From srctran@world.std.com Fri, 27 Aug 1999 11:07:42 -0400 (EDT)
Date: Fri, 27 Aug 1999 11:07:42 -0400 (EDT)
From: Gregory Aharonian srctran@world.std.com
Subject: [Patents] Internet patents - too many, too invalid, too lied about by
 PTO

!19990827  Internet patents - too many, too invalid, too lied about by PTO

(Rachel McCarthy please email me again - your address bounces).

(note: for those who keep track, I think the latest perpetual motion
machine to be patented by the PTO is 5,921,133 issued in July - using
gravity and unbalanced masses)

(pathetic note: yet one more reason politicians are pandering prostitutes:

    Vice President Al Gore shocked scientists Thursday with a statement
    from his office that local school boards had the right to teach
    creationism, although he personally favored the teaching of evolution.

    "The vice president favors the teaching of evolution in public schools.
    Obviously, that decision should and will be made at the local level
    and localities should be free to decide to teach creationism as well",
    said Alejandro Cabrera, a spokesman in the vice president's office.

    The statement, in response to an inquiry from Reuters, came a week
    after Republican presidential front-runner George W. Bush supported
    the teaching of creationism in public schools alongside the theory
    of evolution.
)

     To date, 4282 patents have issued that mention the Internet at least
once in the specification (though not all are Internet applications, some
for example, like patent 5,942,398, "Nucleic acid molecules encoding glutx
and uses thereof" only make mention of something on the Internet, in this
case GENBANK).  Additionally, there are many pre-Internet era patents that
don't mention the Internet but are relevant to/can be asserted against 
Internet activities (like videotex patents).  Whatever the actual number
that are Internet specific, easily over 2000, by looking at the issued
patents' claims, it is obvious that they are not being properly examined.
Commercially, the Internet is a big phenomenon, but intellectually it is
nothing more than a gradual evolution of the GUI, videotex, hypertext and
distributed multimedia activities of the 1980s and early 1990s - it wasn't
suddenly created to provide lots of possibilities to be novel and unobvious.
Maybe the Internet examiners at the PTO are from Kansas.

     Some history.  The first patent explicitly mentioning the Internet, in
its primitive form, is patent 4,438,511 titled "Packetized ensemble modem"
and issued 20 March 1984, and I quote the one mention of the Internet that 
appeared in the background summary:
 
     Several .  .  .  There is a major advantage in having a high speed
     modem that is capable of X.25 protocol compatibility for obvious
     INTERNET communication. There is a further significant advantage
     in having a modem that can communicate over INTERNETS utilizing
     protocols other than.  .  . prior art which are compatibile with
     packet switched networks (not to mention any that are compatible
     with multiple packet switching INTERNET protocols) and which operate
     in both synchronous and asynchonous modes concurrently through
     signal multiplexing.

The most recent Internet patent, issued this week, is patent 5,943,692,
titled "Mobile client computer system with flash memory management using
a virtual address map and variable length data".  Every conceivable feature
and application of the Internet is being patented.

    Below I list 48 titles and numbers of some Aug. 1999 Internet patents,
which not surprisingly cite no non-patent prior art.  Which brings us to
the issue of how credible are these Internet patents, and how credible are
PTO management's comments on the credibility of these patents.  What follows
are some statistics to show that both credibilities rank pretty low - i.e.
the PTO is making no progress in handling software patents, and has made
no progress in the last ten years, if for nothing more than the simple
reason that half of all issued software/Internet patents, as you read this,
are being issued citing no non-patent prior art.  Nothing has changed.

     For the most recent 100 patents mentioning the Internet, patents
ranging between 5,940,472 and 5,943,692 (i.e. all August 17/24 1999),
the average amount of non-patent prior cited (except for two obnoxious
outliers) is 2 non-patent prior art items, and 80% cite less than three
non-patent prior art items.  The 1984 4,438,511 patent shows that since
the Internet was significant enough to be patented and therefore published
circa 1980, there is twenty years of non-patent art to consider.  That is
a tremendous amount of material for which an average of 2 and median of 0
is woefully inadequate - these Internet patents are defacto nothing more
than being registered, since they are being examined in light of less than
a third of the relevant prior art.

     Since this average of 2 and median of 0 is the same (actually for some
reason it is slightlty less - I think the PTO is being overwhelmed) as the
average amount of non-patent prior art cited by a sample of 1000 software
patents I analyzed that were issued in 1994 (as well as a sample of 1000
patents from April 1999, and another sample of 500 early 1990s software
patents), whatever the absolute quality this reflects, it is quite clear
that in the last ten years, for the relative quality, the PTO has made ZERO
progress in handling non-patent prior art for computing.  Go read the
transcripts from the PTO's hearings on software patents four years ago -
few if any of the concerns raised in the testimony have been addressed,
which is why they had to be repeated at this year's hearings.  So when PTO
management touts:

               "We have more access to prior art now than
                      in the history of the PTO"

you have to wonder if they realize how meaningless that is, since unless
the examiners have the time, resources and experience to access this "art",
it doesn't matter how much you have (and having more access now than in
the past isn't a very impressive claim when you consider how little they
had in the past).

     Yet PTO management, in speeches and interviews, is saying that it is
making substantial progress, that the quality of the patents is high, and
that examiners are being supplied with sufficient resources to assess
novelty and obviousness.  But where are their statistics to back up such
bold assertions?  Nowhere.  The result is and will be a flood of Internet
and business method patents of a trivial nature that will be used to harass
legitimate businesses thanks to a PTO unsure of what it is doing (beyond
accusing critics of spreading myths).

     This actually isn't very surprising, given that few if any in PTO
management have advanced degrees or experience in computer science,
software engineering and especially library science, all of which you
need to oversee a large scale effort to insure that examiners get access
to much of the relevant prior art.  A couple PATNEWS readers suggested
that the next Commissioner should be selected using the same logic as was
used for the current head of the IRS, who was choosen mostly for his
experience with information processing systems as opposed to his experience
with tax policy and analysis.

     One reason the PTO keeps on doing so is the patent bar's public
silence, which the PTO can only interpret as tacit approval for their not
taking these issues seriously.  One recent bar failing happened this
spring, the Software Patents panel at the AIPLA conference in Chicago.
If you check the transcript at the AIPLA Web site, you will see no hard
questions on PTO performance asked of the PTO panelists, no panelists
challenging PTO statements, with little said to make PTO management debate
these issues in public.  I guess the committee was too busy schmoozing
(committee members for which I can easily recommend that you avoid using
if you need a software patent lawyer).

     What to do?  Start putting aside lot of money to pay for the many
invalidity opinions you are going to need as the letters of notice start
pouring in from this endless stream of crappy software/Internet/business
patents (I have lists of lawyers who prepare killer invalidity opinions
if you need such help).  And each time you get a letter, write one yourself
to the PTO Commissioner asking him with what data does the PTO justify
its rosy assessments of its handling of these patents.  And after you have
scared away whoever is asserting these patents, write another letter to
the Commissioner telling him that having just had to spend a lot of money
(invalidity opinions, even for crappy patents, cost tens of thousands of
dollars), you don't appreciate cavalier PTO comments such as:

        "[We've] seen a lot of criticism of the patent office out
        there, but I haven't seen anybody file [an Internet patent]
        re-examination request.  If they don't think we're doing a
        great job, then maybe they should step up to the place and
        file a reexamination"

with the implication that busy companies have nothing better to do than
spend even more money filing a reexamination request that they can't
even participate in.

Greg Aharonian
Internet Patent News Service


                              ====================

Patents including the word "Internet" at least once in the specification
issued August 17/24 1999 that cite no/zero/nasa/zilch non-patent prior
art.  Despite the many conferences, journals and books that have been
published in the last ten years explicitly dealing with the Internet,
and even more publications of applicable networking and multimedia
technology, the PTO is asking the public to believe that none of this
vast collection of prior art was worth citing for any of the following
patents.



5,943,692
Mobile client computer system with flash memory management utilizing
a virtual address map and variable length data
 
5,943,674
Data structure representing an interface definition language source file
 
5,943,665
Method and system for performing conceptual joins across fields of database
 
5,943,658
Method for automatic determination of origination ZIP code using caller ID
 
5,943,656
Methods and systems for computerized bill consolidating, billing and payment
authorization, computerized utility bill consolidating, utility billing
access and payment and utility provider consolidated billing systems
 
5,943,649
Configuring an audio interface for different microphone types
 
5,943,480
Method and system for detecting camouflaged congestion on network resource
 
5,943,477
Microlocal calibration of digital printers
 
5,943,423
Smart token system for secure electronic transactions and identification
 
5,943,421
Processor having compression and encryption circuitry
 
5,943,416
Automated survey control routine in a call center environment
 
5,943,406
Telephone call tracking and billing system and method
 
5,943,399
Methods for providing communications to telecommunications terminals
 
5,943,395
Telephone system and processes to enhance access for TDD and/or TTY devices
 
5,943,391
Method and device for a debugger and test data collector
 
5,943,321
Circuit set-up and caching for multimedia multipoint servers
 
5,943,317
Sub-network route optimization over a shared access transport facility
 
5,943,140
Method and apparatus for sending and receiving facsimile transmissions
over a non-telephonic transmission system
 
5,943,093
Software driver digital camera system with image storage tags
 
5,943,048
Method and apparatus for testing a graphic control area
 
5,942,711
Roll-sound performance device and method
 
5,942,710
Automatic accompaniment apparatus and method with chord variety
progression patterns, and machine readable medium containing program
 

5,941,996
Distributed network agents
 
5,941,957
Dependable web page synchronization mechanism
 
5,941,954
Network message redirection
 
5,941,951
Methods for real-time deterministic delivery of multimedia data in a
client/server system
 
5,941,945
Interest-based collaborative framework
 
5,940,871
Method for selectively decompressing operating system ROM image code using
a page fault
 
5,940,865
Apparatus and method for accessing plural storage devices in predetermined
order by slot allocation
 
5,940,855
Method for determining relative cache performance in a computer system
 
5,940,850
System and method for selectively enabling load-on-write of dynamic ROM
data to RAM
 
5,940,847
System and method for automatically correcting multi-word data entry errors
 
5,940,843
Information delivery system and method including restriction processing
 
5,940,823
System for the distribution and storage of electronic mail information
 
5,940,807
Automated and independently accessible inventory information exchange system
 
5,940,799
System and method for securing speech transactions
 
5,940,741
Method and apparatus arranged to reduce traffic on a network channel
 
5,940,739
Multiple access satellite communications network
 
5,940,738
Video pedestal network
 
5,940,611
Method and system for front-end gathering of store instructions within
a data-processing system
 
5,940,593
Simulating a multi-tiered computer environment on a single development
system for debugging
 
5,940,516
Data security method and system
 
5,940,497
Statistically-predictive and agent-predictive call routing
 
5,940,496
Apparatus and methods enhancing call routing within and between call-centers
 
5,940,489
Method and apparatus for detecting and recovering from call waiting
interruptions to modem communications
 
5,940,484
Self-provisioning names database for caller identification services
 
5,940,481
Parking management system
 
5,940,479
System and method for transmitting aural information between a computer
and telephone equipment




