Patently Strategic - Patent Strategy for Startups
Patently Strategic - Patent Strategy for Startups
Means-Plus-Function: The Risk of Losing Your Way
Word choice matters a great deal in the world of patenting. You’re using the English language to draw a picture around highly technical concepts. The precision with which this is done, down to the semantic level, can make all of the difference when it comes to your patent application being rejected or granted – and the future likelihood of your ability to assert your rights or defend against invalidation. Word choice too narrow or overly specific – and you can easily be designed around by competitors. Word choice too broad and only describing what something is vs. what it does and you risk rejection or invalidation for what will be ruled as linguistic tricks to get more coverage than what you actually invented. The tension is real and the case law interpretation is fluid, but it all still comes down to determining if the chosen words will enable a person of ordinary skill in the art to carry out an invention – in the interest of other inventors being able to build on the idea, while also avoiding trespassing with infringement.
One very particular place this tension between breadth of coverage and specificity in enablement arises is with the concept of means-plus-function claim language. In this month’s episode, Dr. Ashley Sloat, President and Director of Patent Strategy here at Aurora, leads a discussion, along with our all star patent panel, into the nuanced world of means-plus-function claiming. The group digs into the statute, explores relevant case law in an analysis of the kinds of word choices that have and haven’t caused problems for inventors, and also provides some great drafting tips for de-risking the use of means-plus-function claim language.
Ashley is joined today by our always exceptional group of IP experts including:
⦿ Kristen Hansen, Patent Strategist at Aurora
⦿ Dr. David Jackrel, President of Jackrel Consulting
⦿ David Cohen, Principal at Cohen Sciences
⦿ Shelley Couturier, Patent Strategist and Search Specialist
Before jumping into the deep with the panel, we also provide a quick primer on key concepts including specification vs claims, Section 112 enablement, functional claim language, and nonce words.
** Resources **
⦿ Show Notes
⦿ Slides
** Follow Aurora Consulting **
⦿ Home
⦿ Twitter
⦿ LinkedIn
⦿ Facebook
⦿ Instagram
And as always, thanks for listening!
---
Note: The contents of this podcast do not constitute legal advice.
WEBVTT
00:05.170 --> 00:08.394
G'day and welcome to the Patently Strategic Podcast, where we discuss all things
00:08.432 --> 00:11.458
at the intersection of business, technology, and patents.
00:11.614 --> 00:15.346
This podcast is a monthly discussion amongst experts in the field of patenting.
00:15.418 --> 00:19.302
It is for inventors, founders, and IP professionals alike, established or
00:19.316 --> 00:22.674
aspiring. And in today's episode, we pick up right where we left off
00:22.712 --> 00:25.794
last month with another but equally important look into the world
00:25.832 --> 00:29.238
of enablement. This topic hearkens back to the fundamental deal of the
00:29.264 --> 00:32.698
patent system. Long term progress of science and the useful
00:32.734 --> 00:36.514
arts is promoted by trading a government granted window of exclusivity
00:36.562 --> 00:40.126
to inventors in exchange for an enabling public disclosure
00:40.198 --> 00:44.170
that teaches the public how to make and use the invention. This predictably
00:44.230 --> 00:47.590
leads to some tension in the system. Inventors understandably
00:47.650 --> 00:51.418
want the broadest possible and most future proof coverage for their ideas.
00:51.514 --> 00:55.810
The health of the broader public innovation ecosystem, however, depends on not wholesale
00:55.870 --> 00:59.302
blocking of iteration and competition by granting exclusive rights
00:59.326 --> 01:02.854
that can go well beyond what was specifically invented in the scope
01:02.902 --> 01:06.414
of any given granted patent. One place in particular this
01:06.452 --> 01:10.146
tension arises is in an area of patent law referred to as means
01:10.208 --> 01:13.726
plus function claiming to someone newer to the world of IP,
01:13.798 --> 01:17.134
this is yet another terribly unintuitive turn of phrase.
01:17.302 --> 01:21.162
Fear not, we'll unpack that in just a bit. For now, suffice it
01:21.176 --> 01:24.918
to say that it is another semantic sharp corner alive and well in our
01:24.944 --> 01:28.374
court system, with two significant Federal Circuit Court rulings as
01:28.412 --> 01:31.966
recent as this past spring. There's a tremendous amount of nuance
01:32.038 --> 01:35.614
in these rulings, in this tension, and in the general understanding
01:35.662 --> 01:39.478
surrounding means plus function claiming to the point of seeming overly
01:39.514 --> 01:42.858
pedantic at first glance. But it's important to remember that
01:42.884 --> 01:46.714
word choice matters a great deal in the world of patenting. You're using the English
01:46.762 --> 01:50.818
language to draw a picture around highly technical concepts. The precision
01:50.854 --> 01:53.970
with which this is done, down to the semantic level, can make
01:54.020 --> 01:57.678
all the difference when it comes to your patent application being rejected or
01:57.704 --> 02:01.218
granted in the future likelihood of your ability to assert your rights or
02:01.244 --> 02:04.638
defend against invalidation word choice too narrow or
02:04.664 --> 02:08.350
overly specific you can easily be designed around by competitors.
02:08.470 --> 02:11.802
We're choice too broad and only describing what something
02:11.876 --> 02:15.562
is versus what it does, and you risk rejection or invalidation
02:15.646 --> 02:19.374
for what will be ruled as linguistic tricks to get more coverage than what you
02:19.412 --> 02:23.158
actually invented. The tension is real, and the case law interpretation
02:23.194 --> 02:27.246
is fluid. But it all still comes down to determining if the chosen words
02:27.368 --> 02:31.062
will enable a person of ordinary skill in the art to carry out an
02:31.076 --> 02:34.398
invention in the interest of other inventors being able to build on the
02:34.424 --> 02:37.878
idea while also avoiding trespassing with infringement. In this
02:37.904 --> 02:41.958
month's episode, Dr. Ashley Slot, president and Director of Patent Strategy here at
02:41.984 --> 02:45.558
Aurora, leads a discussion, along with our all star patent panel, into the
02:45.584 --> 02:48.834
nuanced world of means plus function claiming. The group
02:48.872 --> 02:52.342
digs into the statute, explores relevant case law and an analysis
02:52.366 --> 02:56.074
of the kinds of word choices that have and haven't caused problems for inventors,
02:56.182 --> 02:59.718
and also provides some great drafting tips for derisking the use of
02:59.744 --> 03:02.982
means plus function claim language. Ashley is joined today by
03:02.996 --> 03:06.766
our Always Exception, a group of IP experts including Kristen Hanson,
03:06.838 --> 03:10.918
patent strategist at Aurora dr. David Jackroll, president of Jacqueline
03:10.954 --> 03:14.170
Consulting david Cohen, principal at Cohen Sciences
03:14.290 --> 03:17.686
and Shelley Katarrier, patent strategist and search specialist.
03:17.758 --> 03:21.042
Now, before joining the group, as we often do, we'd like to provide a
03:21.056 --> 03:24.714
short primer on key concepts in this episode for those newer to the world
03:24.752 --> 03:28.330
of patenting. In breaking this down, there's a lot of talk in the episode
03:28.390 --> 03:31.738
about what's described in the spec or specification versus
03:31.774 --> 03:35.434
what's written in the claims. We review this distinction in the prior episode
03:35.482 --> 03:39.202
on American Axel and take an even deeper dive in our patent anatomy
03:39.226 --> 03:42.838
episode from season one. But in brief, the specification describes
03:42.874 --> 03:46.798
the invention and provides context for interpreting the claims.
03:46.954 --> 03:51.106
The claims are the heart of the application and point out the exact invention boundaries
03:51.178 --> 03:54.646
of what the applicant believes she's entitled to, specifically,
03:54.718 --> 03:58.258
what the patent covers. Now, moving on to statute scope
03:58.354 --> 04:01.662
when patents are examined by the patent office or later litigated in
04:01.676 --> 04:04.902
a courtroom, several sections of US statute come into play in
04:04.916 --> 04:08.578
determining if the claims in the patent are eligible, useful, novel,
04:08.674 --> 04:11.770
nonobvious, and enabled or properly described.
04:11.830 --> 04:15.318
Patents can be rejected or invalidated if one or more of the claims are
04:15.344 --> 04:18.858
determined to be otherwise. Problems fall under title 35 of
04:18.884 --> 04:22.678
US. Code in this episode centers around section 112, which is largely
04:22.714 --> 04:26.194
concerned with enablement. We're describing the invention in sufficient
04:26.242 --> 04:29.970
detail to allow it to be practiced by someone skilled in the art without
04:30.080 --> 04:33.858
undue experimentation. Subsection F of section 112
04:33.944 --> 04:37.710
addresses the use of functional language and patent claims with many
04:37.760 --> 04:41.898
conditions that our panel will cover. US. Patent law does provide provisions for
04:41.924 --> 04:45.658
using functional language, or what something functionally does versus
04:45.694 --> 04:49.134
the more traditional approach that defines the structure or the
04:49.172 --> 04:52.998
how it does it of the invention. Another way of saying this is the
04:53.024 --> 04:56.614
claim is functional when it describes a feature by what it does versus
04:56.662 --> 05:00.778
what it is. So why is this referred to as means plus function claiming
05:00.874 --> 05:04.114
despite it not being the only trigger? These claims originally
05:04.162 --> 05:07.734
appeared as phrases worded like a means to turn on a light
05:07.772 --> 05:11.350
bulb, so literally quote a means unquote
05:11.530 --> 05:15.166
followed by or plus some functional outcome.
05:15.298 --> 05:18.858
This generic language provides no specificity as to how the ball is
05:18.884 --> 05:22.134
turned on, potentially including structural elements like
05:22.172 --> 05:25.470
switches and copper wires, but instead encompasses all
05:25.520 --> 05:29.058
possible means by which a bulb can be turned on. While use of this
05:29.084 --> 05:32.598
type of language in its equivalence is legally in play, as you learn
05:32.624 --> 05:36.570
in the discussion, corresponding structure must still be found or
05:36.620 --> 05:40.390
commonly inferred if the claim is not at risk of being ruled indefinite
05:40.450 --> 05:44.298
by the patent office or courts. So the last bit of context we'd like
05:44.324 --> 05:48.130
to provide is around the word nons, which practitioners use with regularity,
05:48.190 --> 05:51.690
and this episode is no exception. To get around the literal means plus
05:51.740 --> 05:55.230
function analysis being triggered by the literal use of the word
05:55.340 --> 05:59.050
means for in a claim, people started using slightly
05:59.110 --> 06:03.226
more specific language and applications. And I do emphasize
06:03.298 --> 06:06.930
slightly these generic words and verbal constructs were still
06:06.980 --> 06:10.198
often void of structure and lacked definite meaning.
06:10.294 --> 06:12.838
Mechanism, module, device,
06:12.994 --> 06:15.910
unit, component, element, member,
06:15.970 --> 06:19.462
apparatus, machine, system. These are referred to as nonce
06:19.486 --> 06:23.242
words. And just like they're a means for cousin, can all kick
06:23.266 --> 06:27.330
off a world of hurt if not properly supported with enabling structure.
06:27.650 --> 06:31.290
All right, that's hopefully enough to send you on your way into the nuanced world
06:31.340 --> 06:34.722
of functional claiming. As with most things, this is a concept that
06:34.736 --> 06:38.106
is neither inherently good nor bad. But knowing what to look for,
06:38.228 --> 06:41.362
how to safely leverage it, and when to avoid it altogether
06:41.506 --> 06:44.694
are key ingredients to staying calm and patenting on.
06:44.792 --> 06:48.630
Now, without further ado take it away, Ashley. All right,
06:48.680 --> 06:52.762
so getting to the real issue at hand, which is meso function claiming
06:52.906 --> 06:55.782
and kind of the state of that and where we're at, and I try to
06:55.796 --> 06:59.374
kind of break today's discussion into device
06:59.482 --> 07:03.238
MPF versus software NPF because there are some nuances
07:03.274 --> 07:06.522
to each, I think. But in general, kind of
07:06.536 --> 07:10.098
the same overarching rules apply. You can
07:10.124 --> 07:13.722
see that in the last 70
07:13.796 --> 07:17.806
years now, that the number of patents that include
07:17.878 --> 07:21.754
means for claim limitations have dropped
07:21.802 --> 07:24.846
precipitously over time. And I
07:24.848 --> 07:27.500
would bet that trend continues into this day.
07:28.550 --> 07:32.360
Or maybe it kind of maybe kind of levels off at about ten, 5% or
07:33.050 --> 07:36.738
so. It's definitely not as common as it used to be, but you
07:36.764 --> 07:40.218
still see it. I know that I still occasionally throw in
07:40.364 --> 07:44.094
some means language into a set
07:44.132 --> 07:47.454
of claims. Not to say it's the only set of claims, but I still
07:47.492 --> 07:50.782
kind of do it just because I feel like it provides
07:50.806 --> 07:54.810
maybe some breath for something from the spec that you didn't actually
07:54.980 --> 07:58.470
claim, but maybe you can still capture what that means for language.
07:59.270 --> 08:02.958
But like I said, there's definitely some short corners to this as well. So in
08:02.984 --> 08:06.966
general, section 112 F, or paragraph six
08:07.148 --> 08:10.760
of the MPEP describes what main plus function is.
08:13.710 --> 08:17.966
Section 112 F is applied if you recite a function without reciting
08:18.038 --> 08:21.802
structure for performing the function and then it limits the claim to the
08:21.816 --> 08:25.430
structure, materials or acts disclosed in the specification or equivalents thereof,
08:25.550 --> 08:29.302
and it's not applicable. Section 112 F is not applicable if you
08:29.316 --> 08:32.520
recite both a function and the structure for performing that function.
08:33.150 --> 08:36.190
And so a lot of times what a lot of the case law that I
08:36.240 --> 08:39.120
read when the courts in particular,
08:39.750 --> 08:42.922
and I guess I haven't read enough about what examiners are
08:42.936 --> 08:47.134
doing. But in courts in particular, when they're trying to identify whether means
08:47.172 --> 08:50.734
plus function applies and whether you have sufficient disclosure to
08:50.772 --> 08:54.818
support that means plus function language,
08:54.974 --> 08:58.198
they basically determine the claimed function. So they
08:58.224 --> 09:01.274
basically say, okay, what's the means term?
09:01.442 --> 09:04.798
Right? Is it some kind of means for doing something? So what
09:04.824 --> 09:08.230
are all the functions that means it's performing? So maybe it's one function,
09:08.280 --> 09:12.098
two functions, four functions. And then they try to identify the corresponding
09:12.134 --> 09:15.454
structures in the written description that perform that function.
09:15.612 --> 09:19.438
And if they can only find one
09:19.584 --> 09:22.714
function, if they can only find support for
09:22.752 --> 09:26.438
one of the multiple functions, it's found indefinite.
09:26.534 --> 09:29.710
So I think it's really interesting in thinking about if you are to use
09:29.760 --> 09:33.566
means plus function language, making sure that you're
09:33.758 --> 09:38.230
clear on all the different functions that you're giving
09:38.280 --> 09:42.900
that means and then making sure each of those functions has
09:43.230 --> 09:45.550
structure in the specification.
09:46.050 --> 09:50.280
So how do you get around MPF or how do you navigate it?
09:50.670 --> 09:54.338
So if means is not used, it must be shown that persons of ordinary
09:54.374 --> 09:58.418
skill in the art would not have understood those term limitations to connote
09:58.454 --> 10:01.678
structure. And so that's what they kind of have to
10:01.704 --> 10:05.014
prove is that, okay, so you didn't use the word means, but you use some
10:05.052 --> 10:08.158
other potential means like word.
10:08.304 --> 10:11.498
So now there is the burden probably of the examiner of the court
10:11.534 --> 10:15.022
to say, well, somebody of ordinary school wouldn't understand that
10:15.036 --> 10:19.610
that had any structure to it. If an MPs determination
10:19.670 --> 10:23.366
is not this design, then the applicant needs to show that a particular claim element
10:23.438 --> 10:27.120
would be understood as connoting structure. And so
10:27.690 --> 10:31.250
this kind of balance of, again, trying to make sure that there's sufficient
10:31.310 --> 10:34.654
structure in the description to support that means,
10:34.692 --> 10:37.922
the function language and this kind of use that means the function dates
10:37.946 --> 10:41.518
all the way back. I mean, the Morse code is,
10:41.604 --> 10:44.470
I think, one of the earliest ones where they kind of talked about a means
10:44.520 --> 10:47.782
of transmitting was
10:47.796 --> 10:52.066
it like electromagnetism or whatever it was, to kind of send
10:52.128 --> 10:55.810
signals, send words. I think that was ultimately
10:56.490 --> 10:59.626
I can't remember if it was actually back at that time or just later and
10:59.688 --> 11:03.898
people kind of retroactively looking at things, but that there
11:03.924 --> 11:07.680
wouldn't have been sufficient structure even in that application to support
11:08.190 --> 11:12.050
the breadth of the claims that were in the Morse code patent.
11:12.230 --> 11:15.362
But one of the more somewhat recent in the last deck,
11:15.446 --> 11:19.058
last decade, last half century or so, is this Halliburton
11:19.094 --> 11:23.062
Oil versus Walker. And so in this case it was
11:23.076 --> 11:26.330
the late 1920s and the oil industry was experimenting
11:26.450 --> 11:30.358
with some different sound echo time methods for measuring the
11:30.384 --> 11:33.470
distance to the fluid surface and deep oil wells.
11:33.590 --> 11:37.020
And so the claims in this application
11:37.590 --> 11:41.074
were the product of experimentation upon this
11:41.112 --> 11:42.670
Lean Wyatt patent.
11:44.590 --> 11:48.558
The investment basically added a mechanical acoustical resonator
11:48.714 --> 11:52.002
to the prior art device for measuring that distance to the fluid surface
11:52.026 --> 11:55.374
and deep oil wells. You can see here the claims recited
11:55.422 --> 11:58.814
a means associated with said pressureresponsive device for tuning said
11:58.852 --> 12:02.214
receiving means to the frequency of echoes from the tubing collars
12:02.262 --> 12:05.958
of said tubing sections to clearly distinguish the echoes from said couplings
12:05.994 --> 12:09.990
from each other. And ultimately the Supreme Court invalidated
12:10.050 --> 12:14.022
Walker's mean plus functional claims because they said it was improper
12:14.166 --> 12:17.522
to claim the most crucial element in terms of what
12:17.536 --> 12:21.482
it will do, rather than in the terms of its own physical characteristics or
12:21.496 --> 12:25.362
its arrangement in the new combination of apparatus. So none
12:25.386 --> 12:28.898
of the claims describe the physical relation of the Walker addition to
12:28.924 --> 12:32.078
that of the old Lair and Wyatt machine. None of the
12:32.104 --> 12:35.394
claims describe the manner in which the Walker edition would operate
12:35.442 --> 12:40.518
together with the prior art machine to make a new apparatus.
12:40.674 --> 12:44.740
And so ultimately the claims failed. And so again,
12:45.490 --> 12:48.494
I used this more than historical one. I didn't deep dive into this case,
12:48.532 --> 12:52.094
but kind of historical perspective. This is where the MPs field
12:52.132 --> 12:55.466
has kind of been going. The most notable case
12:55.528 --> 12:58.250
of recent is this Williams versus Citrix,
12:58.990 --> 13:02.222
and this was in 2015. And the reason I think this one
13:02.236 --> 13:05.694
has gotten a lot more fanfare
13:05.802 --> 13:09.422
or at least play in internet and
13:09.436 --> 13:13.002
stuff like that, is that they started. So historically,
13:13.086 --> 13:16.480
if you used the term means,
13:17.290 --> 13:20.894
it was presumed that it was a section 112 F means
13:20.932 --> 13:24.566
plus function claim. If you didn't use the word means,
13:24.628 --> 13:28.480
it was a very strong presumption that 112s didn't apply.
13:28.990 --> 13:33.278
This Williams versus Citrix online case shifted away from that
13:33.424 --> 13:37.470
and basically saying that there is now no longer a strong presumption
13:37.530 --> 13:40.778
against a means plus function interpretation in the
13:40.804 --> 13:44.594
absence of the means language. And it
13:44.632 --> 13:47.982
also established the nonseward doctrine. And I know David Cohen,
13:48.006 --> 13:51.162
I talked about this the other day and I have a whole slide about nonce
13:51.186 --> 13:53.500
words and nonnounce words.
13:54.490 --> 13:58.578
But what this case ultimately?
13:58.674 --> 14:02.234
Let me take a step back here. So Williamson had invented software for
14:02.272 --> 14:05.910
connecting one of our presenters with geographically remote audience
14:05.970 --> 14:10.022
members. And so in their claims they had a distributed learning
14:10.096 --> 14:15.054
control module. So obviously nothing with means, their meantion
14:15.162 --> 14:18.470
word was a module, distributed learning control module.
14:19.930 --> 14:23.790
And so the district court had evaluated the specification
14:23.970 --> 14:27.966
and concluded that there were no necessary algorithms
14:28.158 --> 14:32.102
in the spec for performing those claimed functions. And ultimately the
14:32.116 --> 14:35.862
Federal Circuit upheld and found that the term distributed
14:35.886 --> 14:39.146
learning control module was indefinite because they basically
14:39.208 --> 14:42.458
said that one of ordinary skill in the art
14:42.544 --> 14:46.418
won't know what kind of meaning or what kind of structure that
14:46.444 --> 14:48.950
this module would have connoted.
14:52.850 --> 14:56.598
This basically led to that heightened burden to make sure that you
14:56.624 --> 14:59.902
have the structure. And in particular, and I'll get into this later, but particularly
14:59.926 --> 15:03.582
for software, you can see in a lot of the case law that they really
15:03.656 --> 15:06.810
want for something that is interpreted as
15:06.860 --> 15:10.206
means plus function in the software world,
15:10.388 --> 15:14.146
you need to have structure and or algorithms
15:14.338 --> 15:18.534
that define what that means for function word or
15:18.572 --> 15:21.810
term is. And so that's kind of what they failed here
15:21.860 --> 15:25.710
is there just was nothing there's, no algorithms or anything like that, that fully
15:27.770 --> 15:31.290
gave the meats and balance of that distributed learning control module.
15:31.610 --> 15:35.118
Right, Ashley, let me add in here, we're very familiar with
15:35.144 --> 15:38.530
this and it is painful.
15:38.650 --> 15:42.630
Williamson visitrix, remember, was the first time that software was called
15:42.680 --> 15:46.162
into play to say these modules
15:46.186 --> 15:49.854
and these means for language pieces are
15:49.892 --> 15:53.082
not properly supported in specs as we see them.
15:53.216 --> 15:56.362
So if you do a means for in kind of a mechanical
15:56.446 --> 15:59.886
case, one skilled in the art might
15:59.948 --> 16:02.934
understand that a fastener could be 40 different things,
16:03.032 --> 16:06.174
even if your spec did not list all of those
16:06.212 --> 16:09.706
things. You may be able to stretch something because one skilled
16:09.718 --> 16:12.570
in the art would stretch something. But with software,
16:12.890 --> 16:15.882
you have no idea what that algorithm might have been. You have no idea what
16:15.896 --> 16:19.940
it means when somebody says decision module or
16:20.570 --> 16:24.826
decision engine or decider
16:24.898 --> 16:28.122
thing or anything that you say module engine or whatnot if
16:28.136 --> 16:32.070
you don't fully explain that in your specification,
16:32.450 --> 16:35.240
there is no support for it.
16:35.570 --> 16:39.282
So it isn't just messy claim language, it's just poor spec
16:39.356 --> 16:43.038
writing. And this case brought that to the forefront because
16:43.184 --> 16:46.798
as a software prosecutor, you can quickly write up flow
16:46.834 --> 16:50.026
charts, but if those flowcharts aren't supported
16:50.158 --> 16:54.202
by what the modules are that are performing those pieces,
16:54.346 --> 16:58.102
and or if you don't take the time to kind of get those examples
16:58.126 --> 17:02.242
in there, it's just too late. It's not supported,
17:02.326 --> 17:05.886
and you're going to get these rejections all over the place. So a
17:05.888 --> 17:09.598
lot of people who are a little bit newer to software
17:09.634 --> 17:13.422
drafting. I would say probably
17:13.496 --> 17:16.662
after 2012. Maybe even later than that.
17:16.856 --> 17:20.106
They will shy away from using means for
17:20.168 --> 17:23.698
and using module even or anything like that in their block
17:23.734 --> 17:28.674
diagrams and their flow charts because they do not want
17:28.712 --> 17:32.718
to take the time to make sure that's properly supported for the
17:32.744 --> 17:36.658
means for language and they don't want to have it on their file
17:36.694 --> 17:40.590
histories. So sometimes you will get these rejections, but it's fine because
17:40.640 --> 17:44.134
you've done your job. Your spec is good, everything's appropriate,
17:44.182 --> 17:47.562
and it's supported. But sometimes you get these rejections and you have to
17:47.576 --> 17:51.178
check that and you may have to remove it from your claim
17:51.214 --> 17:54.654
language because you don't have the proper support. So some
17:54.692 --> 17:58.290
people will shy away from that. But old school software drafting,
17:58.730 --> 18:02.300
you will see module all over the place. You will see means for,
18:02.990 --> 18:06.298
and there are a lot of practitioners who still do this and there's
18:06.334 --> 18:09.762
nothing wrong with it. It's just a little dicey and it's a little more difficult
18:09.836 --> 18:13.050
to make sure your specification meets the standards.
18:14.090 --> 18:17.874
Yes, that's very helpful, Kristin. And the thing that came through to me most was
18:17.912 --> 18:21.534
that, again, just finding all those functions and
18:21.572 --> 18:25.078
making sure there's a full flow chart algorithm
18:25.114 --> 18:29.118
or whatever that clearly supports that function. So just
18:29.144 --> 18:32.682
making sure you got all those functions vetted out. Yes. Even if
18:32.696 --> 18:36.246
it's the dumbest flow chart you've ever seen three boxes of do this, do X
18:36.308 --> 18:40.400
and do Y. I mean, it just has to be in there. And that's support.
18:40.910 --> 18:44.742
Sometimes you need it deeper than that. But yeah,
18:44.816 --> 18:48.022
absolutely. Thank you for that. That kind of brings up the noncers
18:48.046 --> 18:52.182
versus non nonswords. And of course, this is not an exhaustive list,
18:52.256 --> 18:55.650
right? So obviously words
18:55.700 --> 18:58.606
like module, mechanism, unit, component, element,
18:58.678 --> 19:02.818
member, apparatus, machine, system are kind of void
19:02.854 --> 19:06.378
of structure, right? Because there's so many things that
19:06.404 --> 19:10.458
could be considered. Any one of those things from a non nonse word,
19:10.484 --> 19:13.640
the most sounds like nonsense, nonnonse words,
19:14.630 --> 19:18.094
words like circuit, detent mechanism, digital detector,
19:18.142 --> 19:20.670
reciprocating member, connecting, assembly,
19:22.130 --> 19:25.458
seemingly connected joints. Obviously, the thing that
19:25.484 --> 19:30.210
I kind of found in some of these was that the non words had
19:30.320 --> 19:34.410
non structural modifiers. That's like a non structural adjective
19:34.730 --> 19:37.640
and a non structural term. So if you kind of said,
19:39.650 --> 19:43.134
I don't know, like a reciprocating system, maybe that would work,
19:43.172 --> 19:45.978
maybe it's more structural. I'm trying to think of even an example if somebody has
19:46.004 --> 19:49.930
one shout out. But on the other side of the non
19:49.990 --> 19:53.278
words, there's clearly a structural modifier and a non structural term.
19:53.314 --> 19:57.010
So a digital detector, obviously detector
19:57.070 --> 20:00.654
has some level of structure in it. A detent has
20:00.692 --> 20:04.220
some level of structure in it. A circuit has some level of structure in it.
20:04.670 --> 20:08.358
And so the more that you have some sort of structural component to the
20:08.384 --> 20:11.974
term, the better. But what you don't want is stacking
20:12.022 --> 20:15.522
up all these adjectives that are directly tied to the means of
20:15.536 --> 20:19.362
these non structural modifiers that are attached to the means or
20:19.376 --> 20:23.038
some other nonsword. And that seems a lot of people kind of get in trouble
20:23.134 --> 20:26.910
or end up kind of in this mean plus function interpretation.
20:28.850 --> 20:32.790
Do you think that configure two would also be an oddsword?
20:33.290 --> 20:37.362
I think it can be, but again, it depends on
20:37.496 --> 20:40.842
what that function if
20:40.856 --> 20:44.586
you're saying like an element configured to and
20:44.708 --> 20:48.090
maybe that configured two language gives it some structure.
20:48.590 --> 20:51.810
If it's an element configured to attach
20:52.310 --> 20:55.820
element A to element B or
20:56.150 --> 20:59.550
to attach a shaft to a base,
20:59.720 --> 21:04.158
then maybe the element could be considered less of a
21:04.304 --> 21:07.662
nonce word because the structural feature, there could be some kind
21:07.676 --> 21:11.190
of maybe fastening means or something like that because it's connecting,
21:11.810 --> 21:14.754
fixing this shaft to this base.
21:14.912 --> 21:18.246
But from a software perspective, for example,
21:18.308 --> 21:22.366
if you're like an element configured to identify
21:22.438 --> 21:26.278
a user, what is that? Is that a device?
21:26.374 --> 21:29.370
Is that some kind of algorithm? Is it a sensor?
21:29.690 --> 21:33.510
What is that? And so I think it kind of depends what comes
21:33.560 --> 21:37.062
after configured. Yeah, coming from the
21:37.076 --> 21:40.678
other side of the software side, I rarely use configured to unless
21:40.714 --> 21:44.298
I'm tying it to the processor doing it. So processor configured two or
21:44.324 --> 21:47.758
processor and memory configured to carry out these instructions.
21:47.914 --> 21:51.426
I'll use it there, but I'll rarely use it within
21:51.488 --> 21:53.480
my claim body.
21:55.010 --> 21:58.410
I'm sorry, say that again. You would use it where? I would
21:58.460 --> 22:01.542
use it in like a preamble where I'm saying I have
22:01.556 --> 22:04.590
a processor configured to carry,
22:04.910 --> 22:08.034
right, but I wouldn't use it in the body very often for
22:08.072 --> 22:11.238
software, if at all. It's the same issue.
22:11.324 --> 22:15.046
It has to have a little bit more support, a little bit stronger
22:15.118 --> 22:18.954
tie to exactly which algorithm is
22:18.992 --> 22:22.940
configured to, why it's configured to and what the actual steps are.
22:23.570 --> 22:26.922
Rather than if I say an algorithm that includes A,
22:26.936 --> 22:29.540
B and C and I just call out A, B and C,
22:30.050 --> 22:32.838
I have a processor that can do it and that's fine. But if I say
22:32.864 --> 22:36.800
configured to, I have to describe what I mean by configured to.
22:37.790 --> 22:41.490
All right? Like a special processor potentially,
22:41.930 --> 22:44.986
or is it pre programmed, is it an FPGA?
22:45.058 --> 22:48.874
Right? Is it changeable? There's all kinds of just weird odd
22:48.922 --> 22:52.280
things that pop up in the software world for using configured too.
22:53.810 --> 22:56.706
So I'm going to walk through a little bit of case law for some kind
22:56.708 --> 23:00.798
of the device world and a little bit for the software world. So the
23:00.824 --> 23:04.762
Gregory Baron versus Medical Device Technologies case. Federal Circuit
23:04.786 --> 23:08.734
210. Basically, Baron invented an automated biopsy
23:08.782 --> 23:12.486
instrument. Basically the device that he was trying to protect was
23:12.548 --> 23:16.366
charged by pulling back an external guide which was attached
23:16.378 --> 23:20.254
to the cannula along the shaft until the guide locks in place. The locking
23:20.302 --> 23:23.422
function was eleven, that would slide
23:23.446 --> 23:27.190
into a slot. Pressing the other end of the lever would release
23:27.250 --> 23:30.438
the lock and allow the spring to send the cannula forward over
23:30.464 --> 23:33.982
the stylist. So basically like for taking a biopsy
23:34.066 --> 23:38.470
sample and so in Baron
23:38.530 --> 23:42.198
spec or in their claims, they had a release means for
23:42.224 --> 23:45.858
retaining the guide in the charge position. And so
23:45.884 --> 23:49.182
when they did the claim construction, they basically said again, one of the
23:49.196 --> 23:52.858
functions has a release function and a retaining
23:52.954 --> 23:56.914
function, right, retaining the guide in the charge position and releasing the guide
23:57.082 --> 24:00.426
from the charge position. Now Baron tried to say no, it's only one
24:00.488 --> 24:04.206
function, that the other one. I think it was that the releasing part was just
24:04.268 --> 24:07.734
a modifier, but they said no
24:07.892 --> 24:11.274
had bothered function. And unfortunately what they
24:11.312 --> 24:14.974
had in the spec did not support, they didn't have sufficient
24:15.022 --> 24:17.180
structure to support both of those.
24:18.290 --> 24:21.814
And so from a literal infringement in this case perspective,
24:21.982 --> 24:25.782
they needed to find a device that performed both of those structures or
24:25.856 --> 24:29.022
some kind of equivalent function and they couldn't. So not
24:29.036 --> 24:32.574
only were the claims indefinite, but there was also no infringement because
24:32.732 --> 24:36.558
again, that means Term was determined to have two
24:36.644 --> 24:41.538
functions and those functions were not fully supported by the spec and
24:41.564 --> 24:44.214
then for tech licensing core versus video tech.
24:44.372 --> 24:48.058
So this one was interesting because they described a monitoring
24:48.154 --> 24:51.742
means in the spec which described a node to node communication
24:51.826 --> 24:56.646
system and so they did have expert testimony can also obviously
24:56.828 --> 24:59.754
support what one of skill in the art would have thought,
24:59.912 --> 25:04.014
how much structure, whatever term connoted. And so they did have an expert that
25:04.052 --> 25:07.914
testified that the controllers in the circuitry that were
25:07.952 --> 25:11.206
in the spec were very much the corresponding structure.
25:11.338 --> 25:13.450
There was no circuit diagram,
25:13.630 --> 25:17.398
but the necessity of having a circuit
25:17.434 --> 25:20.538
diagram is dependent on the level of one of skill in the art.
25:20.684 --> 25:24.502
So this one was found to be definite because there was sufficient structure.
25:24.646 --> 25:29.158
And so I think one of the more interesting quotes
25:29.194 --> 25:32.382
from this case was that the question is not whether one of skill in the
25:32.396 --> 25:36.498
art would be capable of implementing a structure to perform the
25:36.524 --> 25:39.898
function, but whether that person would understand the written description
25:39.934 --> 25:43.750
itself to disclose such a structure.
25:43.870 --> 25:47.166
So again, the whole crux of the patent system is to
25:47.228 --> 25:50.866
enable one of skill in the art to make and use the invention.
25:50.938 --> 25:54.582
And if you don't have enough structure in there for them
25:54.596 --> 25:57.994
to make and use the invention, then not to say that they couldn't
25:58.042 --> 26:01.978
figure it out and couldn't with some experimentation figure it out, but ultimately
26:02.074 --> 26:06.286
you're not holding up your end of the bargain and you're
26:06.298 --> 26:10.374
not providing the means for that person to make
26:10.412 --> 26:14.790
and use it. So this is some interesting advice.
26:15.470 --> 26:19.074
So tools for checking whether 112 F applies or
26:19.112 --> 26:23.322
how to bolster up your specification in
26:23.336 --> 26:27.198
the world that you decide to use meaningful function language is obviously
26:27.284 --> 26:30.762
make sure that the spec denotes structure, right? Make sure you
26:30.776 --> 26:35.000
have structure in your spec for all the functions that those terms have.
26:36.050 --> 26:39.826
Also look to dictionaries both basic dictionaries
26:39.898 --> 26:43.218
and field specific dictionaries to
26:43.244 --> 26:46.842
determine whether the noun that you're using the note structure as
26:46.856 --> 26:50.418
well, you may find that if you look up a
26:50.444 --> 26:53.898
reciprocating member that that's very clearly known to
26:53.924 --> 26:57.490
have some kind of structure associated with it. So it's
26:57.550 --> 27:00.586
better that that's the case and the alternative
27:00.718 --> 27:04.834
of it not having any structure and then again find evidence of that structure
27:05.002 --> 27:08.974
for the term in the prior art. So is there prior
27:09.022 --> 27:11.862
art references, literature, patents or whatever,
27:12.056 --> 27:15.658
that when people have used that term it can notice some kind of structure.
27:15.754 --> 27:19.218
And again, so all these things are true that even if
27:19.244 --> 27:22.880
you use meaningful function language you should be in a pretty good spot for it
27:23.210 --> 27:27.226
having the support it needs to be found definite. So turning
27:27.238 --> 27:30.186
a little bit to software means function. So there was a few different case and
27:30.188 --> 27:33.978
I kind of put these through these all in one slide just because the
27:34.004 --> 27:36.330
short of it is that a lot of them were found definite and I'm not
27:36.380 --> 27:40.270
surprised. So there was a Dyson versus target
27:40.330 --> 27:44.010
in 2022, this one was found definite. The meaningful function
27:44.060 --> 27:47.454
terms were code, application and system and
27:47.492 --> 27:51.414
VDP versus Visio. Again, Federal Circuit 2022,
27:51.572 --> 27:55.482
storage and processor were found to be definite. And for these it
27:55.496 --> 27:59.238
basically turned on the fact that in
27:59.264 --> 28:03.306
some of the testimony that one of Skilling art would understand
28:03.488 --> 28:06.882
that these terms had some sort of structure, especially in
28:06.896 --> 28:10.806
the system, one was targeted. It was something around like you're in a
28:10.868 --> 28:14.242
target, you're walking around, you're over by the vacuums.
28:14.266 --> 28:18.102
You all of a sudden get a vacuum specific advertisement or coupon or whatever for
28:18.116 --> 28:21.994
the vacuum. It allowed for different targeted advertising essentially
28:22.042 --> 28:25.846
within a combined structure. So in this case, the system was essentially
28:25.918 --> 28:28.866
kind of a system within a building,
28:29.048 --> 28:32.322
the building having kind of different wireless detectors and
28:32.516 --> 28:36.010
things like that. So it's very clear that that system had structure.
28:36.130 --> 28:39.658
And then one of Skill in the art from a code, application, storage and processor
28:39.694 --> 28:43.158
perspective understands what those things are. I don't know to
28:43.184 --> 28:46.378
what degree they describe the processor.
28:46.414 --> 28:50.770
For example, if it was just clear that it was an off the shelf processor
28:50.830 --> 28:54.934
or if it was some kind of configured processor, maybe they had algorithms
28:54.982 --> 28:59.190
in there to support that processor nonsword.
29:00.050 --> 29:03.610
But suffice it to say it was sound definite. When you look at Blackboard
29:03.670 --> 29:06.318
versus Desire to Learn in 2009,
29:06.464 --> 29:09.694
they use an Access Control Manager
29:09.742 --> 29:13.258
language and again, this one was found indefinite.
29:13.294 --> 29:16.626
It's because they could not find any structure that a person's skill in
29:16.628 --> 29:19.818
the art would have understood to have structure. And so one of the things that
29:19.844 --> 29:22.902
they basically said, I think Kristin would agree with
29:22.916 --> 29:26.422
this is that you have to avoid using thin disclosures
29:26.506 --> 29:29.838
for even those elements known in the art. And I think
29:29.864 --> 29:33.574
that especially goes towards these top ones, the Dythem versus
29:33.622 --> 29:38.130
Target and Vdpp versus Visio is that obviously code,
29:38.240 --> 29:42.954
application, storage, processor are pretty well known technology
29:43.052 --> 29:47.014
in the art. That does not mean that you are exempt
29:47.062 --> 29:50.682
from describing them. To enable somebody in the
29:50.696 --> 29:54.498
art, you need to still provide. Is it like Kristen said,
29:54.524 --> 29:58.474
is it some kind of generic computer? Is it some kind of more specifically
29:58.522 --> 30:04.702
programmed processor? And in the case that it is specially
30:04.786 --> 30:08.674
programmed, what are the algorithms that allow it to be specially programmed? If it's generic,
30:08.782 --> 30:12.306
what are the algorithms that is performing that allow it to act as an Access
30:12.368 --> 30:14.300
Control Manager or something like that?
30:15.290 --> 30:18.490
Right? Yeah. And we don't get dinged as often on storage,
30:18.550 --> 30:21.838
system, code, processor, that sort of thing, because we put boilerplate
30:21.874 --> 30:25.758
in for that for exactly that reason. But you could make your
30:25.784 --> 30:29.934
boilerplate for an Access Control Manager just by going through one
30:29.972 --> 30:33.774
iteration of how you're controlling access in
30:33.812 --> 30:37.474
one paragraph. And you could probably pull in each component that's
30:37.522 --> 30:40.340
involved, whether it's software or hardware or both.
30:42.590 --> 30:44.720
You have to remember to do it right.
30:45.830 --> 30:50.358
We often we get writing a. Black diagram because we
30:50.384 --> 30:53.946
need a name to call something. And so we have a
30:54.008 --> 30:57.954
manager here and an engine here and an application here and we
30:57.992 --> 31:01.534
forget sometimes to put the details in that and then we forget
31:01.582 --> 31:05.300
to tie it together too. So there's kind of two components there.
31:05.990 --> 31:09.560
Yeah, and I think in this case, if I recall correctly, but I might be
31:11.090 --> 31:15.654
mixing it up with another one. But this is one where they had some
31:15.692 --> 31:19.710
access control support description effect,
31:19.880 --> 31:23.154
I think around passcodes and things like that. But then
31:23.312 --> 31:27.306
for the specific function or one of the functions that the Access
31:27.368 --> 31:30.702
Control manager controlled, there was
31:30.716 --> 31:34.662
no tying in the spec of that passcode function to that other function
31:34.856 --> 31:38.790
or something. Again, it was one of those weird things where they had said,
31:38.840 --> 31:43.054
well, it's our understanding that this is all the functions of the Access Control manager.
31:43.162 --> 31:45.620
And while, yes, the passcode is one,
31:46.910 --> 31:50.278
how that works is one structural component
31:50.314 --> 31:53.454
for these functions, it doesn't suffice for these other
31:53.492 --> 31:56.586
functions. So again, I think that's where it
31:56.588 --> 31:59.890
gets tricky too. I think that's a big interpretation
31:59.950 --> 32:03.078
part is what are all the functions that they are going to believe,
32:03.164 --> 32:08.190
like the courts or whatever that term
32:09.050 --> 32:12.738
provides, all those functions that it does? And are those the same functions that
32:12.764 --> 32:15.910
you believe that it's doing? Because if those aren't in alignment,
32:15.970 --> 32:19.858
then you're going to have a definite disclosure
32:19.954 --> 32:24.186
issue. Yeah, that's exactly in line with the
32:24.368 --> 32:28.122
advice that I got on means plus function
32:28.256 --> 32:32.060
language just in general in claims is to really
32:32.570 --> 32:36.658
mirror that language almost exactly in the specification.
32:36.814 --> 32:41.674
So if you have like means for comparing
32:41.842 --> 32:45.610
this input to this output, then in the spec you would say means for comparing
32:45.670 --> 32:49.494
this input to this output can include, you know what I mean, and then you
32:49.592 --> 32:53.120
list it out really explicitly with all your structure and stuff like that.
32:53.750 --> 32:57.178
It sounds like kind of similar to what Christmas saying on the boilerplate
32:57.214 --> 33:00.706
language and that some of these cases that Ashley,
33:00.718 --> 33:04.242
you've been going over, part of it is not having the support, but part
33:04.256 --> 33:09.226
of it is not being clear what the means scope
33:09.358 --> 33:13.160
is in a way. Yeah, I would agree.
33:15.710 --> 33:19.470
When I draft, I still really like to do
33:19.520 --> 33:22.902
claims early on and then I plop those into the spec.
33:22.976 --> 33:26.158
That's not to say that I'm not going to add 16
33:26.194 --> 33:30.282
paragraphs up front to kind of give breath and different
33:30.356 --> 33:33.414
versions of things and kind of this overarching view of everything.
33:33.572 --> 33:37.674
But then when you do that, you at least have that what everybody
33:37.712 --> 33:41.706
has agreed to as the most important thing, at least currently.
33:41.888 --> 33:45.620
And then you can build all the alternatives but kind of within
33:46.490 --> 33:49.374
making sure you have that claim language support,
33:49.532 --> 33:53.000
then also can tie those alternatives directly to
33:53.330 --> 33:58.738
the base language. And so in
33:58.884 --> 34:02.734
my experience, it's such a pain when you do things in reverse because
34:02.772 --> 34:06.878
then if you have a spec and you're doing claims,
34:07.034 --> 34:10.538
then you have to make sure you fully import that language
34:10.574 --> 34:13.534
back in there. And it just never ends up, I feel like being as good
34:13.572 --> 34:16.790
as if the claims, at least one set was upfront
34:16.850 --> 34:19.200
and then the spec drafted around those.
34:20.130 --> 34:23.774
So Ashley, sometimes what I end up doing is adding
34:23.822 --> 34:27.290
the claims we generate after to the specification,
34:27.350 --> 34:31.190
but I try to add it in maybe its own examples
34:31.250 --> 34:34.606
so I don't conflate what I had already
34:34.668 --> 34:38.098
done. There are obvious things where you only change
34:38.124 --> 34:41.758
a few words and you just go back and you change the vocabulary. But if
34:41.784 --> 34:44.530
we're adding like ten new claims,
34:44.910 --> 34:48.518
I will often just say, okay, we're going to put this in the flow
34:48.554 --> 34:52.214
chart and it's going to be another two page description.
34:52.262 --> 34:55.862
Just because you never know, right, and maybe you wrote
34:55.886 --> 35:00.178
that two weeks ago. Right. So you really don't recall what
35:00.204 --> 35:02.950
you said? No, that's a good point.
35:03.000 --> 35:06.566
Yeah. Most clients, I would bet, would prefer
35:06.698 --> 35:09.850
an extra few hours spent on an extra
35:09.900 --> 35:13.166
few pages for really making sure claims
35:13.178 --> 35:16.874
were supported versus those same hours spent
35:16.982 --> 35:20.422
trying to smoosh it somewhere into the spec, but then coming out
35:20.436 --> 35:22.730
of it being like, well, I think they're supportive.
35:22.910 --> 35:25.894
Right. Well,
35:25.932 --> 35:29.338
and by that same token, if you have anybody reviewing work like you and I,
35:29.364 --> 35:32.782
once in a while we'll do that and each of us will
35:32.796 --> 35:37.042
have to go back and massage the language, add some content,
35:37.236 --> 35:40.378
or at least outwardly say, like I think I told you this week,
35:40.464 --> 35:43.438
hey, I haven't added this to the spec yet, see what you think of the
35:43.464 --> 35:47.242
claims. So at least we
35:47.316 --> 35:51.382
try to not duplicate the work, but right.
35:51.516 --> 35:54.938
No, it's true. Yeah, it's a fine balance,
35:54.974 --> 35:58.618
that's for sure. So then there's a few more cases here.
35:58.644 --> 36:01.918
There was Media Rights capital versus Capital One Financial Corp. And this
36:01.944 --> 36:05.690
was Media Rights had invented software for preventing
36:05.750 --> 36:09.478
the unauthorized recording of electronic media. And so
36:09.504 --> 36:12.670
they had used terms like compliance mechanism and a
36:12.720 --> 36:16.402
custom media device. And so
36:16.596 --> 36:20.474
what the core had basically said was that the spec does describe how the compliance
36:20.642 --> 36:23.998
mechanism is connected to and interacts with
36:24.024 --> 36:28.226
the other components of the system, what processes the compliance
36:28.298 --> 36:31.658
mechanism performs, and what structural sub components
36:31.694 --> 36:35.326
might comprise the compliance mechanism. But they said ultimately the
36:35.328 --> 36:39.134
mechanism was tied to a general computer and there was no corresponding
36:39.182 --> 36:42.334
algorithm for all the four functions. And so they still
36:42.372 --> 36:45.840
held it indefinite. This one surprised me a little bit just because,
36:46.830 --> 36:50.714
as it says there verbatim from the synopsis
36:50.762 --> 36:54.178
was that it said how it was connected with other
36:54.204 --> 36:58.034
components, how it interacted with other components, what processes
36:58.082 --> 37:01.282
it performed with the structural components components might be,
37:01.416 --> 37:05.498
but there was no algorithms for each of the four functions.
37:05.654 --> 37:09.214
And so that just goes to show even when you kind of
37:09.312 --> 37:13.222
place it into the ecosystem, of the application of this is how it
37:13.236 --> 37:17.206
kind of sits in the bigger picture of everything. If you still
37:17.388 --> 37:21.446
don't vet out those four functions or however many functions it performs,
37:21.578 --> 37:24.770
clearly vet them out, you're still going to be in trouble,
37:24.830 --> 37:28.166
probably. And this is Tech SEC versus
37:28.238 --> 37:31.522
International Business Machines or IBM. So Tech SEC had
37:31.536 --> 37:34.954
invented computer systems for securing computer data.
37:35.112 --> 37:38.410
And so they used words like system memory means
37:38.460 --> 37:41.854
and digital logic means. And so, again,
37:41.892 --> 37:45.254
the court held that the system memory is a specific structure,
37:45.422 --> 37:48.770
digital logic means specifically disclosed in the specification
37:48.890 --> 37:52.442
to comprise structural elements, including system memory and specific modules
37:52.466 --> 37:53.650
and subsystems.
37:56.070 --> 37:59.818
I didn't know this. Kristen probably knows this. But also,
37:59.964 --> 38:03.542
going back to what I said earlier around, if there are dictionaries,
38:03.626 --> 38:07.166
especially field specific dictionaries,
38:07.298 --> 38:10.810
look in there and see what structure
38:11.490 --> 38:14.830
that dictionary puts together with whatever
38:14.880 --> 38:18.142
term you're looking up. So in this world, because the first time I read this,
38:18.156 --> 38:21.602
I was like, structural elements, system memory, specific modules.
38:21.626 --> 38:25.294
I was like, how was that definite? That seems just BS to me.
38:25.332 --> 38:28.930
Right? But when you look up, I guess, digital logic,
38:29.250 --> 38:32.902
a skilled artisan would know that to mean digital circus that
38:32.916 --> 38:39.058
perform Boolean algebra, like it's gates or
38:39.084 --> 38:42.178
gates, basically. Fair enough.
38:42.264 --> 38:45.694
All right. It's weird. It's weird to think of that as
38:45.732 --> 38:49.162
actual hardware, right? Because it's all on our computers. We think it's running in
38:49.176 --> 38:52.910
software, and some of it is, right. You do have software enabled processing.
38:53.030 --> 38:56.594
It's not all just hardware.
38:56.702 --> 38:59.954
You can make these algorithms that function on hardware,
39:00.002 --> 39:03.118
but they function as a soft processor. Right.
39:03.264 --> 39:04.920
So it's not fully clear.
39:06.810 --> 39:10.620
And that's why we need a definite structure. Yeah.
39:11.670 --> 39:15.926
All right. And then the last case I had was NOAA systems versus intuit.
39:16.058 --> 39:19.454
The NOAA system basically described
39:19.502 --> 39:23.062
an automated financial accounting system. The system would
39:23.076 --> 39:26.806
allow businesses or individuals to connect
39:26.868 --> 39:30.818
to the computers of companies with which that entity conducts
39:30.854 --> 39:34.898
business so that information regarding financial transactions could be transmitted
39:34.934 --> 39:38.822
between them. So I kind of liken this to like you go into Wise
39:38.966 --> 39:42.850
or even like your bank has these modules where you can put in information
39:42.960 --> 39:46.680
and then you can connect between different banks or things like that.
39:47.370 --> 39:51.178
So they had described access means and communication means
39:51.264 --> 39:55.298
in their claims, but there was no algorithm disclosed
39:55.334 --> 39:59.062
for one of the two functions for these. And so they
39:59.076 --> 40:02.594
basically said there are really two functions resided, providing access to the file,
40:02.702 --> 40:05.978
and then once access is provided, enabling the performance
40:06.074 --> 40:09.562
of delineated operations. And what
40:09.576 --> 40:12.900
they were saying is an algorithm had to address both of those,
40:13.830 --> 40:17.050
but unfortunately, that was not the case. They didn't have the support for it.
40:17.100 --> 40:20.774
And so the court basically said, to avoid purely functional
40:20.822 --> 40:23.966
claiming, in cases involving computer implemented inventions,
40:24.098 --> 40:27.694
we have consistently required that the structure disclosed in the spec being more
40:27.732 --> 40:30.538
than simply a general purpose computer. And then,
40:30.564 --> 40:34.562
consequently, a means of function claim element for which the only disclosed structure
40:34.586 --> 40:38.150
is a general purpose computer is invalid if the specification
40:38.210 --> 40:41.806
fails to disclose an algorithm for performing that function.
40:41.988 --> 40:45.720
So again, just really hit home run like
40:46.050 --> 40:52.838
dead on its face, like super clear of what they want anyway.
40:52.924 --> 40:57.962
So the kind of some of the tips that I kind of thought about I
40:57.976 --> 41:01.994
don't do this, I don't do mental function very often because in
41:02.032 --> 41:06.354
general I was kind of brought up in the world that it's
41:06.462 --> 41:09.770
just not something you do. I do recognize the value in having,
41:09.820 --> 41:13.838
like I said, maybe a claim set that has some meaningful function language but
41:13.984 --> 41:18.234
I would sprinkle in meaningful function, make sure it's not the only claim
41:18.282 --> 41:21.854
type but you can maybe sprinkle it in a little bit and assume that
41:21.892 --> 41:25.622
terms will be construed as MPs and dropped accordingly. So now in
41:25.636 --> 41:29.454
this world where if you use the term
41:29.502 --> 41:33.038
means 112 bath certainly applies. If you don't use
41:33.064 --> 41:35.500
means, there's a strong presumption that it doesn't apply.
41:36.670 --> 41:40.334
That's no longer the truth, right? That's no longer the case. If you
41:40.432 --> 41:44.562
use non means terms there's
41:44.586 --> 41:47.798
still a likelihood that it can be construed as a means plus function
41:47.884 --> 41:51.254
term. You kind of have to take this view that anytime you're using
41:51.292 --> 41:55.218
anything that's remotely close to an ons word that assume
41:55.254 --> 41:59.222
that it's going to be viewed as means of function either by the examiner or
41:59.236 --> 42:02.750
by a later court body to make sure that you draft accordingly.
42:03.370 --> 42:06.158
And then even structure is known in the art to be given structure in the
42:06.184 --> 42:09.760
description. I think that one kind of surprised me through all of this
42:10.750 --> 42:13.974
research and stuff and then for software include algorithms
42:14.022 --> 42:17.558
for all the functions for that means function term. And so like Kristen said,
42:17.584 --> 42:21.170
even if it's here's the functions, I'm going to do some lame
42:22.150 --> 42:25.538
flowchart for each of them. At least there is something there that kind of
42:25.564 --> 42:29.202
shows what inputs are being received, was being done with those inputs
42:29.226 --> 42:33.090
and what the output is so that you have that very literal
42:33.210 --> 42:37.418
description. And one more thing to add there, I always find
42:37.444 --> 42:40.994
it very interesting that is
42:41.032 --> 42:44.920
kind of the requirement for an application
42:45.430 --> 42:48.998
to get through and to be allowed and then as soon
42:49.024 --> 42:52.730
as you file a continuation, of course your flowchart is
42:52.840 --> 42:56.342
nonexistent because you can't add new matter but you happen to pull it
42:56.356 --> 42:59.942
from the spec language. But they're fine with that in a
42:59.956 --> 43:02.450
continuation or two or three or four or ten,
43:02.500 --> 43:05.798
right? So it's kind of interesting as you
43:05.824 --> 43:10.298
prosecute further in a family that just gets a little lighter for
43:10.324 --> 43:13.998
what they require adding flowcharts
43:14.034 --> 43:17.774
later, do you mean you can't? So if
43:17.812 --> 43:21.014
you were to write a set of claims that doesn't have a flow chart because
43:21.052 --> 43:24.926
now let's say you write that has
43:24.988 --> 43:28.538
five different end steps than your first flow chart and
43:28.564 --> 43:32.018
first claim you have support for it
43:32.044 --> 43:35.930
likely because you described it. But like
43:35.980 --> 43:39.702
in Europe, if you didn't describe it in one embodiment and you're
43:39.726 --> 43:43.490
pulling that from five areas of the spec, it won't fly.
43:44.050 --> 43:47.622
So in the US. They really kind of say it's
43:47.646 --> 43:51.158
fast and loose, it's fine, it's in there you have the
43:51.184 --> 43:55.170
language even though you didn't tie it to the same permutation.
43:55.350 --> 43:58.240
Right. So it's interesting.
43:58.810 --> 44:02.634
It's also why I don't often do a lot of divisionals
44:02.682 --> 44:05.994
in Europe unless it's a specifically different claim
44:06.042 --> 44:09.858
set than what I've already gotten allowed. Right. Well, this gets so increasingly
44:09.894 --> 44:12.520
expensive too. Yes,
44:13.510 --> 44:15.820
it's tricky. I think that's where you also have to,
44:17.410 --> 44:20.440
if the budget and time allows and all that good stuff,
44:21.370 --> 44:26.294
working with the inventor of the companies to understand this
44:26.332 --> 44:30.474
is the current iteration. But what do future iterations
44:30.522 --> 44:33.722
and if there's one, you know, like we're for sure going to
44:33.736 --> 44:36.638
do this in the next two years. The other way of doing it, we're only
44:36.664 --> 44:40.082
doing it this way now because it was the cheapest, fastest way to market but
44:40.096 --> 44:43.574
we're going to do it the other way later than making sure you have the
44:43.612 --> 44:47.606
flow charts for that later implementation or at least
44:47.668 --> 44:51.002
really good support to avoid this kind
44:51.016 --> 44:54.158
of stuff. Yeah. The other thing that I was going to add
44:54.184 --> 44:57.762
is like I find myself using functional language
44:57.906 --> 45:01.998
I don't know about often. But sometimes. Especially if a client
45:02.034 --> 45:05.474
has described a few different alternatives to do
45:05.512 --> 45:09.138
something and they really want the breadth in the independent
45:09.174 --> 45:12.974
claim I'll use the functional language to kind of cover
45:13.132 --> 45:17.490
three or four different embodiments but then a lot of times I'll use dependent claims
45:17.610 --> 45:21.038
to clarify that we are in this
45:21.064 --> 45:24.402
thing is then it's the functional language
45:24.426 --> 45:28.058
and use the structural language and the dependent claims. And I guess
45:28.204 --> 45:32.418
my thought or hope is that it's
45:32.454 --> 45:36.698
even further clarifying and supporting the
45:36.724 --> 45:40.550
definiteness of the functional language in the independent claim.
45:41.170 --> 45:44.622
And remember, it's not bad language, it's just not bad language.
45:44.646 --> 45:47.162
You just have to take care when using it.
45:47.296 --> 45:50.934
And if you get to a point where it's just easier
45:50.982 --> 45:54.566
to write out those dependent claims and then you can only file 20,
45:54.628 --> 45:57.710
just pull them out and put them in the stock only. There's nothing
45:57.760 --> 46:00.998
wrong with that. Yeah, I really
46:01.024 --> 46:04.322
like that. I think we as partitions kind of said
46:04.396 --> 46:07.446
forget the value independent claims
46:07.458 --> 46:10.710
and I mean, obviously from the claim set perspective,
46:10.830 --> 46:14.870
the value but the giving the breast value
46:14.920 --> 46:17.598
of independent claims. I think it's easy to kind of be focused on the dependent
46:17.634 --> 46:22.154
claims from alternative embodiments and almost
46:22.192 --> 46:25.346
an explanation of what different things are, what they mean or
46:25.408 --> 46:28.646
giving structure. But I think kind of forgetting the two
46:28.768 --> 46:32.858
that they also because there's a dependent claim that
46:32.884 --> 46:36.690
is narrower than the independent claim, then by virtue of that dependency,
46:36.750 --> 46:40.360
you have this great breath in the independent claim. So if you can import
46:40.750 --> 46:44.838
that same relationship into the stack and you're also kind of similarly
46:44.874 --> 46:48.642
giving that same breath and relationship and then like that disclosure.
46:48.666 --> 46:52.538
And that definiteness. All right, well, that's all I had for
46:52.564 --> 46:56.150
you all today. I'm glad that nobody ever seemed
46:56.770 --> 47:00.760
participatory and intuitive. Nobody fell asleep, so that's a win.
47:02.530 --> 47:06.278
Now, using this language and software makes me sweat. I do
47:06.304 --> 47:09.710
it because I've worked with many older practitioners, like old
47:09.760 --> 47:13.470
school practitioners who like it and want that coverage,
47:13.590 --> 47:17.246
but every time I do it, I have to double check everything
47:17.308 --> 47:20.020
I've done. When it comes to a claim like that,
47:22.150 --> 47:25.862
yeah, I haven't done much. I think the one that comes to mind was
47:25.996 --> 47:29.994
like language on a sensor module. But I think that has because the sensor
47:30.042 --> 47:33.590
is kind of the modifying term that I think it
47:33.640 --> 47:36.782
connotes a decent amount of structure that obviously they're sensors. I mean, you still have
47:36.796 --> 47:40.634
to kind of say what that sensor module is doing and what things
47:40.732 --> 47:44.450
perform that function. In that case, you're probably tying it to some specific sensor.
47:50.090 --> 47:54.090
Just thinking aloud here, this whole discussion
47:54.770 --> 47:59.420
seems to be similar or
47:59.990 --> 48:03.738
analogous in a way to the question of product
48:03.824 --> 48:07.590
by process versus product composition.
48:10.590 --> 48:15.060
Do you see that as an analogy? Also product,
48:15.750 --> 48:19.654
you make a product here's a product which is made this
48:19.692 --> 48:24.814
way. I mean, and really it's a
48:24.852 --> 48:29.122
method claim and you
48:29.136 --> 48:33.430
know, it's much fuzzier than if you just recite the composition.
48:36.070 --> 48:39.820
I haven't kind of looked into case law around product by process.
48:41.230 --> 48:45.374
If a product
48:45.412 --> 48:49.194
by process claim if somebody could somebody be found infringing
48:49.242 --> 48:52.802
if they did a similar process but didn't end up with the
48:52.816 --> 48:56.138
product in the chemical arts, I imagine that could be a place
48:56.164 --> 49:00.842
where that happens. Probably not in a lot of other arts, but obviously
49:00.916 --> 49:03.830
it's assuming you're performing all the process steps.
49:04.150 --> 49:08.198
My understanding of product by process claims is that the
49:08.224 --> 49:12.110
process limitations are more or less irrelevant
49:12.910 --> 49:16.514
and that if someone makes the same product
49:16.612 --> 49:20.210
by a different process, or if that's in the prior art,
49:20.320 --> 49:22.790
let's say, then the claim wouldn't be allowable.
49:23.410 --> 49:27.014
So I almost never really never use
49:27.052 --> 49:30.558
those for that reason. But again, I'm not that familiar
49:30.594 --> 49:33.220
with all the latest case law on that either.
49:33.670 --> 49:37.206
Yeah, okay. Sounds like an opening. David for David Cohen
49:37.278 --> 49:40.790
for another presentation.
49:43.610 --> 49:45.080
Let me walk that back.
49:47.390 --> 49:50.658
Hey. Yeah, I like it.
49:50.684 --> 49:54.440
I would love to hear something on that because I hardly ever run across it.
49:55.370 --> 49:58.954
Well, thank you everybody, I appreciate it. Was a good dynamic discussion,
49:59.002 --> 50:01.950
so I love it. Thank you, Ashley.
50:02.690 --> 50:06.534
Thanks Ashley. It's a great topic. Yeah, thanks everybody. All right,
50:06.572 --> 50:10.914
that's all for today, folks. Thanks for listening. And remember to check us out@aurorapatins.com
50:10.952 --> 50:14.070
for more great podcasts, blogs, and videos covering all things
50:14.120 --> 50:17.382
patent strategy. And if you're an agent or attorney and would like to be part
50:17.396 --> 50:20.674
of the discussion, or an inventor with a topic you'd like to hear you're discussed,
50:20.782 --> 50:24.582
email us at podcast@aurorapatant.com. Do remember that
50:24.596 --> 50:28.078
this podcast does not constitute legal advice. And until next time, keep calm
50:28.114 --> 50:28.800
and patent on.
