Ex Parte DUTTA et al

Patent Trial and Appeal Board

Dec 26, 2018

12814756 (P.T.A.B. Dec. 26, 2018)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
12/814,756 06/14/2010
74739 7590 12/28/2018
Potomac Law Group, PLLC (Oracle International)
8229 Boone Boulevard
Suite 430
Vienna, VA 22182
Debasis DUTTA
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www .uspto.gov
ATTORNEY DOCKET NO. CONFIRMATION NO.
2011-0234US01 1043
EXAMINER
FLEISCHER, MARK A
ART UNIT PAPER NUMBER
3629
NOTIFICATION DATE DELIVERY MODE
12/28/2018 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
patents@potomaclaw.com
bgoldsmith@potomaclaw.com
eofficeaction@appcoll.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte DEBASIS DUTTA, BRIAN GASP AR,
JULIAN CHALLENGER, and DINESH ARORA
Appeal2017-009329
Application 12/814,756 1
Technology Center 3600
Before JASON V. MORGAN, IRVINE. BRANCH, and
MICHAEL M. BARRY, Administrative Patent Judges.
MORGAN, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Introduction
This is an appeal under 35 U.S.C. § 134(a) from the Examiner's final
rejection of claims 1, 3-8, 10, 11, and 13-20. Claims 2, 9, and 12 are
canceled. Appeal Br. 18-20. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.
1 Appellants identify Oracle International Corporation as the real party in
interest. Appeal Br. 2.
Appeal2017-009329
Application 12/814,756
Invention
Appellants disclose using a data mining tool to predict "the future
performance of [an] employee and/ or the likelihood that the worker leaves at
some point in the future." Abstract.
Representative Claim (key limitations emphasized)
1. A non-transitory computer-readable medium having instructions
stored thereon that, when executed by a processor, cause the processor
to foresee employee characteristics, the foreseeing comprising:
collecting attributes associated with each of a plurality of workers,
wherein the attributes relate to at least one of backgrounds,
responsibilities, or compensations of the workers;
applying a regression technique to the attributes to find a first function
of the attributes that outputs a performance rating of at least one of the
workers;
using the first function to predict a future performance of a specific
worker;
applying a classification technique to the attributes to find a second
function of the attributes that outputs a voluntary termination of at
least one of the workers; and
using the second function to predict a future likelihood of attrition of
the specific worker.
Rejections
The Examiner rejects claims 1, 3-8, 10, 11, and 13-20 under
35 U.S.C. § 101 as being directed to patent-ineligible subject matter. Final
Act. 4---6.
The Examiner rejects claims 1, 4---6, 8, 11, 14--16, 18, and 19 under
35 U.S.C. § I03(a) as being unpatentable over April et al. (US
2011/0015958 Al; published Jan. 20, 2011) ("April"), Sureka (US
2008/0077544 Al; published Mar. 27, 2008), Menon (US 2009/0307025
2
Appeal2017-009329
Application 12/814,756
Al; published Dec. 10, 2009), and Scarborough et al. (US 7,472,097 Bl;
issued Dec. 30, 2008) ("Scarborough"). Final Act. 6-12.
The Examiner rejects claims 3, 7, 10, 13, 17, and 20 under
35 U.S.C. § I03(a) as being unpatentable over April, Sureka, Menon,
Scarborough, and Y oungjo Lee et al., Generalized Linear Models with
Random Effects: Unified Analysis via H-likelihood, CRC Press (2006)
("Lee"). Final Act. 12-14.
35 U.S.C. § 101 REJECTION
Examiner's Determinations and Conclusions
In rejecting claim 1 as being directed to patent-ineligible subject
matter, the Examiner determines the claimed invention is directed to
"predicting employee characteristics," which is "a species of fundamental
economic practices, methods of organizing and assessing and predicting
human activities and performance," and, thus, directed to an abstract idea.
Final Act. 4. The Examiner also determines the claim recitations "all entail
organizing information through mathematical correlations [and] hence are
directed to further abstract ideas." Id. at 5. The Examiner determines claim
1 does not include "additional elements that are sufficient to amount to
significantly more than the judicial exception because the additional
limitations ... are insignificant data gathering steps such as insignificant
pre[-] and post-solution activities." Id.
Principles of Law
To be statutorily patentable, the subject matter of an invention must be
a "new and useful process, machine, manufacture, or composition of matter,
or [a] new and useful improvement thereof." 35 U.S.C. § 101. There are
implicit exceptions to the categories of patentable subject matter identified
3
Appeal2017-009329
Application 12/814,756
in§ 101, including: (1) laws of nature; (2) natural phenomena; and (3)
abstract ideas. Alice Corp. v. CLS Bank Int 'l, 573 U.S. 208, 217 (2014).
The Supreme Court has set forth a framework for distinguishing patents with
claims directed to these implicit exceptions "from those that claim patent-
eligible applications of those concepts." Id. (citing Mayo Collaborative
Servs. v. Prometheus Labs., Inc., 566 U.S. 66 (2012)). The evaluation
follows a two-part analysis: (1) determine whether the claim is directed to a
patent-ineligible concept, e.g., an abstract idea; and (2) if so, then determine
whether any element, or combination of elements, in the claim is sufficient
to ensure that the claim amounts to significantly more than the patent-
ineligible concept itself. See Alice, 573 U.S. at 217-18.
"[ A ]11 inventions at some level embody, use, reflect, rest upon, or
apply laws of nature, natural phenomena, or abstract ideas." Mayo, 566 U.S.
at 71. We "'must be careful to avoid oversimplifying the claims' by looking
at them generally and failing to account for the specific requirements of the
claims." McRO, Inc. v. Bandai Namco Games Am. Inc., 837 F.3d 1299,
1313 (Fed. Cir. 2016) (quoting TL! Commc'ns LLC Patent Litig., 823 F.3d
607, 611 (Fed. Cir. 2016)). We, therefore, look to whether: (1) the claims
focus on a specific means or method that improves the relevant technology
or (2) the claims are instead directed to a result or effect that itself is the
abstract idea and merely invoke generic processes and machinery. See
Enfzsh, LLC v. Microsoft Corp., 822 F.3d 1327, 1335-36 (Fed. Cir. 2016).
That is, we look to whether the claims are "specifically designed to achieve
an improved technological result in conventional industry practice." McRO,
837 F.3d at 1316 (emphasis added) (citing Alice, 573 U.S. at 222-24).
4
Appeal2017-009329
Application 12/814,756
If a claim proves to be unpatentable as a result of the two-part
analysis, no additional determination regarding preemption is necessary.
"While preemption may signal patent ineligible subject matter, the absence
of complete preemption does not demonstrate patent eligibility," as
"questions on preemption are inherent in and resolved by the § 101
analysis." Ariosa Diagnostics, Inc. v. Sequenom, Inc., 788 F.3d 1371, 1379
(Fed. Cir. 2015) (internal quotation marks and citation omitted).
Appellants' Contentions and Our Analysis (Step 1)
Appellants contend the Examiner erred by failing "to provide a court
decision or provide a proper identification of an abstract idea." Appeal Br.
5; see also Reply Br. 2. The Examiner, however, properly determines that
the claim recitations are directed to fundamental economic practices related
to "predicting employee characteristics." Final Act. 4.
Specifically, the claimed invention's application of regression and
classification techniques to predict, based on collected attributes, future
worker performance and attrition merely juxtaposes methods of calculating
that, individually and in combination, are abstract. See SAP Am., Inc. v.
Versata Dev. Grp., Inc., 107 USPQ2d 1097, 1110 (PTAB 2013) (non-
precedential), aff'd, 793 F.3d 1306 (Fed. Cir. 2015). The ideas of seeking to
predict worker performance and seeking to predict worker attrition are forms
of risk management that, like methods of hedging against price fluctuation
risk or mitigating settlement risk, are directed to abstract fundamental
economic practices. See Alice, 573 U.S. at 219-21. The Specification
acknowledges that "[ o ]rganizations utilize human resource management
processes to attract appropriately skilled employees, integrate them into the
organization, assess and develop their competencies, and retain their
5
Appeal2017-009329
Application 12/814,756
commitment." Spec. ,r 2. The claimed invention, which uses mathematical
techniques as part of these processes (e.g., assessing an employee's
competencies and commitment) merely adds the use of mathematical
concepts without making the underlying abstract idea-the fundamental
economic practice of managing risks associated with the use of human
resources by predicting employee performance and attrition-patent eligible.
See RecogniCorp, LLC v. Nintendo Co., 855 F.3d 1322, 1328 (Fed. Cir.
2017), cert. denied, 138 S. Ct. 672 (2018) ("A claim directed to an abstract
idea does not automatically become eligible merely by adding a
mathematical formula").
Furthermore, "[u]sing a certain technique (i.e., classification) for
predicting attrition and a completely different technique (i.e., regression) for
predicting performance" (Appeal Br. 6) merely adds two abstract concepts
together (i.e., two fundamental economic risk management practices) in a
manner that fails to render claim 1 non-abstract (see RecogniCorp, 855 F .3d
at 1327). As the Examiner correctly notes, "merely using two basic and
different, but well-known techniques, for analyzing data for two distinct
purposes, one for predicting performance and one for predicting class
membership, does not establish unconventionality." Ans. 7. The overall
claim continues to be directed to managing risks associated with the use of
human resources by predicting employee performance and attrition-a
fundamental economic practice.
For these reasons, we agree with the Examiner that claim 1 is directed
to an abstract idea. Final Act. 4.
6
Appeal2017-009329
Application 12/814,756
Appellants' Contentions and Our Analysis (Step 2)
Appellants contend the Examiner erred because
use of a computer to apply a classification technique for
predicting attrition and a regression technique for predicting
performance is clearly a technical solution to the tedious problem
of determining employee characteristics based on attributes
associated with each of a plurality of workers, and is not a
solution that can be easily performed manually or using routine
computer data storage and mathematical operations.
Appeal Br. 6-7. That is, Appellants argue the invention of claim 1
"improve[s] the functioning of [a] computer." Id. at 9.
Appellants' arguments are unpersuasive because "mere automation of
manual processes using generic computers does not constitute a patentable
improvement in computer technology." Credit Acceptance Corp. v.
Westlake Servs., 859 F.3d 1044, 1055 (Fed. Cir. 2017). Here, the
Specification generically describes a number of computer technologies that
could perform the claimed techniques without detailing how such computer
technologies are improved. See, e.g., Spec. ,r,r 27 ("Processor 22 may be any
type of general or specific purpose processor"), 37 ("functionality may be
performed by hardware ... or any combination of hardware and software"),
73-77, Fig. 1.
Furthermore, contrary to Appellants' arguments that the claimed
solution cannot be performed "using routine computer data storage and
mathematical operations" (Appeal Br. 7), the Specification describes both
the storage and mathematics used broadly as routine, well-known
technologies and techniques. For example, the Specification discloses that
"[ c ]omputer readable media may be any available media that can be
accessed by processor 22 and includes both volatile and nonvolatile media,
7
Appeal2017-009329
Application 12/814,756
removable and non-removable media, and communication media." Spec.
,r 28. The Specification also discloses multiple classification and regression
techniques, denoting some simply by name. See, e.g., id. ,r,r 41
("classification algorithms include Naive Bayes, Decision Tree, ... "), 42
("regression algorithms include Multiple Regression ... ").
For these reasons, we agree with the Examiner that claim 1 is
"specified at a very high level of generality and invoke[ s] well understood
techniques for the statistical analysis of data." Ans. 7. Moreover, the
invention of claim 1 merely recites, in the preamble, use of generic computer
elements as a tool. See Credit Acceptance, 859 F .3d at 1055 ( citing Enfzsh,
822 F.3d at 1335-36). Therefore, we agree with the Examiner that claim 1
does not have additional limitations sufficient to transform the underlying
abstract idea into patent-eligible subject matter. Final Act. 5.
Accordingly, we sustain the Examiner's 35 U.S.C. § 101 rejection of
claim 1, and claims 4--8, 11, and 14--20, which Appellants do not argue
separately with respect to this rejection.
Dependent Claims 3, 10, and 13
With respect to claim 3, which depends from claim 1, the Examiner
determines that "outputting information in a graphical X-Y chart ...
constitute[ s] ... insignificant post-solution activity associated with data
gathering." Final Act. 5. Appellants contend the Examiner erred because
"the recited graphical representation provides benefits that are significantly
more than an abstract idea." Appeal Br. 8. We agree with the Examiner,
however, that "the use of such graphical outputs, especially in the context of
statistical methods, is a well understood and commonplace practice such as
use of scatter plots and the like." Ans. 9. The Examiner correctly concludes
8
Appeal2017-009329
Application 12/814,756
that "[ s ]uch typical and usual types of displays are not generally seen as
constituting significantly more than abstract ideas." Id. (citing Elec. Power
Grp., LLC v. Alstom S.A., 830 F.3d 1350, 1352 (Fed. Cir. 2016)). The
Specification's illustrations of the claimed XY chart (e.g., Spec. Fig. 13a)
show that the claimed XY chart is merely a scatterplot-a basic visualization
of two-dimensional data points that has its origins in the first half of the
nineteenth century (see Michael Friendly and Daniel Denis, The Early
Origins and Development of the Scatterplot, Journal of the History of the
Behavioral Sciences, Vol. 41(2), 103-30, 128 (Spring 2005)). The use of
this well-known visualization technique does not make the invention of
claim 3 significantly more than the underlying abstract idea of managing
risks associated with the use of human resources by predicting employee
performance and attrition-an abstract fundamental economic practice.
Accordingly, we sustain the Examiner's 35 U.S.C. § 101 rejection of
claim 3, and claims 10 and 13, which Appellants do not argue separately
with respect to this rejection.
35 U.S.C. § 103(A) REJECTIONS
Examiner's Determinations and Conclusions
Menon discloses examples of general behaviors, such as consistent
low performance on metrics, that form general employee satisfaction
indicator set 206. Menon ,r 46, Fig. 2. Menon teaches that an "employer
may use a knowledge base, database, or other information to evaluate the
employee satisfaction behavior." Id. ,r 52. "The expected attrition time
measurement for [an] employee may be calculated based on ... the
employee satisfaction behavior." Id. ,r 60.
9
Appeal2017-009329
Application 12/814,756
In rejecting claim 1 the Examiner relies on Menon's employee
satisfaction behavior teachings, which include performance metrics, to teach
or suggests using a first function to predict a future peiformance of a
specific worker. Final Act. 8 (citing, e.g., Menon Fig. 9, ,r,r 46-50, 61); see
also Ans. 11 ("the rejection clearly attributes prediction of performance to
Menon"). The Examiner concludes it would have been obvious to modify
the teachings and suggestions of the other cited references "to include
worker performance predictions as taught by Menon" because "the claimed
invention is merely a combination of old elements." Final Act. 8.
Appellants ' Contentions and Our Analysis
Appellants contend the Examiner erred because "Menon merely lists
'examples of general behaviors' which may include 'consistent low
performance.' Such behaviors are used as input information in various
embodiments of Menon to predict attrition, but Menon is completely silent
regarding any performance related output such as a prediction of
performance." Appeal Br. 14.
Appellants accurately characterize the use of "performance" as an
input predictor rather than as a prediction output. Menon teaches that an
employee with "consistent low performance" exhibits of the negative
employee satisfaction indicators that places the employee at higher risk of
attrition, which can be reported as expected attrition date. See Menon
,r,r 46-53, 60. Importantly, the Menon does not predict employee
performance. Rather, Menon uses employee performance to predict
attrition. The Examiner does not present findings or explanation showing
that it would have been obvious to predict employee performance based on
10
Appeal2017-009329
Application 12/814,756
the cited teachings of Menon. Nor does the Examiner show that the other
references cure the noted deficiency of Menon.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 103(a)
rejection of claim 1, and the Examiner's 35 U.S.C. § 103(a) rejections of
claims 3-8, 10, 11, and 13-20, which contain similar recitations
DECISION
Because we affirm at least one rejection of each claim, we affirm the
Examiner's decision rejecting claims 1, 3-8, 10, 11, and 13-20.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 41.50(±).
AFFIRMED
11
APPENDIX
Excerpts from Michael Friendly and Daniel Denis, The Early Origins
and Development of the Scatterplot, Journal of the History of the Behavioral
Sciences, Vol. 41(2), 103-30 (Spring 2005)
Application/Control No. Applicant(s)/Patent Under Patent
Appeal No.
Notice of References Cited
12/814,756 2017-009329
Examiner Art Unit
3629 Page 1 of 1
U.S. PATENT DOCUMENTS
*
Document Number Date
Country Code-Number-Kind Code MM-YYYY Name Classification
A US-
B US-
C US-
D US-
E US-
F US-
G US-
H US-
I US-
J US-
K US-
L US-
M US-
FOREIGN PATENT DOCUMENTS
*
Document Number Date
Country Code-Number-Kind Code MM-YYYY Country Name Classification
N
0
p
Q
R
s
T
NON-PATENT DOCUMENTS
* Include as applicable: Author, Title Date, Publisher, Edition or Volume, Pertinent Pages)
u The Early Origins and Development of the Scatterplot, Journal of the History of the Behavioral Sciences, Vol. 41(2), 103-30
Spring 2005
V
w
X
*A copy of this reference is not being furnished with this Office action. (See MPEP § 707.05(a).)
Dates in MM-YYYY format are publication dates. Classifications may be US or foreign.
U.S. Patent and Trademark Office
PT0-892 (Rev. 01-2001) Notice of References Cited Part of Paper No.
Jownal of the History ~{the Behavioral Sciences, Vol. 41(2), 103-130 Spring 2005
Published online :in Wiley Interscience (v,,-wv1.interscience.wiley.com). DOI 10.1002 ;Jhbs.20078
© 2005 Wiley Periodicals, Inc.
THE EARLY ORIGINS AND DEVELOPMENT OF THE SCATTERPLOT
MICHAEL FRIENDLY AND DA:"IIEL DENIS
Of all the graphic forms used today, the scatterplot is arguably the most versatile, polymor-
phic, and generally useful invention in the history of statistical graphics. Its use by Galton led
to the discovery of correlation and regression, and ultimately to much of present multivari-
ate statistics. So, it is perhaps surprising that there is no one widely credited with the inven-
tion ofthis idea. Even more surprising is that there are few contenders for this title, and this
question seems not to have been raised before.
This a.iticle traces some of the developments in the history of this graphical method, the ori-
gin of the term scatterplot, the role it has played in the history of science, and some of its
modern descendants. We suggest that the origin of this method can be traced to its unique ad-
vantage: the possibility to discover regularity in empirical data by smoothing and other
graphic annotations to enhance visual perception. © 2005 Wiley Periodicals, Inc.
Of all the well-known graphical devices used today for the display of quantitative data,
the most ubiquitous, at least in popular presentation graphics-pie charts, line graphs, and bar
charts-in their modern form, are generally attributed to William Playfair 1 (1759-1823). All
of these were essentially one-dimensional (ID) or at least topographically equivalent to ID
forms (or perhaps 1.5D, where the .5 was a dimension of time, a categorical variable, or an-
other variable, showing an apparently functional or exact relation).
The next major invention, and the first true two-din1ensional (2D) one, is the scatte,plot.
Indeed, among all the forms of statistical graphics, the humble scatterplot may be considered the
most versatile, polymorphic, and generally useful invention in the entire history of statistical
graphics. Tufte (1983) estimated that between 70 and 80 percent of graphs used in scientific
publications are scatterplots; see also Cleveland and McGill (1984) for modern developments.
To distinguish it from other graphic forms, we define a scatterplot as a plot of two vari-
ables, x and y, measured independently to produce bivariate pairs (xi, y;), and displayed as in-
dividual points on a coordinate grid typically defined by horizontal and vertical axes, where
there is no necessary functional relation between x and y. The definition for the tenns scatter
diagram, plot, from the Oxford English Dictionary ( OED2) gives the three essential charac-
teristics: (a) two variables measured on the same observational units, (b) plotted using points
referred to (typically orthogonal) axes, and (c) designed to show the relation between these
variables (typically as how the ordinate variable, y, varies with the abscissa variable, x):
scatter diagram, plot (Statistics), a diagram having two variates plotted along its two
axes and in which points are placed to show the values of these variates for each of a
nmnber of subjects, so that the form of the association between the variates can be seen.
1. Line graphs of time series go back at least to Robert Plot (1685), and to Christopher Wren's 1750 "weather clock,"
a mechanical device for automatically recording a graph of temperature. The idea of a bar chart, to represent a math-
ematical function, rather than data, goes back to Nicholas Oresme (1482). But Playfair's 1786 Commercial and
Political Atlas and his subsequent Statistical Breviary (Playfair, 1801) are generally acknowledged as the first mod-
ern use and explanation of these graphic forms to portray empirical (largely economic) data.
l'vfICHAEL FRIENDLY is a professor in the Psychology Department, York University, Toronto, ON,
Canada, E-mail: fi'iendl)@yorku.ca.
DANIEL DENIS was a graduate student in psychology at York University, now at the University of}l1ontana.
E-mail: daniel.denis@umontana.edu
103
128 MICHAEL FRIENDLY AND DANIEL DENIS
Concurrently (i.e., in the 1970s), researchers began to develop software to allow users
to interact directly with a collection of several linked displays of the same data by selecting
a subset of observations or a rectangular region in a display or table (with a mouse) so that
those selected observations were identified and highlighted in all other (linked) views of
the same data. More recent extensions and combinations of these ideas, and widely avail-
able software implementations too numerous to cite here, have led to explosive growth of
interactive data-visualization methods, many of which exploit the polymorphic adaptability
of the basic scatterplot.
SrnvIMARY AND CONCLUSIONS
In the first half of the nineteenth century, all the modern forms of data display were in-
vented (Friendly & Denis, 2001 ), but the scatterplot-a bivariate plot of points designed to
show the relation between two separate quantitative variables-was a relatively late arrival,
perhaps the last to be introduced. We have argued that the idea of such a plot would not have
occurred to Playfair, who viewed the largely economic data with which he was concerned as
comparisons of distinct series of measm·ements across time or circumstance ( country, prod-
uct, imports vs. exports, etc.), but in Playfair's time such comparisons were conceived indi-
rectly, by overlaying several graphs on the same page.
The need for a scatterplot arose only when scientists had the need to examine bivariate
relations between distinct variables directly. As opposed to other graphic forms-pie charts,
line graphs, and bar charts-the scatterplot offered a unique advantage: the possibility to
discover regularity in empirical data (shown as points) by adding smoothed lines or curves
designed to pass "not through, but among them," so as to pass from raw data to a theory-
based description, analysis, and understanding. Indeed, in the cases of both Herschel and
Galton, it may be argued that the plots of the smoothed relations were primary, in both pres-
entation and use.
The term scatter diagram did not enter the statistical lexicon m1til the early 1900s, when
it was illustrated as the method to "see the relationship" between variables. In this same pe-
riod, the scatterplot proved its worth in several remarkable discoveries.
Modern enhancements of the basic scatterplot include a variety of methods designed to
show more than a 2D view, to deal effectively with more complex data, and to allow the user
to work interactively with collections of bivariate displays.
From the material we have presented, it appears that credit for the origin of the scatter-
plot in modern form belongs to J. F. W Herschel, though of course there were earlier roots in
graphing functions and in mapmaking, as we described in our opening section. As far as we
are aware, Herschel's 1833 paper was the first to both describe the construction of a basic
scatterplot and suggest its use for smoothing empirical bivariate data. (J. H. Lambert [1760]
had earlier described curve fitting and interpolation for empirical data, in the context where
a functional relation was expected.) More generally, we have also attempted to provide a brief
history of statistical and graphical ideas that led to the modern use of the scatterplot.
ACKNOWLEDGMENTS
This work is supported by Grant 8150 from the National Sciences and Engineering Research Council of Canada. Vle
are grateful to Paul Delaney for insightful discussion on astronomical data, and to !es Chevaliers: Antoine de
Falguerolles, Ian Spence, Antony Unwin, and Howard Wainer for helpful comments on earlier drafts. Two anonymous
reviewers and the editor helped us sharpen our presentation and arguments. A copy of this article with color figurns
may be found on the first author's Web site at http:/ /vvvvw.math.yorku.ca/SCS/Papers/scat.pdf.