.
At common law, one crucial factor in determining liability for defamation is the provider’s
relation to the content—whether the provider
functioned as a carrier, distributor, or publisher of
the defamatory content. In the United States the
common law has been overridden by a federal
statute, a holdover portion of the CDA, 47 U.S.C.
230. A key part of the CDA survived judicial
scrutiny. Section 230 immunizes ISPs for their
users’ defamation: “No provider or user of an
interactive computer service shall be treated as
the publisher or speaker of any information provided by another information content provider.” 32
Moreover, the First Amendment shields speakers
from liability for much speech about public figures. In Canada ISPs must still find their fit within
the traditional categories, where they can escape
liability if they are carriers or distributors, transmitting data without discrimination, preference,
or regard for content, or may face liability as publishers if they exercise editorial control over material. Thus, while Canadian and U.S. service
providers share the right to remove content voluntarily, those in Canada do not have the broad
discretion or protection enjoyed by those in the
United States, and may be compelled to take
down allegedly defamatory content (e.g., postings to message boards) under threat of suit.
Copyright
U.S. copyright law has also evolved more quickly—perhaps even hastily—than Canadian law in
230
Regional Overviews
the subjects targeted for filtering. Indeed, it is
developers first, and users second, who determine what gets filtered when such software is
implemented.
Although CIPA mandates the presence of filtering technology in schools and libraries receiving subsidized Internet access, it effectively delegates blocking discretion to the developers and
operators of that technology. The criteria
“obscene,” “child pornography,” and “harmful to
minors” are defined by CIPA and other existing
legislation, but strict adherence to these (vague)
legal definitions is beyond the capacity of filters
and inherently subject to the normative and technological choices made during the software
design process. Moreover, while CIPA permits
the disabling of filters for adults and, in some
instances, minors “for bona fide research or
other lawful purposes,” 29 it entrusts school and
library administrators with deactivating the filters,
giving them considerable power over access to
online content. Once FCC certification requirements have been met, it is these individuals who
shoulder the burden of ensuring access to constitutionally protected material. 30
In the single known U.S. attempt to install
filtering deeper into the network, the
Commonwealth of Pennsylvania in 2004 authorized the state attorney general (AG)’s office to
force ISPs to block Pennsylvania residents’
access to sites the AG’s office identified as child
pornography. A district court struck this regulation down on First Amendment grounds of overbreadth because the filters’ imprecision blocked
substantial lawful speech unrelated to child
pornography. 31 Since both possession and distribution of child pornography are criminal in the
United States, service providers do respond to
requests to remove it from their networks and
report it to the National Center for Missing and
Exploited Children when they encounter it.
Defamation
As in other national contexts, the potential for
legal liability for other civil violations, including
defamation and copyright, constrains the publishers of Internet content and certain service
providers in the United States and Canada.
These pressures can have a “chilling effect” on
lawful online content and conduct and can
threaten the anonymity of users. The content and
court adjudication of such laws is “state action,”
even when the lawsuits and threats are brought
by private individuals or entities.
At common law, one crucial factor in determining liability for defamation is the provider’s
relation to the content—whether the provider
functioned as a carrier, distributor, or publisher of
the defamatory content. In the United States the
common law has been overridden by a federal
statute, a holdover portion of the CDA, 47 U.S.C.
230. A key part of the CDA survived judicial
scrutiny. Section 230 immunizes ISPs for their
users’ defamation: “No provider or user of an
interactive computer service shall be treated as
the publisher or speaker of any information provided by another information content provider.” 32
Moreover, the First Amendment shields speakers
from liability for much speech about public figures. In Canada ISPs must still find their fit within
the traditional categories, where they can escape
liability if they are carriers or distributors, transmitting data without discrimination, preference,
or regard for content, or may face liability as publishers if they exercise editorial control over material. Thus, while Canadian and U.S. service
providers share the right to remove content voluntarily, those in Canada do not have the broad
discretion or protection enjoyed by those in the
United States, and may be compelled to take
down allegedly defamatory content (e.g., postings to message boards) under threat of suit.
Copyright
U.S. copyright law has also evolved more quickly—perhaps even hastily—than Canadian law in
230
Regional Overviews