Tuesday, March 3, 2015

Board okays Examiner's reliance on drawings alone, despite silence as to scale

Takeaway: The Applicant appealed an obviousness rejection In an application directed to a semiconductor structure. The Examiner relied solely on prior art figures to teach relative dimensions as claimed. The Applicable argued that such reliance was improper as a matter of law because the reference did not mention scale. The Board disagreed, finding "sufficient detail and precision [in the drawings] to allow some findings and conclusions with regard to the relative size of the depicted structures. Moreover, Romano's written description supports those findings and conclusions." (Ex parte Sato , PTAB 2014.)

Details:
Ex parte Sato
Appeal 201201276; Appl. No. 11/686,892; Tech .Center 2800
Decided:  March 20, 2014

The application on appeal was directed to an LED. The limitations at issue involved the relative dimensions of structures within the LED.

During prosecution, the Examiner rejected originally filed claim 9 as obvious over Kwak in view of Applicant Admitted Prior Art (AAPA).
     9. A method of manufacturing a semiconductor light-emitting device, the semiconductor light-emitting device including a semiconductor layer and a first electrode (24) on a first surface (11) of a semiconductor substrate (10) and a second electrode (40) on a second surface (12) of the semiconductor substrate (10), the semiconductor layer including a light-emitting  region (22A), the method comprising the steps  of:          
     after forming the semiconductor layer on the first surface (11), forming the first electrode (24) on the first surface (11) so as to correspond to at least the light-emitting region (22A);  
     arranging a recessed section (12A) with a depth larger than the thickness of the second electrode (40) on the second surface (12) so as to form a step section (12B) projected from the recessed section (12A) in a region other than the recessed section (12A) in the second surface (12);
     forming the second electrode (40) on at least the recessed section (12A) of the second surface;          
(Emphasis added.)
For the "depth" limitation highlighted above, the Examiner relied solely on the figures in Kwak – not on the specification of Kwak. In analyzing Kwak Fig. 17, the Examiner mapped via hole h3 to the claimed recessed section and conductive layer 175 to the claimed projection.

The Applicant argued only the dimensional ("depth") limitation of claim 9. However, the Applicant did not distinguish Kwak, instead merely traversing the Examiner's sole reliance on Kwak's drawing. Citing to a variety of cases from the Federal Circuit and CCPA, the Applicant asserted that reliance on the drawings was improper under case law because the specification of the reference was silent on dimensions (absolute and relative). The Applicant further asserted that the specification "fails to indicate that the drawings are to scale" so that "apparently the proportions shown in FIG. 17 are apparently the result of the draftman's selection rather than the embodiment of an intention to represent the device to scale." 

The Examiner maintained the rejection in another Office Action, and continued to rely solely on the drawings. In response to Applicant arguments, the Examiner pointed to MPEP 2125 ("the description of the article pictured can be relied on, in combination with the drawings, for what they would reasonably teach one of ordinary skill in the art"). The Examiner further explained his reading of Kwak:
... Kwak clearly discloses in figure 17 that the depth of the recessed section h3 is larger than the thickness of the second electrode 175, and the one of ordinary skill in the art would easily recognize this relationship from the figure of Kwak. Furthermore, if the second electrode 175 was thicker than the recess h3, then it would be level with the bottom of, or even protrude from the bottom of, layer 150. However, Kwak clearly shows second electrode 175 to be recessed within the recessed section h3. The Examiner submits that one of ordinary skill in the art would easily recognize that the depth of the recessed section is larger than the thickness of the second electrode 175.
Later in prosecution, Applicant added another claim with relative dimensions. This one involved width rather than depth:
     13. A method of manufacturing a semiconductor light-emitting device, the method comprising the steps of:
     forming a first electrode on a projection section, said projection section being a contact layer on a first conductivity type cladding layer;
     forming a second electrode in a recessed section of a substrate, the width of said recessed section being larger than the width of said projection section.
(Emphasis added.)
The Examiner rejected claim 13 as anticipated by or (in the alternative) obvious over Romano. Once again the rejection relied solely on drawings. In analyzing FIG. 15 of Romano, the Examiner mapped area 516 to the claimed recessed section and layers s574 and 576 to the claimed projection.
The Applicant filed an After Final Response and handled the claim 13 rejection similar to the previous handling of claim 9. Namely, the Applicant argued that case law prohibited the Examiner from relying solely on the figures of the reference. In the Advisory Action, the Examiner responded to the claim 13 argument by once again referring to MPEP 2125 and explaining what inferences a POSIA would draw from the figure alone. The Advisory Action concluded with this statement

[O]ne of ordinary skill in the art would easily recognize this [relative width]structural limitation . Hence, it would have been obvious to one of ordinary skill in the art at the time the invention was made to make the width of the recessed section larger than the width of the projection section since Romano clearly suggests this structure in figure 15. (Emphasis added.)

The Applicant appealed the pending claims. Once again, the only argument made for independent claims 9 and 13 was improper reliance on the drawings. For claim 9, the Applicant merely repeated the arguments made in the Response to Final. For claim 13, the Applicant repeated previous arguments but added additional arguments.

One of the arguments newly added for claim 13 addressed two cases cited in MPEP 2125. The Applicant first distinguished In re Aslanian, arguing that Aslanian dealt with a design patent rather than utility patent. The Applicant then addressed In re Mraz by quoting sections in which the appellate court had approved of use of a drawing showing the features at issue "with great particularity," while contrasting this with an "enlarged section of a small drawing obviously never intended to show the dimensions of anything." The Applicant asserted that Romano, like the disapproved drawing in Mraz, "fails to proffer an intent that the width of cavity (416)(516) is larger than the width of layers (474,476)(574,576)."


In contesting the rejection of claim 13, the Applicant also used a teaching of the width limitation in Applicant's own specification to argue that the Examiner had used hindsight:
In the absence of any disclosure within Romano [of the width limitation] or any other objective supporting evidence, the line of reasoning within the Office Action appears to have been merely an extraction from the Appellant's own specification.


The Examiner's Answer basically repeated the earlier rejection of claim 9, but did add an annotated version of Kwak's Fig. 17. The Answer varied the rejection of claim 13 by also relying on an annotated Fig. 14 of Romano.

The Answer also explicitly stated (first the first time) a rationale for the obviousness rejection of claim 13: "in order to maximize the heat dissipation from device, as discussed in Romano [citation to section omitted]." The Examiner also responded to the Appeal Brief arguments of hindsight by asserting (again) that "one of ordinary skill in the art would easily recognize" in Romano's drawing, the width relationship recited in claim 13.

After receiving the Examiner's Answer, the Applicant filed an Amendment cancelling independent claim 9. Independent claim 13, added during prosecution, remained on appeal as obvious over Romano.

The Board framed the issue as "... whether Romano 'describes' [under § 102] or suggests [under § 103,] the claimed relative widths." The Board affirmed and explained as follows:
     [W]e look to the drawings, but consider them in light of the disclosure as a whole to determine what Romano would have disclosed and suggested to one of ordinary skill in the  art. ... The Examiner here is not attempting to use the drawings of Romano to establish precise dimensions, but is citing the drawings for showing generalized proportions. While we agree with Appellants that the drawings are not disclosed by Romano as being to scale, the drawings are drawn with a sufficient detail and precision to allow some findings and conclusions with regard to the relative size of the depicted structures. Moreover, Romano's written  description supports those findings and conclusions. ....In explaining Figure 9, Romano states that the cavity 216 has a length and width larger than that of the p-contacts 240 situated above it. (Internal citations omitted) .... 
... Romano's Figures 10-14 show the progressive steps of forming a semiconductor device with a filled cavity 416. All of the figures depict the various layers and structures within the device with different thicknesses. It is, therefore, reasonable to believe that Romano intended to convey relative differences in thickness even if not exactly to scale. Figures  11-14, like  Figure 9, include multiple p-contacts (440) over cavity 416. Therefore, it is reasonable to believe that, like the p-contacts of Figure 9, the p-contacts 440 of Figures 10-14 are intended to be of smaller width than the width of the cavity 416.
My two cents: Weak Applicant argument. It was clear from the prosecution history that the Examiner did not rely on dimensions, measurements, or values, which means In re Wilson and In re Wright were inapposite. Similarly, the Examiner relied on general rather than precise proportions, so In re Olson and Hockerson-Halberstadt v. Avia were inapposite.

Still, this decision is a good review of the case law governing drawings as prior art.

24 comments:

  1. Similarly, the Examiner relied on general rather than precise proportions, so In re Olson and Hockerson-Halberstadt v. Avia were inapposite

    Whether the Examiner relied upon general rather than precise proportions is immaterial. When drawings are not to scale, you cannot make ANY inferences whatsoever regarding the relative proportions. For example, in a semiconductor device, one layer might be 100x larger than the next layer but I'll draw them the same size for clarity purposes. As another example, the width of the gate may be 50X the width of the vias, but I'll draw them the same size to fit them on a piece of paper.

    the drawings are drawn with a sufficient detail and precision to allow some findings and conclusions with regard to the relative size of the depicted structures
    Utter and complete BS.

    For example, look at Fig. 11, which is very similar to Fig. 15 relied by the Examiner. Layer 464 is disclosed as "having a thickness of from about 0.5 μm to about 1.5 μm." Layer 460 is disclosed as "having a thickness of from about 5 μm to about 20 μm" (about 10x the thickness of 464, while in illustration 460 is only 40% larger). Layer 470 is described as having "a thickness of about 20 nm." Looking at the drawings, 470 is about 1/12th the size of 460, but by the written description, that difference should be 1:250 the size (at the minimum) and could be as much as 1:1000. Both 466 and 472 are described as having the same thickness (i.e., "about 0.1 μm") yet there is a substantial difference between the illustrated thickness. As such, the drawings are EXTREMELY not to scale. Moreover, while there appears to be some attempt to show some relative scale between the thicknesses, these are grossly inaccurate and not always correct.

    Moreover, even if one could gauge some relative scale between the thickness, there is no evidence that same relative scale also applies to the widths.

    All this being said, the reference does explicitly teach (within the written description) that the width of the cavity exceeds the width of the p-contacts. Its amazing (well, maybe not) that nobody caught this.

    Right result -- wrong analysis.

    ReplyDelete
  2. You must be thinking of mraz as mraz talks about a canadian patent "wilson" as the primary ref. and mraz stands for drawings can anticipate but must show POSITA all claimed features and how put together. long time no see. I've been using anki to memorize case names works for me... (edited marz to mraz)

    ReplyDelete
  3. This is easily one of the worst appeal and reply briefs I've ever seen. Complete garbage. All that bold, italicized underlining. Distracting, unnecessary and amateurish. Just plain horrible. What on earth are they trying to accomplish with that? It's like, "Oooh, look, we're really serious that these arguments are powerful and persuasive because we've bolded, underlined, AND italicized these words here!!! Can I haz reversal now?" What an eyesore these POS briefs are. Here's a tip for these hacks: when every other word in your argument is bolded, italicized, and/or underlined, it's not more persuasive or powerful, it's just distracting, annoying, and childish. Stop it. Sheesh.

    In the paragraph spanning pages 5-6 of the decision, the Board noted that the examiner cited (on pages 17-18 of the Answer) col. 10, lines 46-50 and 62-65, of Romano et al. (U.S. 6,744,072) for "a suggestion within Romano" to have the width of the recessed section larger than the width of the projection section to maximize heat dissipation. There is absolutely no suggestion whatsoever in those passages of Romano et al. for that "suggestion." None. Whatsoever.

    I didn't see anywhere in either the appeal brief or the reply brief where the appellant very simply argued that the evidence supplied by the examiner (i.e. col. 10, lines 46-50 and 62-65 of Romano et al.) did not resolve the factual question by a preponderance. Instead the briefs are all a bombast of bold, italicized and underlined crapola. Lots of case law cites that are pointless.

    Here's another tip for these hacks: you have to actually address the evidence the examiner supplies to resolve the factual questions. Address it head on. If it's insufficient to meet the examiner's burden of a preponderance, explain why. Clearly, simply, and directly. Without all the bold, italicized, underlined garbage.

    ReplyDelete
    Replies
    1. > you have to actually address the evidence the examiner supplies
      >to resolve the factual questions.

      Absolutely.

      I focused on the decision's discussion about inferences from the drawings, and that is what initially piqued my interest.

      But stepping back a bit, Yeah, I see that this case is indeed a prime example of not addressing the Examiner's specific findings (contentions). Which is a very common Applicant mistake.

      Delete
    2. These guys completely missed the boat. It's all well and good to argue the case law on what the drawings can be fairly relied upon to disclose or suggest to one of ordinary skill, but once you've done that you have to take the next step and explain why the examiner's reliance on and conclusions from the drawings do not meet the examiner's burden.

      This case is also a great example of the Board's policy of not helping appellant out at all. The portion of Romano et al. cited by the examiner was total horsesh!t for the "suggestion" the examiner cited it for. But appellant never addressed it. It's not even clear if the Board even read the column and line numbers of the reference. The Board just said, "The examiner cited col. 10, lines 46-50 and 62-65 of Romano et al. for that suggestion" and left it at that. The APJ's were not gonna make the argument for appellant. So appellant had to make it for themselves. Something these hacks failed miserably at.

      However, just to be fair, it's entirely possible that they did address the portion of Romano et al. cited by the examiner, but if they did it would have been buried in the midst of all that truly awful bold italicized underlining. If they did, I didn't see it. And it's a darn good bet the Board didn't see it amidst all that cr#p.

      Delete
    3. If I was the client, and those guys sent me that brief to review, I wouldn't even look at it beyond my initial disgust for how just plain awful it looks. I would have sent it back with instructions to get rid of all that bold italicized underlined cr#p and send it back to me when it's actually readable.

      Delete
    4. Here's a tip for these hacks: when every other word in your argument is bolded, italicized, and/or underlined, it's not more persuasive or powerful, it's just distracting, annoying, and childish. Stop it. Sheesh
      I have some Examiners who do exactly the same thing. They reproduce the claim language and half it is either italicized, bolded, or underlined with no explanation as to what means what.

      Delete
  4. explain why. Clearly, simply, and directly.

    Patent attorneys don't seem to write this way. Why not? Don't they get training on this?

    ReplyDelete
    Replies
    1. Patent attorneys don't seem to write this way. Why not? Don't they get training on this?
      Few people do this -- this isn't limited to patent attorneys. Writing clearly is a learned art.

      Delete
  5. "Patent attorneys don't seem to write this way. Why not? Don't they get training on this?"

    Too many think that there's some requirement, in response to a rejection, to "put on our case of patentability" so to speak. So they concoct this "story" about why the claims are patentable. Instead of doing what is required by Rule 111 and explaining why the examiner's case of unpatentability is insufficient. It causes them to ignore what the examiner said in a misguided effort to get out what they want to say. This case is a good example of that type of terrible advocacy.

    It should be, and is, very simple: 1) what is the examiner's burden? 2) what evidence did the examiner provide to meet that burden? 3) why doesn't the evidence provided satisfy the burden?

    That's it. Nothing more required.

    As for training, law students are taught how to write. Some get it, some don't . Within firms, training is mostly informal, and new practitioners pick up the "style" of those they work for. Some "styles" are horrible, for example the style in this case. Some styles, like a lot of what goes on at the PTO, are passed down from generation to generation and just accepted as "good" or "proper." It's an imperfect model, but one that's been used for a very long time.

    ReplyDelete
  6. BTW, these guys cite Sponnoble on page 7 of their reply brief. If there's one case that's in the ICU on life support after KSR, it's Sponnoble. These guys should be embarrassed.

    ReplyDelete
    Replies
    1. If there's one case that's in the ICU on life support after KSR, it's Sponnoble.
      Why do you say that?

      Delete
    2. Sponnoble says, "A patentable invention, within the ambit of 35 U.S.C. § 103, may result even if the inventor has, in effect, merely combined features, old in the art, for their known purpose, without producing anything beyond the results inherent in their use." That can't be argued with a straight face in view of KSR.

      There are aspects of Sponnoble that are still good law, e.g. the discovery of a cause of a problem may itself indicate non-obviousness, but the portion of the case quoted above is deader than a door nail now.

      Delete
    3. That can't be argued with a straight face in view of KSR.
      Sure it can. The quoted language from Sponnoble is "patentable invention ... may result ..." (emphasis added). Compare that to KSR's statement that "The combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable result" (emphasis). In both Sponnoble and KSR, the Courts indicated that such a combination can be obvious but isn't necessarily obvious. These are consistent with the statement from KSR that "As is clear from cases such as Adams, a patent composed of several elements is not proved obvious merely by demonstrating that each of its elements was, independently, known in the prior art."

      Personally, I don't see the inconsistency.

      Delete
    4. The statement from the CCPA in Sponnoble that a combination of known features may be non-obvious even "without producing anything beyond the results inherent in their use" is no longer good law in view of KSR. Citing that passage from Sponnoble, which these clowns did, to a panel of APJ's is borderline unethical, IMO. It's also just plain stupid. As Karen noted in her post, and I noted in my comments, the appellant had, at least with respect to claim 13 and the Romano et al. reference, a good argument that the examiner had failed to establish by a preponderance that all of the claimed features were within the scope and content of the prior art and/or that there was a reason to combine the references in the manner claimed. Citing some case on obviousness that's 38 years before the Supreme Court's decision in KSR most likely gave the APJ's the impression that the appellants had no good arguments. They did have good arguments. They just didn't make them. And the APJ's were not going to make them for the appellants.

      Delete
    5. The statement from the CCPA in Sponnoble that a combination of known features may be non-obvious even "without producing anything beyond the results inherent in their use" is no longer good law in view of KSR
      I disagree. The quote I reproduced from KSR said "likely" -- not necessarily.

      Both Sponnoble and KSR agree that you cannot prove obviousness just by alleging that all the elements were known in the art -- even if the result of the combinations were predictable.

      If you think these cases are inconsistent, please explain further.

      Delete
    6. I did. Sponnoble would allow a finding of non-obviousness of the combination even "without producing anything beyond the results inherent in their use." KSR is pretty clear that a "combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable results," then discussing U.S. v. Adams, Anderson's-Black Rock, and Sakraida. I don't recall seeing any arguments from the hacks that filed this appeal that there was teaching away, or a lack of reasonable expectation of success, or that the combination was beyond the level of ordinary skill. And again, if they made those arguments, then I apologize to them, but in my defense it's hard for me to recall anything from their brief(s) without the image of all that ridiculous bold, italicized underlining clouding my memory.

      Delete
    7. You still haven't said why they are inconsistent. Based upon the same fact pattern (i.e., combination of familiar elements according to known methods yielding no more than predictable result), Sponnoble says a patentable invention may result, and KSR says that it is likely obvious. Neither decision states one result or the other must necessarily occur.

      I don't recall seeing any arguments from the hacks that filed this appeal that there was teaching away, or a lack of reasonable expectation of success, or that the combination was beyond the level of ordinary skill
      Not necessarily -- the Examiner still has to present a reasoning with some rational underpinning to support the Examiner's finding of obviousness. Merely identifying the elements isn't enough.

      Also, Sponnoble stands for "a patentable invention may lie in the discovery of the source of a problem even though the remedy may be obvious once the source of the problem is identified." This part is still cited in the MPEP.

      Delete
    8. Looks to me like you guys are arguing at cross-purposes here.

      *Of course* new KSR and old Spoonoble both say *maybe* -- we all know there is no per se obviousness. But we also know there is a PF case of obviousness, and KSR gives several rationales which suffice as a PF case. That's exactly what KSR brings to the table -- a shift away from TSM and a focus on other rationales.

      >Both Sponnoble and KSR agree that you cannot PROVE
      >obviousness just by alleging that all elements were known in the art --
      >even if resulting combination was predictable.

      (capitals emphasis added.)

      OK, sure, the Examiner can't PROVE obviousness by making findings of (a) combination of known elements (b) leading to a predictable result.

      The above does, however, constitute a PF case. And in prosecution, that PF case is usually the real battleground.

      Delete
    9. FWIW, the Applicant here cited Spoonable for its use of the word "inherent" -- to accompany its argument that the dimensional relationship wasn't inherent in the reference. At no point did the Applicant make a 103 (i.e, Rationale-to-Combine) argument.

      The Examiner's proferred Rationale-to-Combine was a TSM from Applicant's own Background. I didn't study the TSM enough to say whether it should have been challenged.

      Delete
    10. That's exactly what KSR brings to the table -- a shift away from TSM and a focus on other rationales
      No ... not really. The other "rationales" could easily be read as examples of teachings, suggestions, or motivations. Also, many/(most?) examiners still use a form of TSM and I have seen very few that strictly apply one of the KSR rationales described in MPEP 2143.


      No again. I suggest you read the requirements again (see MPEP 2143). Merely alleging that the elements were known and lead to a predictable result is NOT enough. In the mechanical and electrical arts, nearly all the results are "predictable." As such, so long as the elements are known -- and nearly all are known, then nearly everything in the mechanical and electrical arts would be obvious.

      Delete
    11. >other "rationales" could easily be read as examples of
      >teachings, suggestions, or motivations.

      So, what, you're saying that substitution is an example of a rationale for combining? Well, sure. That's why I've referred to substitution as a "generic" or "other" rationale. Guess we agree on this point.

      So maybe you and I are simply using the terminology differently. When I say "TSM" I mean an explicit teaching in the reference itself used as a rationale for the combination. When I say "generic rationale" (or "other rationale" in this comment thread), I mean something other than TSM.

      >many/(most?) examiners still use a form of TSM

      If "TSM" means statements from the references, then Yes, that's my experience too.

      Even so, Examiners can and do use the non-TSM rationales from KSR. Unfortunately, as you note, plenty of them do *not* apply these rationales properly. Instead of making a showing that element X in reference is a well-known equivalent to element Y in claim, the Examiner simply says "obvious to substitute Y for X." Without any predicate findings of fact.

      >Merely alleging that the elements were known and lead to a
      >predictable result is NOT enough

      OK, I'll tighten my sloppy terminology.

      It's not enough to *allege*.

      According to MPEP 2143(B), what *is* enough for a Prima Facie case is making these 4 predicate findings of fact: (a) only difference is the combination; (b-1) using known method to combine; (b-2) each element merely performs same function; (c) results are predictable.

      Delete
  7. Ok Ok I found In re Wilson 136 USPQ 188 looking at the appellant's brief. Nice old J. Rich case, not in MPEP and not in google scholar.

    ReplyDelete