About having inhouse patent databases and possible liabilities, there
is no clear consensus, at least based on comments I collected from lawyers
who get my PATNEWS newsletter. Here is some comments I distributed last
October.
Greg Aharonian
DATE[19991021
TITLE[Inhouse patent database: triple damages or not????????????
TEXT[
Last spring I posted a question: if a company builds an inhouse
database of issued patents for their engineers to browse through for
educational and avoidance issues, are they liable for triple damages
if they are sued for infringing one of the patents that appear in the
database? I have been asked to build such databases a few times, but
each time non-IP corporate counsel found out and shut down the project.
The first time I asked the question, lawyers responded back with
answers along the lines of YES, NO, YES and NO, and MAYBE (the original
set of answers appears at the end). Those answers prompted some more
answers that appear below, along the lines of NO, YES, YES, MAYBE and
MAYBE.
Thus from nine lawyers the view is three NOs, four YESs, and three
MAYBEs. Look folks, 47 years after the Patent Reform Act of 1952,
there should be a definitive opinion about the conflict between two
fundamental aspects of the patent system: transferring knowledge to the
public for use, and triple damages for knowledgable infringement of such
patents.
====================
Greg-
Putting up this kind of Web page would just HAND anyone who OWNS one of
these 1000 patents a GREAT shot at treble damages, should there turn out
to be any infringement. Kind of makes their day. Definitely would be a
major blunder.
====================
Greg:
There is one point concerning the patent database that the Corporation is
considering putting on its intranet: in my experience as litigation
counsel, I have found that the marketing and technical people who try to
read a patent and then write memos (or even make notes) interpreting the
patent create a horrible written record. The interpretation is often
wrong -- not necessarily because they misunderstand the technology but
often because they do not understand how the technology is described in
legalese -- and usually contains some statement that can be explained but
that is bad as written.
For instance, I am currently trying to work my way back from a senior
engineer who wrote a memo that started with the phrase "Given that the
[X] patent is valid ...". When I asked him what he meant, without
prompting he said that it meant: "Given that [X] got a patent to issue ..."
Of course, that is not what he wrote and not what will the jury will be
shown (no doubt, hundreds of times).
I realize that I have not yet commented on the willfulness issue ... but
that is because I don't think I can add much to the debate you have already
republished. It's a close question, but there are other issues that must
be decided. Will the intranet track who accesses which patent when? (My
vote: no.) What limits will be put on who has access? (My vote: see next
question.) Will there be any patent-related training for the people who
have access before they can do so? (My vote: yes.) On balance, it probably
makes sense to provide the patents but it is not without risk, and I think
that the Company is wise to address the issue head on ... they are not being
too cautious.
====================
The general purpose of the patent database is to give the corporate
scientists (generally engineers) the benefit of reviewing the closest
patented art to their area of interest (generally a problem area in a new
product design). If the engineers successfully design around the patent
there is no infringement issue. If, with knowledge of the patent, they
copy the patented solution, we could, and should assume the copying
and, therefore, the infringement, is willful. These are the easy answer
extremes.
The next easiest issue/answer is the imputed knowledge of the
corporation of the patents through access by their engineers to the
patent database and the corporation's knowledge of its own products.
This seems no different than the constructive knowledge we're all charged
with based on the existence of the patent system, and the law is settled
that the patentee must give notice. This is true even if we assume a
customized database which is focused on the corporation's product classes.
We can't assume actual knowledge of infringement without proof.
Finally, if the corporation's actual knowledge of infringement does exist,
wilfulness is based on the corporation's conduct following actual
knowledge; maybe, depending on the circumstances, following actual
notice of the patentee's intent to enforce. At this point the issue is
intent; and if they intend to infringe following actual knowledge and
actual notice, they run the risk of treble damages.
One last comment. Its sometimes argued that allowing engineer access
to such a patent database will increase the likelihood of unconscious
copying. It may, but it seems that trade journals are the more likely
source of design copying, especially in the faster developing technologies
where designs are reported in trade journals years before they appear in
issued patents. In its best use, the patent database will provide the
corporation with the ability to both design its way around infringement
liability while at the same time possibly gaining a leading edge patent
with its design around.
====================
Another view. This is a confusing area because the case law seems to
support both points of view, and many lawyers, looking at only one case from
either "line of cases" has a potentially incorrect view. A good explanation
of all of this can be found in the article "The Patent Marking and Notice
Statute: A Question of "fact" or "act"? at 9 Harvard Journal of Law &
Technology at 429 1996. The article discusses what is still the seminal
1894 Supreme Court case of Dunlap v. Schofield, 152 U. S. 244 (1894)on the
meaning of the notice provision of the patent marking and notice statute.
The two lines of cases stem from the Dunlap Court's statement that
"notification under the notice provision of the statute involved an
AFFIRMATIVE FACT to be plead and proven by the patentee." One line of cases
(including Stryker which one of your respondents mentioned) says "Notice
requires an affirmative ACT by the patentee before damages begin to run."
The line of cases which latched on to the "affirmative ACT" requirement
stems from Muther v. United Shoe Machinery, 21 F. 2d 773 (D. Mass. 1927)
which has been followed at the Federal Circuit in Amisted Indus. Inc. v.
Buckeye Steel Castings Co. 24 F.3d 178 (FC 1994) and Devices for Medicine,
Incv. Boehl, 822 F.2d 1062 (FC 1987). The second line of cases (which says
that any source of notice will do) stems from the sixth circuit case of
Warner v. Tennessee Products Corp. 57 F.2d 642 (6th Cir, cert denied 287
U.S. 632 (1932), the holding being that actual notice of infringement
fulfills the notice requirement regardless of whether or not the patentee
performed an affirmative act.
This whole issue is further complicated by the fact the the patentee still
(regardless of how the defendant got the notice) must plead and prove the he
did have notice and that he infringed. This proof that he had notice is
often difficult and one gets into some of the questions raised by your
commenters, such as "is looking at the patent on a web site really proovable
notice?" etc. again the courts will likely use the "totality of
circumstances" test to make a determination. Additionally, note that a
patentee is not charged with marking an invention which is a process (not
"an article") so if one is only charged with infringing method claims, one
still must plead and prove that the defendant had notice, but this
"affirmative act" requirement is arguably irrelevant.
Finally because of all of this confusion, many prudent lawyers, out of an
abundence of caution, will say "don't set up this comprehensive map of
patents to your technology for your engineers to review" because it is
perhaps more likely than not that this set of circumstances would be the
wrong kind.
====================
The problem is that this isn't a legal question. It is a business risk
question, albeit partially premised on legal liability issues. So what
you have is:
Benefits: Increased knowledge of technical staff; increased ability
to focus on patentably inventions (by reason of knowing the prior art);
the synergy that comes from a thorough review of an art area; etc.
Risks: As stated, triple damages for willful infringement.
Of course, neither the benefit nor the risk is readily quantifiable.
Technical staff will clearly say that benefits outweight risks, because
they want the benefits. It sounds like corporate counsel is risk
adversive, so he or she is likely to be concerned no matter how small
the risk.
It might be that the best strategy is to focus on minimizing risk, as
opposed to trying to "weigh" the benefits and risks. Once the risks are
minimized, then the balance between risk and benefit may be clearer.
So, what can be done? In the first case, make the patent survey really,
really broad. Put 2,000 cases in it instead of 1,000. Put scientific
articles, etc. in it also. Make it huge. Why? At some point, a really
big internal database isn't any practically any different, in terms of
corporate knowledge, than a publicly accessible database (like PTO's or
commercial sites). It is a decent argument to make to a jury.
Also, document the selection criteria, and make it automatic (without
human discriminative intervention) -- so that the client doesn't have an
employee who selects (and thus in some sense "knows") cases, but rather
selection is done by word searches, art areas, etc., in a mechanical
fashion.
Think of some sort of corporate description/disclaimer (and put it on the
first page of the database) -- make sure that the selection criteria is
documented, etc., clearly document that nobody has reviewed the contents
to see if it applies to the company's business or products, etc., clearly
document that the info is derived from public sources and is simply made
available as a research aid, etc.
Past that, somebody's got to bite the nut and decide. If anyone can come
up with a calculus for actually determining the risk on such nebulous
factors, then I really want to see it (and patent it).
====================
THE ORIGINAL PATENT NEWS STORY
19990422 Question: corporate patent searching and triple damages?
I recently posed the following problem and question, asking for advice
from lawyers. Well the advice came in, but didn't really resolve the
issue. Here's the problem:
A company wants to do a comprehensive patent survey for their area
of technology, creating on an intranet a set of interconnected HTML
pages as a guide to about 1000 patents for their scientists and
engineers (who like the idea). On the other hand, the corporate
counsel is worried that doing so makes the liable for willful
infringement triple damages in any future infringement lawsuit over
any of the 1000 patents.
I have always thought this to be a strange contradiction of the patent
system. On one hand, the patent system is supposed to promote the
creation and distribution of new technology in return for IP rights.
But the triple damages clause seems to contradict that spirit by
discouraging corporations from seeking out these issue patents.
Question: are they worrying too much or not?
The answers I received to date are yes, no, yes and no, and maybe. Some
more answers would be nice to receive. I know a fair number of startups
involved with software that are facing this problem, and get confusing
advice.
==========
Greg, re the willful infringement question -- strange (contradiction) but
true, thanks to the Federal Circuit.
Making the database available to corporate patent counsel would clearly
give him/her actual notice of anything he/she looked at. However, the
mere availability of the database should not constitute actual notice any
more than access to Lexis/Patents or the IBM patent web site would. With
regard to the engineers and scientists, their access may or not constitute
actual notice of what they find for purposes of treble damage exposure.
The argument against notice is that they are not competent to interpret
and evaluate the scope of claims (although, in fact, many are). On the
other hand, once they communicate what they find to patent counsel, any
protection that might exist is lost.
There is a good article in the 1996 AIPLA Quartely Journal by Robert Bolan
and William Rooklidge on what it takes (or should take) to impute actual
knowledge of a patent to a company for purposes of wilfull infringement
(Vol. 24, No. 1pp. 157-190).
One of the more vexing, and as yet unanswered, questions in this area
relates to when a little bit of knowledge triggers a "duty of further
inquiry." Here's a hypo:
Your star sales rep emails corporate patent counsel that while making a
call on a customer recently, the customer remarked that the sales rep from
Company A (your main competitor) had told the customer that Company A had
just been granted a patent on their Product X which, by the way, is a
direct replacement for your Product Y. Any duty on the part of patent
counsel to punch Company A's name into the assignee field in the boolean
search screen of the IBM web site and look at what comes up?
==========
Yes, that company is worrying too much about triple damages and
liability for willful infringement. In the old days, when no technologies
existed for easily examining either the larger patent landscape or
individual patents, some courts tended to regard even the knowledge of a
competitor's patents as suggestive of willful intent to infringe.
The case of Stryker Corporation v. Intermedics in 1996 began to reverse
this thinking. Successful companies today need to have, if anything, a very
patent-savvy R&D program if they wish to avoid potential infringement
minefields in advance. New technologies and patent databases help make such
a patent-savvy R&D effort possible.
Of course, old thinking still survives. Witness the statement of Apple
Computer, in response to Imatech's $1.1 billion suit for willful
infringement, that it has "no policy or procedure for determining, prior
to the manufacture or sale of a product, if the product or its method of
manufacture infringes a patent owned by another."
In the old days, claiming ignorance made sense. Today, it's dumb dumb
dumb. If anything, it's an invitation to a shareholder lawsuit over
negligence in Apple's IP practices.
==========
Greg:
My two cents on the corporate patent web page. VERY BAD IDEA.
As you said, there is a contradiction in the law.
1) You cannot willfully infringe a patent you don't know exists.
2) On the other hand, once you know a patent exists, there is a
willfulness potential.
Suppose you find yourself in court as a defendant fighting the patent owner
of some patent displayed on your web page. Obviously, there is colorable
infringement (otherwise you probably wouldn't be in court). Now suppose the
allegedly infringing aspect of your product was designed after the company
learned about the patent. Finally, the company never obtained an expert
non-infringement or invalidity opinion.
How are you going to argue your way out of this one? Even if there is no
evidence that any engineer even looked at the patent, it still looks bad.
I wouldn't want to be in that position.
You want to hear something even worse?
Suppose I went out of my way to copy something in a patent. I make a carbon
copy of the patented apparatus, undersell the patent owner, and get rich.
The patent owner sues me, and I stop selling product on that day. Am I in
trouble, triple damages and the works? Not if the patent owner did not mark
his or her products with the patent number. The patent owner cannot recover
ANY damages until he or she personally makes me aware of the patent (even if
I went out of my way to find the patent and copy it).
Bottom line, don't do the web page.
==========
The answer is very easy, "It depends." If the intranet is put us as a
collection of good ideas to copy, and the engineers do copy things out of
the patents and put them into products, that sounds like the definition of
willful infringement.
On the other hand, if the things are put up as a guide to avoiding
patent infringement and the engineers are told to check with patent counsel
before using the teaching of the patents in products then it may be a good
defense that there is no willful infringement.
There is also the case of infringing a patent in the library even though
the particular engineer who infringed never used the library. Again, it
seems to me that if someone does a search using the library and concludes
there is no infringement, they are better off than not having the library.
The risk (to use your example of IBM and prior art) is having the
library which could have shown infringement but no one did a search.
==========
Yes, they are worrying too much. As usual, the lawyers are all wet.
Willful infringement (and many other legal transgressions) requires an
*intent* to commit a violation, and simple knowledge, or an information
system by itself, would not seem to be an equivalent.
As Judge Rafeedie said in the highly relevent and memorable case of tiny
CELERITAS v ROCKWELL INT'L, the corporate defendents at Rockwell had 'an
intent to misappropriate from the beginning' and were 'hedger and liars'
--overqualifying for willful infringement.
I also notice a parallel to Rule 56, or that rule which requires KNOWN
prior art to be submitted -- effectively discouraging searching. So
ignorance is viewed as a shield -- how pathetic. The author of that
ambiguous statute would be well advised to learn about the wide policy
implications, incentives and motivations caused by seemingly simple
legal code.
This archive was generated by hypermail 2b29 : Fri Aug 10 2001 - 15:58:03 EDT