PersonalWeb Tech v. Apple (Fed. Cir. 2017) [personalweb]
Following an administrative trial, the PTAB found the challenged claims of PersonalWeb’s Patent No. 7,802,310 obvious based upon a combination of two prior art references. On appeal, however, the Federal Circuit has vacated the judgment – holding that the Board’s factual findings were not supported by substantial evidence. In particular, the Federal Circuit could not find substantial evidence for the conclusions (1) that the prior art taught each element of the challenged claims or (2) that PHOSITA would have been motivated to combine the references to form the invention as claimed.
Substantial Evidence: For me, the substantial evidence rule is a misnomer because it suggests that a decision is based upon a substantial amount of evidence (and thus seems like a reasonable approach). I would rename it as the “scintilla rule” because, under the rule, factual conclusions made by the PTAB are affirmed on appeal so long as supported by “more than a mere scintilla” of evidence. I previously wrote:
[T]o be clear, the appellate court will affirm a factual determination even when that determination is likely wrong so long as some amount of evidence supports the determination. We have all heard the adage that ‘reasonable minds can differ.’ I think of the substantial evidence rule as a relaxed version that ‘somewhat reasonable minds can differ.‘ Or, in other words, the PTO’s factual determinations will be affirmed if somewhat reasonable.
Although the PTAB burden is not great, the Board must fully explain its conclusions and their bases. (This is a major difference when compared with the ‘black box’ of a jury trial). As part of its explanation, the Board must also specifically contend with evidence that “detracts from an agency’s decision.” The court explained in its 2002 Lee decision:
For judicial review to be meaningfully achieved within these strictures, the agency tribunal must present a full and reasoned explanation of its decision. The agency tribunal must set forth its findings and the grounds thereof, as supported by the agency record, and explain its application of the law to the found facts.
In re Lee, 277 F.3d 1338, 1343 (Fed. Cir. 2002). That is, the Board, “must articulate ‘logical and rational’ reasons for its decision.” (quoting Synopsys).
Here, the court explained that the rational in an obviousness case includes
- Particularly identifying where each element of each claim is taught by the prior art references. Here, the Board failed to explain the particular location within Woodhill that taught the claim limitation of comparing a name with a “plurality of values.” (yes, that’s right.)
- Explaining the motivation to combine.
Explaining motivation to combine is a big deal because it comes up in almost every patent case before the PTO. Here, the Board simply wrote that:
“a person of ordinary skill in the art reading Woodhill and Stefik would have understood that the combination of Woodhill and Stefik would have allowed for the selective access features of Stefik to be used with Woodhill’s content-dependent identifiers feature.”
On appeal, the Court found that justification inadequate.
Indeed, the Board nowhere clearly explained, or cited evidence showing, how the combination of the two references was supposed to work. At least in this case, such a clear, evidence-supported account of the contemplated workings of the combination is a prerequisite to
adequately explaining and supporting a conclusion that a relevant skilled artisan would have been motivated to make the combination and reasonably expect success in doing so. . . .A brief explanation may do all that is needed if, for example, the technology is simple and familiar and the prior art is clear in its language and easily understood. See Ariosa, 805 F.3d at 1365–66. On the other hand, complexity or obscurity of the technology or prior-art descriptions may well make more detailed explanations necessary. Here, the Board’s explanation is wanting. Apple’s attempts in this court to explicate both the Board’s explanation and the underlying evidence do not persuade us otherwise.
On remand, the Board will give it another go – and we’ll see whether the exercise of actually explaining its reasoning causes the Board to change its mind as well.
= = = = =
My explanation of the case skipped over details of the patent at issue. Claim 1 is shown below and basically includes three steps: (a) sending a “content based name” for a data-item from one computer to another; (b) check to see if the name is on a list; and (c) grant or deny access to the data-item depending upon the result of (b).
1. A computer-implemented method … comprising the steps:
(a) at a first computer, obtaining a content-based name for a particular data item from a second computer …, the content-based name being based at least in part on a function of at least some of the data which comprise the contents of the particular data item, wherein the function comprises a message digest function or a hash function, and wherein two identical data items will have the same content-based name; and
(b) … a processor at said first computer ascertaining whether or not the content-based name for the particular data item corresponds to an entry in a database comprising a plurality of identifiers; and
(c) based at least in part on said ascertaining in (b), determining whether or not access to the particular data item is authorized.
The patentee claims that Apple’s iTunes and iCloud services infringe.
A couple of days ago, I was looking over comments on another old thread on this site. This was an exchange between two regular comment posters (not identified here by name). An earlier comment mentioned “Graham v. John Deere” as the only post-1952 SCOTUS opinon concerned with obviousness under 103. A little later, a comment mentioned “Anderson’s Black Rock” and “Sakraida”, whereupon the other poster said that he did not concern himself with pre-1952 SCOTUS precedents, which were irrelevant in so far as they had allegedly been swept away by the 1952 act (or words to that effect – I am going by memory). So far as I could see, that earlier comment thread did not include any correction pointing out that “Anderson’s Black Rock” was actually decided in 1969, and “Sakraida” in 1976, but, as the Wikipedia article on “Anderson’s Black Rock” points out, CAFC did not deign to cite or follow these holdings until decades later, when cert had been granted by SCOTUS on KSR.
The Wikipedia summary of “Anderson’s Black Rock” accords with my memory of reading the opinion: “It stands for the proposition that, when old elements are combined in a way such that they do not interact in a novel, unobvious way, then the resulting combination is obvious and therefore unpatentable.”
This seems to be on point here. If one stores in a database information as to who has access to particular files, one may identify files by appropriately-generated names, but how the names are generated, and whether or not those names are “content-based” or not, seems irrelevant. And similarly using hash functions and the like to generate filenames, or partial filenames seems “obvious”, even without explicit prior art. So it seems difficult, at least to me, to envisage how one could argue that this claim is more than the predictable use of prior-art elements according to their established functions.
If one stores in a database information as to who has access to particular files, one may identify files by appropriately-generated names, but how the names are generated, and whether or not those names are “content-based” or not, seems irrelevant. And similarly using hash functions and the like to generate filenames, or partial filenames seems “obvious”, even without explicit prior art. So it seems difficult, at least to me, to envisage how one could argue that this claim is more than the predictable use of prior-art elements according to their established functions.
This kind of straightforward reasosing, of course, wipes out >95% of the “logic on a computer” junk that’s out. Therefore, it must be wrong! Welcome to US patent law where we need to listen to the patentee’s expert tell everyone that clearly it’s not obvious because … otherwise someone would have done it before! Also, look at all the money the deep pocketed person is making! That just proves that they deserve to lose to the plaintiff, who is merely incredibly rich. That third house isn’t going to pay for itself, after all.
The other way to approach the issue, of course, is to simply recognize that data is data. The instructable computer — the actual (old) “technology” — doesn’t care what meaning the human layers on top of the data. It could be “content data” or “Granny’s favorite DVD data” or “authorization access data” or “move the cartoon mouth data”. The computer doesn’t care, just as the telescope doesn’t care whether it’s “looking” at your neighbor sunbathing or a solar eclipse. The logic-executing machines is just a machine that carries out logic operations on input data, according to instructions written by people who understand logic and who understand the language the instructions need to be written on.
The meanings layered on top of data are abstractions and they should either be ignored when looking at obviousness or or it should be recognized that what is being protected by the patent claim is not “technology” but the abstraction layered over the technology (in which case ineligible). Same result either way: junk is junk is flushed down the tubes. And the analysis is pretty much exactly the same, too. Just different words and — oh, I almost forgot — in the latter case we don’t have to spend a half million bucks listening to paid experts give speeches regarding “secondary factors.”
“The logic-executing machines is just a machine that carries out logic operations on input data, according to instructions written by people who understand logic and who understand the language the instructions need to be written on.
The meanings layered on top of data are abstractions and they should either be ignored when looking at obviousness or or it should be recognized that what is being protected by the patent claim is not “technology” but the abstraction layered over the technology (in which case ineligible). Same result either way”
this reply fits here as well (in addition to pointing out that Malcolm’s “Same result either way” is nothing more than a thinly veiled
the Ends justify the Means, so ‘whatever'”):
So how is your Britney Spears CD doing running your computer operating system…
Oh wait, apparently there IS some functional relationship and difference that you want to obfuscate about.
Funny that too, this is exactly the point upon which you have volunteered an admission against interests as to knowing and understanding the controlling law (the exceptions to the judicial doctrine of printed matter).
Prof. Crouch, is this not tantamount to L I E ing for Malcolm to so constantly prevaricate on a point of law that he has relinquished?
This is not even at a level of him wanting (or suggesting) a change to the law.
This is him openly advocating a position unsustainable given his own admissions.
I seem to remember reading somewhere that this site was “America’s leading patent law source” – seems not to be the case when the law is trampled so routinely as Malcolm does…
Not to mention MM that if you sell patented computers, and every use of those computers intended to be supported by their designers is actually considered an ‘improvement” to your patented device, your patent right is whittled away to nothing, because nobody can actually use your patented device without running into a thicket of other “rights” holders to do so.
Yup.
The programmable computer is a tool for applying logical operations to data. The tool is a proxy for the human mind. Humans decide what logical operations they want the computer to carry out and they instruct the computer by writing those instructions down in a language that the computer understands.
Watching the patent maximalists kick dust over these basic facts is fascinating, and also informative of their lack of integrity and candor. Also very, very well documented.
“The tool is a proxy for the human mind.”
LOL
Anthropomorphication
Your “proxy” position is pure fallacy and just this week I pointed out to you this Zombie of the “Mental Steps Doctrine” that you ascribe to.
Of course, you denied it – only to now put your foot into your mouth.
Classic Malcolm.
“intended to be supported ”
What does that mean? That the designers envisioned every use of the patented computer before they left the factory? They knew someone was going to use one to manage shadow accounts and someone else was going to use them to take dictation and someone else was going to use them look for intelligence in signals received from the VLA in New Mexico?
Please…
I guess that means that the Wright Flyer should not have been patented because the guys that made the cloth intended the cloth to be used to harness the wind and the guys that made the rope intended the rope to be used to pull on things.
Distant, you know why Rich and crew ignored those cases, right? They relied heavily on pre-52 case law which Rich contended was no longer good law because whole point of enacting the new §103 can was to overrule that old case law on invention.
Even today, the IPO puts out the propaganda that §103 overruled that prior case law regarding “invention.” I bet you these people are going to tell Congress a pack of lies on this point, never telling Congress the truth about what the Supreme Court has previously held on this issue.
“Rich and crew“…
A bit of unethical misrepresentation of what Congress enacted, eh Ned?
Tell me again if your duties of not misrepresenting the law during your advocacy efforts…
Anon, listen, as a private citizen GS Rich had a right to his opinion and he had a right to argue his opinion about what the intent and purpose for enacting §103, but as a judge, he was obligated to follow the law. And in this, he failed because he continued to advocate his position and he ignored Supreme Court authority that was directly contrary and inconsistent with his views.
§103 is but one example of many. His Benson case was overturned. He had reasoned that because, in the opinion of the C.C.P.A., a programmed computer was a special machine and therefor eligible under 101 that a mathematical process that was conducted on a computer must also be eligible subject matter. And of course, his logic was impeccable if the premise was true. But when the Supreme Court overturned his ruling, did he ever question his premise? No. Rather he doubled down on it, citing this prior C.C.P.A. authority in Alappat, and then citing Alappat in State Street Bank.
My God the man was a piece of work, and perhaps the worst judge this country has ever seen, not because he was not smart or well-educated or otherwise a competent lawyer. No. He was a bad judge because he was unethical.
Obligated to follow the law – the law he knew better than anyone else…
Ned, your protest is laughable.
By the way, in case you still haven’t grasped the issue: it is not the Supreme Court that gets to write the statutory law that is patent law.
(You also get it wrong with Benson, seeing as he knew better than anyone else what the Court’s dicta in Benson was – and was not – controlling over, which led directly to Diehr and Chakrabarty for a very politicized-policy driven Court.
Are you still trying to place the Court above the Constitution? Why have you not supplied your State attorney oath? Yes, Ned – we both why, don’t we?
Also, Distant, it would be interesting to note the exact holding of Anderson’s BlackRock:
Key word there, Ned: “inherent”
How many times have I stressed this to you?
‘Start with two piles of identical components, each pile including a radient heat burner, a spreader, a tamper and screed, a chassi, and sufficient fasteners to combine the elements. Configure one pile of components for continuous paving. Now race with the components to pave a portion of a road. The unchanged original components that were not configured cannot do what the configured components can do. Thus, the new capability shows that the unconfigured pile cannot do what the configured pile does, and thus the arrangement itself is a machine component. Further, this shows that the supreme court was wrong to conclude that the combibation added nothing to the claims of Black Rock. Quod Erat Dicendum.’
How very odd, Ben. I see what you are attempting to do, but you just don’t quite make it.
There are about 21 cases in the Supreme Court that deal with obviousness after 1952, including the very important KSR decision.
adv: SY,DI(obvious! & patent) & DA(aft 1952)
that’s a Westlaw search, by the way
A little later, a comment mentioned “Anderson’s Black Rock” and “Sakraida”, whereupon the other poster said that he did not concern himself with pre-1952 SCOTUS precedents, which were irrelevant in so far as they had allegedly been swept away by the 1952 act (or words to that effect – I am going by memory). So far as I could see, that earlier comment thread did not include any correction pointing out that “Anderson’s Black Rock” was actually decided in 1969, and “Sakraida” in 1976, but, as the Wikipedia article on “Anderson’s Black Rock” points out, CAFC did not deign to cite or follow these holdings until decades later, when cert had been granted by SCOTUS on KSR.
I feel fairly certain that I am the discussion participant who characterized Black Rock and Sakraida as pre-1952. Mea culpa. Clearly those two were decided after 1952. I cannot remember why this point was important to the discussion, but to the extent that it was, clearly everything else I might have said on the subject is unsound by extension.
“clearly everything else I might have said on the subject is unsound by extension.”
That’s a pretty droll attempt to try to make the opposite point, that everything else that you said on the subject somehow must be sound…
😉
I cannot remember why this point was important to the discussion, but to the extent that it was, clearly everything else I might have said on the subject is unsound by extension.
That’s an honorable thing to say. Thus anon must defile it.
Honor had NOTHING to do with such false modesty, Mr. Snyder.
Such false pretenses serve no one.
Having a shot at practicing Claim 1 of the patent.
On a UNIX or Linux box (perhaps accessing remotely over a network), creating a directory with files Tom.txt, Dick.txt and Harry.txt, whether the contents of these files differ.
Adjusting file permissions in standard fashion ad libitum, e.g.,
chown wheel Harry.txt
chmod g+x Harry.txt
Run the following command in Bourne shell (or bash):
for i in *
do
mv $i FILE_`sha1sum $i | sed -e ‘s/ .*$/.txt/’`
done
===
At this stage, or at least if someone in group wheel attempts to access files in this directory, all steps in the claim should have been practiced.
First of all, the directory listing is as follows:
FILE_008a2812c080f119689be2b9c9bc9192fd21c3f6.txt
FILE_0b96bf3bcbb0d720d62d4b4400ba954c885bdb8d.txt
FILE_fc998c59b0a166f506c897ea358239ae8637e7d2.txt
Next, those in group wheel, and only those in group wheel, should be able to “execute” the first of the files.
The abstract of the ‘310 patent makes it clear that a “data item” can comprise a file. All my files have been given a “content-based name” in such a way as to ensure that files with identical content will receive the same “content-based name”. It seems to me that other claim limitations in (a) have been satisfied.
The directory containing these files, in itself, is the “database” of (b). If I type the command
ls FILE_f*.txt
then a processor will ascertain whether or not the “database” contains the relevant “identifier”. If I don’t infringe by listing the directory locally, then I presumably would by accessing the host computer over a network from within or without the jurisdiction.
And distinct users can ascertain whether or not they have access to the data item in accordance with (c).
===
Patent was published in 2010, filed 2007, priority date 1995. Of course there is a long chain of patent applications between 1995 and 2007. And whilst the words “authorize” and “authorized” occur frequently in the claims, they appear comparatively rarely in the specification – as though the author of the specification was unaware back in 1995 of having “invented” the combination of “content-based names” and authorization based on a “database” that is so central to Claim 1, and to the CAFC opinion.
Luckily I am outside the jurisdiction…
I have used standard UNIX and Bourne shell in seeking to implement the claim, because this is really old in relation to 1995. And whilst the SHA1 hash function dates from around that period, the MD4 and MD5 hash functions were developed and made available by Ron Rivest back in 1990 and 1991.
Hindsight bias? When I have come across such hash functions before, it has occurred to me before that they could be used as part of filenames in automatically naming files to avoid clashes. I am sure that many have had such ideas, which are such as surely likely to occur to a competent PHOSITA.
Correction:
chgrp wheel Harry.txt
not
chown wheel Harry.txt
Dennis:
I think you meant –contend– : “, the Board must also specifically content with evidence that “detracts from an agency’s decision.””
and here I think you meant : — Court–: “On appeal, the Board found that justification inadequate.”
Thanks!
“Indeed, the Board nowhere clearly explained, or cited evidence showing, how the combination of the two references was supposed to work.” I am so delighted to see this articulated by a court.
I don’t care how junky these claims are, I practice in the tangible arts (no X configured to do Y limitations), but routinely get rejections with this level of thought put into what it means to modify the primary reference in the suggested manner, usually articulated as “combined according to” or some such nonce expression used to gloss over that detail. It’s called a prima facie case only because a reviewer can check each box, yes, the form paragraph is completely filled out with words, not because those words make some kind of sense. Shout out to the examiners that draw the motivation from the “therefore, what is desired” statement from the last paragraph of the background section of the poorly drafted secondary reference.
That’s right.
For what it’s worth, those “cookie cutter rejections” aren’t the norm but 19 times out of 20 it means that the client is going to end up with the patent claim they want and it’s not going to cost them much more to prosecute to issuance.
I don’t see anything remotely “new” or interesting in this decision. What’s important is to recognize the game being played by the patentee. There is no hope here. And the CAFC did the right thing by remanding.
Malcolm is somehow not “anti-patent” even as he directs his feelings not at poor examination, but rather at those who may dare want to have a patent…
Same old advice, Malcolm: get into a line of work in which you can believe in the work product.
The cognitive dissonance is just not healthy for you.
“comparing a name with a “plurality of values.””
I’m going to go out on a limb here and state that they probably just found it implicitly present.
“a person of ordinary skill in the art reading Woodhill and Stefik would have understood that the combination of Woodhill and Stefik would have allowed for the selective access features of Stefik to be used with Woodhill’s content-dependent identifiers feature.”
Should have been :
a person of ordinary skill in the art reading Woodhill and Stefik would have understood that the combination of Woodhill and Stefik would have allowed for the selective access features of Stefik to be used with Woodhill’s content-dependent identifiers feature … and would have done so because/with the motivation of x.
Alza Corp. v. Mylan Labs., Inc., 464 F. 3d 1286, 1291 (2006)
“There is flexibility in our obviousness jurisprudence
because a motivation may be found implicitly in the prior
art. We do not have a rigid test that requires an actual
teaching to combine . . .
Even if the claims were not pure rubbish, teaching nothing that any PHOISTA would not know from a basic understanding of how a computer works.
The mandatory TSM test is alive and well, despite KSR. Long live the counter-revolution!
It sounds like you are right: tsM is back.
“[O]bviousness concerns whether a skilled artisan not only could have made but would have been motivated to make the combinations or modifications of prior art to arrive at the claimed invention.”
Dennis, know who the attorney was in In re Lee: none other than Richard H Stern.
You and he are quite a team. Stern made the Board (and the examiner) explain itself and you are asking the same of the Federal Circuit with respect to appeals from the patent office in connection with rule 36 affirmances
Something about “blind squirrels” comes to mind…
The claim at issue here is a logic claim. The logic is “determine if something is authorized, wherein said determination comprises comparing input data X to stored data Y.”
The “field of use” is “content-based names” associated with (wait for it) … data.
This is g@rbage. It doesn’t belong in a patent system and the idea that we waste millions of dollars and thousands of hours arguing about whether it’s obvious to instruct a logic processing machine to perform basic logical operations is a farce.
There is no “there” there. It’s just lawyers arguing about words that describe logic. If the IPO has its way, we’ll see lawyers arguing about whether one party is entitled to own legal arguments embodied by logic that’s “written in a book manufacture” or “processed by a computer”. Or maybe just a method of “indexing” those arguments or expressing them as a “data structure.” That’ll be fun, won’t it?
A “logic claim”…
Is that like the (proverbial) logic copyright that I keep asking you about (and then watch you run away – every time)…?
Yeah, thought so.
What a putz.
You keep saying this, and it it keeps making no sense.
You can get a copyright on “logic”. What is a “how-to” book if not a copyrighted item of logic?
And you can get a copyright on your software code. It’s logic, and it’s copyrighted.
You cannot obtain a copyright on logic, young man.
Don’t confuse the protection afforded to a book of math with protection of math itself (to use an analogous example).
Which emphasizes that software is neither logic, nor math.
It may incorporate and use both, but the difference is what drives the protection (for different aspects) in the different IP realms.
Copyright: aspects of expression
Patent: aspects of utility.
That you recognize the IP protection of aspects of expression afforded by copyright, but fail to grasp that the SAME logic applies to protection of the different aspects of utility with the IP protection of patents is a simple logical failure on the part of the “software is not patent eligible” “cohorts.”
You’ve been asked before, “anon”, but — since your silly “argument” rests entirely on this premise — can you show everyone the law that says everything in the universe is either copyrightable or patentable?
That’s a rhetorical question, of course.
Just put a s0 ck in it already. For your own sake.
That is an interesting – albeit worthless – moving of goal posts / straw man fallacy.
No one has ever said anything about “the law that says everything in the universe is either copyrightable or patentable” let alone any type of comment about how this straw man is somehow “your silly “argument” rests entirely on this premise.”
Maybe you want to provide a link, seeing how you like to boast about your “archive” skills of arguments made on this blog…
Yeah, that’s what I thought.
What a putz.
You keep saying this, and it it keeps making no sense.
When has that ever stopped “anon”?
C’mon, Martin. Don’t feed it.
…”it”…
Reminds me of way back when, when Malcolm routinely played the role of “Jamie” of the movie “Silence of the Lambs.”
Yay 11 years of blight!
And just for Mr. Snyder’s enjoyment, your own “logic” here is once again akin to the Big Box of Protons, Neutrons, and Electrons.
You cannot patent those items themselves, but configurations of those very same items are patent eligible all the live long day long.
As everyone knows, this specification discloses precisely zero new configurations of any atoms.
That’s not “my logic.” That’s just a fact.
It happens to be a fact that reveals for the ten zillionth time that you’re a know-n0thing hypocrite who couldn’t argue his way out of a paper bag but don’t let that stop you from reciting your “Magic Box of Protons” script. It’s very deep and very serious! Just like you. We’re all very impressed, as far as you can tell.
You rather miss the point that the Big Box is a slap at your attempted “logic”.
Again.
(Which just goes to show that your ad hominem is clueless and empty ad hominem – yay 11 years of the sAme blight)
your attempted “logic”.
Again: there’s no “attempt at logic.”
There’s just the fact of the complete absence of recited novel structure in your Super Dooper Important Logic Claims, and that fact is (boo hoo hoo!) biased against you. So you wallow in the mud and weep.
[shrugs]
Structure in claims…
Yet again you want to ploy the fallacy that one optional claim format is somehow more than merely optional…
“Go figure Folks”
” and the idea that we waste millions of dollars and thousands of hours arguing about whether it’s obvious to instruct a logic processing machine to perform basic logical operations is a farce.”
But that wasn’t the argument was it. The argument was about whether the PTO showed that it was obvious to check a content-based name against a list, the content-based name being based at least in part on a function of at least some of the data which comprise the contents of the particular data item, wherein the function comprises a message digest function or a hash function, and wherein two identical data items will have the same content-based name.
But that wasn’t the argument was it. The argument was about whether the PTO showed that sky was blue
Fixed to save time.
“This is g@rbage. It doesn’t belong in a patent system and the idea that we waste millions of dollars and thousands of hours arguing about whether it’s obvious to instruct a logic processing machine to perform basic logical operations is a farce.”
So how is your Britney Spears CD doing running your computer operating system…
Oh wait, apparently there IS some functional relationship and difference that you want to obfuscate about.
Funny that too, this is exactly the point upon which you have volunteered an admission against interests as to knowing and understanding the controlling law (the exceptions to the judicial doctrine of printed matter).
Prof. Crouch, is this not tantamount to L I E ing for Malcolm to so constantly prevaricate on a point of law that he has relinquished?
This is not even at a level of him wanting (or suggesting) a change to the law.
This is him openly advocating a position unsustainable given his own admissions.
I seem to remember reading somewhere that this site was “America’s leading patent law source” – seems not to be the case when the law is trampled so routinely as Malcolm does…
Comments are closed.