[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Final text: Software developers comments on the WIPO database treaty (fwd)
Final text of the Software Developers letter on WIPO treaty.
jamie
---------- Forwarded message ----------
Date: Fri, 22 Nov 1996 04:04:14 -0500 (EST)
From: Gordon Irlam <gordoni@base.com>
Gordon Irlam
2310 Rock St.,
Mountain View CA 94043
(415) 903 0811
Commissioner of Patents and Trademarks
Box 4, Patent and Trademark Office
Washington, D.C. 20231
Attention:
Ms. Carmen Guzman Lowrey
Associate Commissioner for Governmental and International Affairs
November 21, 1996
Dear Ms. Lowrey,
We are responding to the request published in the Federal Register of
October 17, 1996 for Comments on the Chairman's Text of the WIPO
Diplomatic Conference on Certain Copyright and Neighboring Rights
Questions.
These comments are directed at the Chairman's Text of the Basic
Proposal for the Substantive Provisions of the Treaty on Intellectual
Property in Respect of Databases. We do not seek to offer comments on
any of the other proposals on the table at this conference.
As computer industry business leaders and as software developers, we
are at the forefront of the National Information Infrastructure (NII)
and the emerging Global Information Infrastructure (GII), and as such
hope we may be able to offer a number of insights into some of the
possible effects of this treaty on the NII and GII.
We believe that the Internet represents the foundations of the GII,
and as such we look to the Internet to assess the effects of the
database treaty on the GII.
(1) Effects of the treaty on the Internet routing infrastructure
At the lowest level, the computers forming the Internet,
communicate through the exchange of packets of information. These
packets pass from one system to the next until they reach their
destination. Typically a packet needs to pass through around 20
different systems before it reaches its final destination.
Analogous to the postal system, each packet has an address printed
on it that identifies the final destination. These addresses are
effectively 9-digit numbers, and are known as IP addresses.
Unlike the postal system these addresses are not defined
geographically. Instead, a particular IP address might be located
anywhere in the world. Intermediate systems are aware of the
topology of the Internet, and know where the computer with a
particular IP address is located. Each intermediate system knows
who to forward a packet to next to ensure it reaches its final
destination.
Intermediate systems know where each IP address is located through
the exchange of what is termed routing information. The systems
exchange information identifying where each IP address is located.
Computers send messages to connected systems to saying: "I am
here", and the connected systems then collect up and send on this
information to other systems on the Internet. This information is
sent in the form of routing tables listing a set of IP addresses,
and for each IP address listing its current location.
These routing tables fall under the domain of the proposed
database treaty. As such, a network provider would be able to
claim ownership of the routing table constructed from the routing
information provided by its subscribers. Doing this would allow
the network provider to prevent, or control the way in which
others can make use of such routing information. Apart from the
network provider, there is no other practical method of obtaining
this routing information. Such routing information is essential
for network connectivity, and, as such, could be used as a price
lever by large network connectivity providers (PSI, UUNet, MCI,
Sprint) to force smaller providers out of the marketplace. An
example end-user license for the use of such routing tables might,
for instance, prohibit the exchange of the information contained
in the routing table with anyone other than a customer of one of
the large network providers.
(2) Effects of the treaty on the Internet domain name system
Users on the Internet aren't required to know the IP address of
the hosts to whom they are talking. Instead, they refer to
computers through textual names, such as "www.uspto.gov". The
Domain Name System is the component of the Internet responsible
for translating these textual host names into IP addresses, to
which packets can then be sent.
The domain name system is administered by Network Solutions
Incorporated under a time limited contract with the NSF. Parties
contact Network Solutions telling them the names of new machines,
and Network Solutions publishes a database on the Internet that in
effect contains the IP address of each machine.
The domain name system meets the definition of a database given as
specified by the treaty. No practical way exists to obtain the
information contained in this database other than either directly
or indirectly through the data supplied by Network Solutions.
The entire data contained in the DNS database is made available by
Network Solutions to 10 or 20 specialized systems known as root
name servers that are owned and operated by independent third
parties. These root name servers are then contacted by the users
of the Internet whenever they need to translate a host name into
an IP address, so that they can then contact a particular named
host.
Since it was drafted prior to the treaty, it isn't clear from the
Network Solutions's contract with the NSF, whether Network
Solutions would be able to claim ownership of the DNS database, or
if the database belongs to the NSF. It seems plausible that
Network Solutions would be able to claim ownership. We are
however unable to make this claim with certainty.
The ability to claim ownership of databases such as the DNS
database could have a potentially severe chilling effect on the
Internet. Network Solutions position as sole provider of the DNS
database would enable them to charge a high price for access to
it. This information contained in this database is fundamental to
a user's ability to navigate around the Internet, and essentially
Network Solutions would end up owning the Internet.
It seems vital to us that in evaluating the effects of this
treaty, possible serious real world ramifications such as this
must be very carefully analyzed and understood.
(3) Effects of the treaty on Internet search services
There are a number of Internet search services which users use to
find information on the Internet. Examples of popular search
services today include Alta Vista, Excite, Infoseek, and Verity.
Information on the Internet is contained in individually owned and
managed repositories called web sites. Each web site contains
numerous documents.
The search services work by visiting each web site on the
Internet, retrieving the site's contents, and from the contents
constructing an index of the information contained on the web
site. To perform a search, a user types in a few key words. The
search service is then able to use its indices to return to the
user a list of all web sites containing the specified words. The
search services do not return the contents of the indexed web
sites, but merely tell the user the names of the relevant web
sites.
The generation of an index for a web site currently does not fall
under the scope of the Copyright Act. Under the proposed database
treaty, each web site on the Internet fully meets the definition
of a database. And, under this treaty the ability to construct an
index of a web site, which of necessity involves retrieving the
information from the web site would appear to fall under the
"right to authorize or prohibit the extraction" terms of the
proposed treaty.
It is impractical for search services to gain permission to index
every web site on the net, and as a result the only options would
be for search services to either flout the law, hoping that their
infringements won't be prosecuted, or to cease providing this
valuable service to the users of the Internet.
Some might believe that the uses of data that we describe here may not
be prohibited by the treaty. They may imagine the extraction is not
substantial, or they may think the extraction is not significant to
the value of the database. Such beliefs are wrong. The extraction
that occurred here is prohibited by the treaty:
(1) Extraction of "all" of the database occurred which would be prohibited
In the three situations considered, it was not a portion of the
database that was extracted but the entire database. Routing
involves the entire routing tables being transfered from one
system to another. DNS involves the transfer of the entire DNS
database from Network Solutions to the root name servers.
Searching requires the transfer of the entire contents of a web
site to the search system for indexing.
Article 3 of the treaty prohibits the unauthorized extraction or
utilization of a database. Extraction is defined as:
"the permanent or temporary transfer of all or a substantial
part of the contents of a database to another medium by any
means or in any form"
Thus, since the contents of the entire database have been
transfered, whether the contents also meet the definition of "a
substantial part" is moot. The treaty already prohibits the
extraction which has taken place.
(2) The effects of extraction would be "substantial" and thus prohibited
The treaty prohibits the extraction of the entire database. Even
if extraction of the entire database was not prohibited, or even
if the extraction of the entire database had not taken place, the
effects of the transfer would still be considered "substantial",
and thus prohibited by the treaty. The treaty defines "a
substantial part" of the contents of a database as:
"any portion of the database, including an accumulation of
small portions, that is of qualitative or quantitative
significance to the value of the database"
In assessing the significance of the transfers that have taken
place on the value of the database, we need to do so with respect
to the value of the database once this treaty has been enacted,
and not with respect to the value of the database today. Today,
the market value of the DNS database is negligible. Multiple root
name servers provide access to the contents of the DNS database to
anyone on the Internet. The root name servers do not charge for
access to the DNS database, and if they attempted to do so they
would not be successful. Users would simply contact other root
name servers willing to provide the same information for free.
Following the enactment of this treaty, and possibly following the
expiration of its contract with the NSF, Network Solutions could
set up its own root name servers, access to which required the
payment of a fee. Network Solutions could then claim that the
independently owned and operated root name servers through their
transfer and subsequent provision of free access to the DNS
database, had a "significant" negative impact on the value of the
DNS database. Thus the transfer of the DNS database from Network
Solutions to the independent root name servers would under the
terms of the treaty be considered "substantial". This would allow
Network Solutions to prohibit the independent root name servers
from obtaining copies of this information. This would leave
Network Solutions as the sole provider of DNS information, and
enabling it to charge a substantial fee for access to this
information. The significance of the damage this would cause can
not be understated.
The same considerations apply to the routing and searching
databases. Today, the value of the originating databases that
provide this information are insignificant. Following the
enactment of this treaty, the value of these databases would
become significant because the owners of these databases could
then use them to establish a monopoly. In the case of searching
this might involve mandating the use of a web site specific search
interface for searching a particular web site, and for which a fee
is charged. The effect of the existing Internet-wide search
services on the ability of a site to charge for a site specific
search service would be "significant". The indexing of such sites
by the Internet-wide search services would thus involve the
extraction of a "substantial" part of the database. Under the
terms of the treaty the operation of the Internet-wide search
services would thus be prohibited.
In the case of routing, large network connectivity providers may
institute a prohibition on the redistribution of the contents of
the routing tables by smaller providers without the payment of a
fee. Routing tables are essential to the operation of the
Internet. As such, re-distribution of them has a "significant"
negative effect on their value. This means any re-distribution of
them would be considered "substantial", and able to be prohibited.
We should emphasize that the treaty prohibits the extraction of
the entire database. As a result, each of the situations we are
considered here is already prohibited by the treaty. What we are
doing here is showing that even if the treaty did not prohibit the
extraction of the entire database, but instead only prohibited
actions that had a significant effect on the value of a database,
the harmful effects of the treaty would be unchanged.
We hope these comments help highlight some of the serious issues
raised by the proposed database treaty. We believe the treaty has
been developed through abstract theoretical considerations, and that
many of the practical consequences of this treaty have not been
carefully considered.
If recent U.S. court decisions make it necessary to make changes in
database protection law, we believe this should be achieved through
carefully considered legislative changes to the U.S. Copyright Act.
No form of Sui Generis protection is desirable, and especially not one
achieved via an international treaty. We feel it will only be
appropriate to consider an international treaty defining the scope of
protection for databases if and when an acceptable set of changes have
been made to the U.S. Copyright Act.
The WIPO database treaty is based on an inadequate understanding of
the scope of its applicability and its resulting effects. We find the
treaty to be ill-considered and inappropriate. In view of the effect
it would have on the Internet and the future U.S. computing
infrastructure, we urge the U.S. work to have this treaty removed from
further consideration at the WIPO conference.
Thank you for considering our opinions.
Respectfully,
Gordon Irlam
Internet Software Developer -- Mountain View, California
L. Peter Deutsch
ACM Fellow
ACM Software System Award co-recipient
President, Aladdin Enterprises -- Menlo Park, California
Keith D. Lauver
President, MicroAssist, Inc. -- Northfield, Minnesota
Michael Cohen
President, Vermont Business Visions, Inc.
-- North Springfield, Vermont
Luther Welsh, Jr.
President, NDA Technologies, Inc. -- Tustin, California
Michael Tiemann
Director, Cygnus Support -- Mountain View, California
Earl Green
President, Crystal Wind Communications, Inc.
-- Crystal River, Florida
Steve Shannon
President, Coffeehaus Networks, Corp. -- Boston, Massachusetts
Brian Bartholomew
President, Working Version -- Cambridge, Massachusetts
Rodney H. Kay
President, Kay Communications -- Tampa, Florida
Dean Anderson
President, Plain Aviation, Inc. -- Nashua, New Hampshire
John Gilmore
Entrepreneur -- San Francisco, California
Christopher H. Wysocki
VP & CTO Data Life Associates, Inc. -- Verona, New Jersey
Philip Peake
CEO, Newberg Webcenter -- Newberg, Oregon
John R. Levine, Ph.D.
Co-author of "The Internet for Dummies" and other books
Internet lecturer/consultant and owner, I.E.C.C.
-- Trumansburg, N.Y.
Tim Pearson
Author of "Internet Marketing", published by John Wiley & Sons
Director, FutureLink -- San Francisco, California
F. E. Potts
Author of "F. E. Potts' Guide to Bush Flying", ACS Publishing
Recipient Aviation/Space Writers "Award of Excellence"
Owner, The Web Outpost -- Tucson, Arizona
Gail McGowan Mellor
Owner, Earth Sciences/Writer's Inc. -- Louisville, Kentucky
Dan Connolly
Research Scientist -- Cambridge, Massachusetts
W. Curtiss Priest
Director, Center for Information, Technology & Society
-- Melrose, Massachusetts
Laura Kusumoto
Director of Production, LVL Interactive
-- Santa Clara, California
Phil Karn
Software Engineer -- San Diego, California
Bodo Parady
Staff Engineer, Sun Microsystems -- Menlo Park, California
Blaise Pabon
Network Analyst -- Mountain View, California
Fen Labalme
Software Architect -- San Francisco, California
Joseph Kiniry
Graduate Student & Consultant -- Pasadena, California
David S. Miller
Software Engineer -- East Brunswick, New Jersey
James Salsman
Systems Analyst -- Mountain View, California
Lars Poulsen
Engineering Manager -- Goleta, California
Kathy Mitchell
Software Engineer -- Mountain View, California
Greg Merrell
Senior Network Consultant -- Cupertino, California
Robert Gunn
Computer Consultant -- Houston, Texas
Elizabeth Yow
-- Norfolk, Virginia
Verna Desatoff
-- Fort Collins, Colorado
Joshua M. Paulson
Christian Greuel
Computer Artist -- Mountain View, California
Todd Clayton
System Software Engineer -- San Francisco, California
Eric Titmas
Senior Staff Engineer -- Mesa, Arizona
Yuri Sorkin
Independent Mathematician -- Los Angeles, California
Jonathan Ryshpan
Software Project Manager -- Oakland, California
Greg Limes
Software Engineer -- Mountain View, California
Frank Arnold
Webmaster, Catholic Charities -- San Jose, California
Marla Pereira
Software Engineer -- California
Adam Rifkin
Graduate Student -- Pasadena, California
Timothy G. Lang
Programming Director, Community Access Center
-- Kalamazoo, Michigan
Ray Baco
Technical Support Supervisor -- Durham, North Carolina
William Arnold, P.E.
Principal Consultant, Pythia Corporation, Inc.
-- Indianapolis, Indiana
Jack Buchanan, MSEE, MD
Engineer and Physician -- Memphis, Tennessee
Dan Kegel
Software Engineer -- California
Stanley T. Shebs, Ph.D.
Software Manager -- Mountain View, California
Ed Tynan
Senior Program Manager -- Chandler, Arizona
John Stoner
Internet Consultant -- Fox River Grove, Illinois
Robert Stacy
Law Student -- Chapel Hill, North Carolina
Brian Harvey
Lecturer, Computer Science Division, Univ. of California,
Berkeley -- Berkeley, California
Rick Crawford
Computer Security Researcher, Univ. of California, Davis
-- Davis, California
Henry Schaffer
Professor -- Raleigh, North Carolina
Prasad Wagle
Software Engineer -- Mountain View, California
John Sadler
Software Engineer -- Palo Alto, California
Richard Swan, Ph.D.
Computer Scientist -- Portola Valley, California
David Hough
Software Engineer -- San Jose, California
Rob Snevely
Software Engineer -- Sunnyvale, California