PATNEWS: Sightsound waving wet spaghetti; nice review of State Street
Gregory Aharonian
Interesting week for software patenting. Monday's New York Times,
Business section, has two articles on the ongoing messes caused by a
software patenting system seemingly spiraling out of control (I am trying
to be profoundly poetic).
First, Teresa Riordan's regular column on patents (page C2) reports
on the controversy surrounding Cybergold's patent on paying consumers to
view Internet advertisements, 5,794,210. In the article, Nathaniel
Goldhaber, president of Cybergold (Berkeley), and James Love, director of
the Consumer Project on Technology (Washington, DC), trade barbs over the
validity of Cybergold's patents, with Mr. Love planning to press Congress
to conduct hearings on Cybergold's patents as well as other method of
doing business patents. Mr. Love is quoted as saying "This is typical of
the poor quality of review they give at the PTO.", which I repeat because
he is out-outraged in the second Times article.
At one point, Mr. Goldhaber is quoted as saying "People who globally
criticize patents on the Internet are just wrong", obviously ignoring the
fact that what people are complaining about are not Internet patents, but
the PTO's inability to properly examine Internet patents. Which Nathaniel
will probably learn to his regret, if he evers tries to assert his '210
patent, which was not properly prior-art searched by either Cybergold or
the PTO, and is a Gail Hayes patent, two strikes looking for a third strike
to declare it out of the ballpark of validity (yes, a really horrible
warped cliche) for at least 103 reasons I can think of.
====
The second article is by Denise Caruso (page C4), half of which is
profound observations by some patent news crackpot on the Internet, and
the other half about another controversial software patent, Sightsound's
patent for downloading, for a fee, music over networks. Ms. Caruso's
article can be seen at:
http://www.nytimes.com/library/tech/99/02/biztech/articles/01digi.html
She reports on the controversy involving Sightsound and MP3.com, with
Sightsound threatening action against MP3.com for infringing two patents
that Sightsound acquired that basically disclose methods of downloading
audio and video over networks for a fee transmitted over the networks
(the patents are 5,191,573 and 5,675,734). I list below two rather
humorous letters (except to MP3) that Sightsound sent to MP3.com. Mike
Robertson, president of MP3.com, expressed his outrage at the crap being
thrown against the electronic walls of his company (yea, yea, another
crappy cliche), "I just have to wonder if the Patent Office has completely
lost its mind".
His unhappiness, because the Sightsound patents have the same problems
that too many other of these electronic commerce and Internet patents have:
not only were they not well searched during the patent examination process,
they were not well searched before they were asserted (which I think is
just plain unethical). My advice - don't succumb to Sightsound's bluff.
The PTO should do everyone a favor and reexam this patent on their own -
an easy kill than the Compton's reexam.
====================
What follows are some pompous and blustery letters from Sightsound to
MP3. Had Sightsound spent as much time researching the validity of the
patents they bought as they did writing these letters, these letters would
be much less hilarious to those who know the history of the technology.
January 27th, 1999
VIA FACSIMILE AND U.S. CERTIFIED MAIL
Michael Robertson
President - MP3.com
3550 General Atomics Court Bldg 14
San Diego, CA 92121-1194
Dear Mr. Robertson,
This letter is to inform you that SIGHTSOUND.COM is the owner of
United States Patents 5,191,573 and 5,675,734 which control, among other
things, the sale of audio or video recordings in download fashion over the
Internet. I'd like to invite you to join our Limited Patent Licensing
Program.
I'm sure you are aware of the Secure Digital Music Initiative (SDMI)
which is intended to foster, among other things, the selection of an open
architecture, secure format for the download sale of music by the end of
this year. While SIGHTSOUND.COM is participating in SDMI and determining
whether the format will be appropriate for inclusion in a Long Term Patent
Licensing Program, SIGHTSOUND.COM is offering a One Year Limited Patent
License to companies like yours so that you can receive the protection of
Patents 5,191,573 and 5,675,734. The royalty rate for the License is one
percent (1%) of the total price charged to customers per transaction for
the download sale of music or other audio recordings.
Please contact me at your earliest convenience at the number listed
below. However, please understand that if you do not become an authorized
licensee, you MUST immediately cease and desist from selling music, or
other audio recordings, over the Internet in download fashion. I have
enclosed copies of both patents for your review. I look forward to hearing
from you. Sincerely,
Christopher J. Reese
Vice President and General Counsel
SIGHTSOUND.COM
733 Washington Road Suite 400 Mt. Lebanon, PA
(412) 341-1001 Voice (412) 341-2442 fax
====
February 1, 1999
VIA FACSIMILE
Michael Robertson
President - MP3.com
3550 General Atomics Court Building 14
San Diego, CA 92121-1194
Dear Mr. Robertson,
Although we haven't heard directly from you, we have read quotes
attributed to you, stating that you are not selling music in download
fashion. An article as recently as January 15, 1999 in Webnoize stated.
"[MP3.com] last year formed the Digital Automatic Music (DAM) program,
which enables participants to sell downloadable music and earn a 50%
royalty check." If the article is inaccurate and the download sale of
music is not in your plans, maybe you won't need a license.
In the meantime, I wanted to offer a thought for your readers. I
wonder how many of them are familiar with Elisha Gray. Most people are
not, yet he is the individual who filed for a patent on the telephone
two hours after Alexander Graham Bell. The point is that the granting of
a patent -- and with it, a licensed monopoly -- has never been without
controversy. However it is embraced by our Constitution (Art 1, Sec. 8)
and is the way our country has succeeded throughout generations in
encouraging innovation.
For those who think the patent system should be abolished, I probably
won't be able to persuade them otherwise. But for others, I would suggest
the way to judge a company that own patents, is to see how it uses its
licensed monopoly. Some patent-holders do not offer licenses at all and
seek to shut down competitors. Others charge upfront fees or high royalty
rates that price out smaller companies and inhibit new companies from
forming. Our current licensing structure includes a very reasonable
royalty raste and no-upfront payments at all. It is designed for all
companies, large and small, and we hope that, with our help, the SDMI
format will enable us to create a Long Term Licensing Program that will
be open to the smallest garage band as well as the largest record label.
I look forward to discussing these matters with you. Sincerely,
Christopher Reese
Vice President and General Counsel
SIGHTSOUND.COM
733 Washington Road, Suite 400
Mt. Lebanon, PA 15228
(412) 341-1101 Voice
(412) 341-2442 Fax
==============================================================================
A lot of people have asked me what I think of the recent State Street
decision. In short, the CAFC said "STOP USING 101 TO REJECT SOFTWARE -
enough decisions already". Given the current Mutually Assured Crap
strategies being pursued, the filings will grow rapidly to levels high
enough to ensure new levels of bad patents being issued, a retirement
plan for patent litigators and busters.
Now, for a more professional commentary on the State Street case, a recent
issue of Fenwich & West's IP Bulletin had a very nice article, which
appears below. And while they hedge their commentary, the authors are
some of the very few (but hopefully growing) number of patent attorneys
PUBLICLY expressing doubts about the ability of the PTO to handle software,
Interent and method of doing business patents. Until this problem is
openly discussed, it will never be solved.
==========
Software Patents and Internet Business Methods
in the Wake of State Street Bank
by Robert R. Sachs (r...@fwpa.com) and
Michael J. Sacksteder (michael_s...@fwpa.com)
Fenwick & West, Palo Alto - http://www.fwpa.com
Fall 1998
Earlier this year, the United States Court of Appeals for the Federal
Circuit took a step that some believed would have a major impact on the
patentability of computer software and the conduct of business on the
Internet. In State Street Bank & Trust Co. v. Signature Financial Group,
Inc., 149 F.3d 1368 (Fed. Cir. 1998), the Federal Circuit clarified the
tests for determining whether the subject matter of a computer-based
business application is subject to patent protection. In the wake of the
court's decision, one media headline screamed, "Floodgates Open for Patent
Cases." Another report predicted "Big Headaches for Internet Shopkeepers."
In reality, however, State Street Bank marks the culmination of an
incremental, sensible and somewhat predictable evolution in the law
concerning what "inventions" are "patentable" under Section 101 of the
Patent Act. Nevertheless, although the legal change wrought by the case is
relatively modest, the case and its aftermath raise important issues to be
considered by companies seeking to obtain patent protection for their
software, as well as by those who do business on the Internet.
State Street Bank v. Signature Financial Group
In State Street Bank, the Federal Circuit addressed two patentability
issues that relate to software and the conduct of business on the Internet.
Signature's patented invention was a data processing system that permitted
the administrator of a particular type of "hub and spoke" mutual fund
partnership to monitor and record the flow of financial information and to
make the calculations needed to maintain the arrangement. In particular,
the patent's claims were directed to a computer system that computed various
asset values and other financial data about the funds. The district court
held the patent invalid under Section 101 both as an unpatentable
mathematical algorithm and as an unpatentable business method.
The Federal Circuit reversed on both grounds. First, the court
effectively eliminated the "mathematical algorithm" exception to
patentability and replaced it with a test that considers whether any
invention - including one that incorporates mathematical algorithms - is
"reduced to some type of practical application, i.e., 'a useful, concrete
and tangible result'." A practical application is evidenced by the
transformation of data. According to the court, permissible transformations
are not limited to changes in physical matter - even a pure transformation
of numbers is acceptable as long as it produces a useful result:
Today, we hold that the transformation of data, representing
discrete dollar amounts, by a machine through a series of
mathematical calculations into a final share price, constitutes
a practical application of a mathematical algorithm, formula,
or calculation, because it produces a "useful, concrete and
tangible result" - a final share price momentarily fixed for
recording and reporting purposes . . .
The requirement that the useful result be "concrete and tangible" apparently
was satisfied by the fact that the share price was "momentarily fixed for
recording and reporting purposes," that is, stored in a computer memory.
After dispatching the mathematical algorithm exception, the court turned
its attention to the so-called "business method exception," which, according
to one school of thought, precluded patenting methods of doing business. The
court took the opportunity to "lay this ill-conceived exception to rest,"
correctly noting that neither the Federal Circuit nor its predecessor court
had ever invoked the exception to invalidate a patent. Instead, cases in
which the court had invalidated patents for business methods had really
turned on other issues, such as obviousness, the lack of novelty or the
attempt to patent an abstract idea. The court clarified that in the future,
the issue of whether a patent's claims are directed to patentable subject
matter under Section 101 "should not turn on whether the claimed subject
matter does 'business' instead of something else."
State Street Banks Impact
Around the time the State Street Bank opinion was published last summer,
a number of seemingly broad patents covering Internet business methods and
software applications issued, followed closely by press releases describing
the patents and media reports touting a major shift in the way such patents
are handled by the U.S. Patent and Trademark Office. One much-hyped patent,
U.S. Patent 5,794,207, is directed to the Internet company Priceline.com's
method for running a "reverse auction" over the Internet, in which a
potential buyer makes a conditional offer, secured by a credit card, to
purchase items such as plane tickets at a price specified by the buyer. The
offer is passed along to a number of potential sellers; if one accepts the
offer, payment is transferred to the seller via the buyer's credit card and
the sale is complete. Other patents were directed to other, similarly simple,
methods of operating on the Internet. In light of these patents, a brave new
world seemed to beckon.
But the novelty of this world is greatly exaggerated. As the State Street
Bank court pointed out, the business method exception was already largely
illusory. Moreover, the PTO has been granting patents for software for years,
notwithstanding any exception for mathematical algorithms. Under the PTO's
Examination Guidelines for Computer-Related Inventions, issued in 1996,
software was generally considered patentable subject matter if it fell into
one of the Guidelines' so-called "safe harbors." Under this approach, a
software invention was likely to be patentable if it a) received real-world
data and processed it, or b) produced a realworld output. Manipulation of
information inside a computer alone was more difficult, but still not
impossible, to patent.
State Street Bank opens the latter category to potential patentability
by invigorating a somewhat forgotten notion in patent law. An old series of
cases held that the essence of patentability was the transformation of matter
from one form to another. For example, a lathe is patentable because it
takes an item, such as a piece of wood, and makes something different from
it, such as a table leg.
State Street Bank brings the transformation test into the 1990s. Now the
transformation can be of "data" - i.e., information or just numbers. The
transformation of real-world data is no longer required, nor is a real-world
effect. All that is necessary is a practical application that produces a
useful result. While this new threshold is quite low, and certainly makes
it much easier to demonstrate patentable subject matter, there was no dearth
of software patents before State Street Bank. Nevertheless, the perception
that the "flood gates are open" exists, and the notion that Internet and
software patents are now easier to obtain may lead to a sharp increase in
applications for such patents.
State Street Bank and the PTO
Given the low threshold for patentable subject matter, the State Street
Bank decision makes it clear that the real issue of determining software
patentability is whether the invention meets the other standards for
patentability - specifically novelty and nonobviousness.
In this regard the PTO continues to improve in examining software patent
applications, though some observers believe the PTO has some difficulty
implementing the standards of novelty and nonobviousness to this field.
For example, patent examiners - typically young engineers - may lack the
expertise or broad familiarity with business software necessary to evaluate
patents that cover Internet business models. In addition, the resources
available in the PTO for locating nonpatent software prior art are not
totally comprehensive. Publications concerning software are not completely
organized into a comprehensive, easily accessible database, nor are the
myriad previously created software products and processes, which can be
very difficult to locate. Under these circumstances, the PTO may sometimes
issue software patents for ideas that are not really new and nonobvious -
such as retrofitting an existing business model for use on the Internet.
This does not mean that potential Internet patentees should avoid
seeking protection for their inventions. Indeed, because State Street Bank
has increased the perception that this type of patent is now easier to
obtain, it is more important than ever for an inventor to apply for
appropriate patent protection.
The important thing now is to make sure these patents have staying
power. Because a patent that can easily be invalidated by potential
infringers is obviously of limited value, inventors should work carefully
with their patent attorneys to obtain patents that can withstand a serious
validity challenge. The following precautions are especially important in
the software field:
- Critically evaluate the inventiveness of your idea. Consider what
portion of your software invention or Internet business method is really a
new, nonobvious development in the art. The patent with the broadest coverage
is not useful if it is not sustainable. Sometimes, the greatest competitive
value comes from a precisely crafted patent that provides reliable coverage
for the key solution provided by the invention. Do not assume that taking an
existing business method and putting it on the Internet is patentable. Often,
in spite of the hype or positive customer reaction, it is not.
- Provide the examiner with as much relevant prior art as is
reasonably available. Because patent examiners often lack access to and
knowledge of software-related publications, products and processes, it is
important to submit to the PTO as much relevant prior art as reasonably can
be located. This effort may include a search of nonpatent prior art, which
is far less likely to be available to the examiner than previously issued
patents. While the circumstances will not always justify an exhaustive
prior art search, the option should be considered.
- Craft the scope of claims and disclosure at multiple levels of
abstraction. The patent should define the essential functional aspect of
the invention independently of the details for its implementation. The goal
is to capture the essence of the invention in a way that does not limit it
to the description of the implementation detail. However, claims should be
written that cover the functional and structural design and architecture,
along with any implementation features that have strategic value.
Although these considerations are relevant to the prosecution of any
patent, they are especially important here. While State Street Bank may not
be the earthquake in the law of patentability that some believe it is, the
perception still exists. As the resulting stampede to the Patent Office
proceeds, it is important to obtain patents that will remain viable in the
face of any challenge.
========================================
Greg Aharonian
Internet Patent News Service