Here are the few specimen entries for the Android Makers Encyclopedia "F" topics:

The following is a WordPerfect "Subdoc" from the printed editions...

FACE MEASUREMENTS

This topic is the instructions for taking the various measurements of the face. Making these correctly help assure your android's likeness as desired.

Now that the points used by anthropologists, artists, forensic detectives, cosmetic surgeons, and now android (and doll) makers are documented, a step or sub-step of the android making process can be described. This part is to actually make the measurements. These measurements are the measured distances between two anthropometric points.

These measurements a re done in one of two ways. One is the straight line point-to-point. The other measurement is like the "Great circle-arc" which takes a chordal measurement as the tape is placed on the surface.

     MEASURED ATTRIBUTES: Use the 60 scale of an engineers scale for more resolution when working from a photograph. These measurements use many of the points described in the previous section. The order of these measurements starts with the most basic and then those that are more detailed.

Face height

Upper face height

Skull-base width

Face Width

Orbital face height

Mandible height

Lower face height

Upper chin height

Chin height

Forehead Width

Intercanthal width

Eye fissure width

Biocular width

Orbit height

Lower eyelid height

Upper eyelid height

Eye fissure height

Eyebrow height

Mouth width

Half mouth width

Cutaneous upper lip height

Upper lip height

Lower lip height

Philtrum width

Cutaneous lower lip height

Upper vermillion Height

Ear insertion height

This leaf could be copied as a working paper for your android's measurements.

Admittedly, the previous schema appears a little bit disjoint because not all of the overall measurements or assignments are together in one location while the facial anthropometrics are in another. Unfortunately, computation dependencies are a higher priority in this generation of worksheet. Hopefully, successive revisions and versions can reduce this effect.

The tedious chores in this step is measuring and/or extrapolating the respective measurements from a working copy of the picture or image of your desired android. You will want a working copy because the necessary projections and extrapolations will mark up the portrait. Great care is still required because the prompted cell in the worksheet must assume that the numeric values you enter come from the likeness you desire.

Fortunately, the "seed" values in the worksheet can help you in this process. Their values may suggest a reasonable range for the particular entry. Unfortunately, these numbers may not represent the "specimen provided" (see "STEP ONE: DEFINE YOUR ANDROID" topic) very well, if at all. Therefore, if you measure something radically different, then you may assume that there is an error in measurement (an error in the model is remotely possible, but not very probable).

FACE STUDY

Constructing a Hatfield Type Android begins with a feminine bias like the beginning of a human body. Since the particular artifacts and art work of this encyclopedia are intended to be female where that matter is pertinent, a question arises:

What makes an attractive feminine face? What is masculine attractiveness versus feminine beauty?

"Masculine attractiveness" is almost an oxymoron for The Android Maker, but that is another story. Maleness, related to masculinity, is associated with a body where form seems to follow function more so than a nubile female; form is less important than function. "Smooth and silky" is feminine where "tough and `knarly'" are usually masculine attributes.

Defining a pretty face has been a matter of discussions for centuries. In a way, most agree that "I don't know what it is, but I know it when I see it."

However, some have applied scientific methodologies to this question. Farkas [p.95] found that "women with above average [read "attractive"] faces had a larger biocular width than nose width." This seems to mean that slender noses or wider spaced eyes are more attractive.

Farkas used two pictures; one to illustrate an "attractive" face and the other to depict a "below average" face. The aspect that seemed to provide the most difference is about the mouth level and behind the lips. This line of attention seems to pass between the molars. The attractive face is thinner than the below average face along this line.

There were a few other technique used to make the below average face appear worse. The eyes were whitened out so only the eye fissure was pre sented. Also, the subject's hair was parted to form a bilateral "bang" that can make any face less appealing. Shaping her eyebrows a bit could improve her appearance at the eye level too.

    Reviewing this, I thought of an experiment. I cavitated my cheeks as I watched the effect in a mirror. With coaxing and practice, a better-than-life portrait could be taken of someone who is not shaped the most desirable.

L. M. Boyd has said a few interesting things about this matter in his column. "The perfect oval face is a rarity among women." [15 May 1990] is one example.

Apparently he may have seen some of Farkas' (or his peers) work when his column cited that the averages of facial characteristics; anthropometrics of beauty contestants were evaluated in three dimensions. The "winners were almost exactly average. ... It is averages that's beautiful." [7 Sep. 1990]

The French philosopher Rene Descartes;
    "was particularly attracted to eyes imperfectly aligned. But almost any distinctive characteristic about the eyes tends to heighten romantic appeal, according to our Love and War man's sources. It's said to contribute to the intense interest so many western men show in Far Eastern women." [13 Nov. 1989]

What makes the female eyes more attractive than men or some more attractive than other women? It is not eye color: The man's eyes and the woman's (no makeup apparent) in an available test set appear to be the same. Spacing? The man's and the woman's both measured 160 on the 60 scale of an engineers scale. Eye fissure height-to-width appears a candidate; giving the bigger, rounder eye appearance , but some attractive women have narrow eye fissures. The "one eyebrow" of the man is definitely male. Female skin is generally smoother and sleeker, and their eyebrows are generally neater; i.e.; with a defined edge. A frequent complaint by females about crossdressers is that male eyelashes are more prominent.

    One of my observations is that pretty faces seem to have the least prominent supraorbital ridges. This pair of skull ridges may be softened by extra female and some particular racial tissue thicknesses.

Although some female faces do not meet the anthropometric "average" for standard attractiveness, some girls appear quite "precious." How can this be if their anthropometric ratios are not "average?"

Descartes, as previously cited, suggests that some imperfections are, in fact, desirable. Common or familiar racial or ethnic factors may be a part of this answer.

Examine the specimen (assuming that you have picture of a female) or your desired android in your copy: Is she attractive? If so, is it because she is "average" meaning that her facial measurements and ratios are near the arithmetic mean for her population by race? Is her face soft and sleek and otherwise void of features that could be described as "masculine;" with such affectations like "squarish" or "chiseled?" Or is she pretty because there is some endearing quality to her image?

Perhaps there is a combination of answers. An anthropometrically pretty face may still have the "darling imperfection."

LESS THAN ATTRACTIVE

This subtopic is the appropriate place to insert this mini-dissertation about this aspect of appearance standards. The previous section has been about attractiveness, so this one is for those less than attractive (and for "compare and contrast"). After all, making a "troll" may have certain commercial value.

There are several degrees that are substantially less than attractive. See "Degrees of Ugly."

FEMUR LENGTH

(Engineering) A "mini investigation" was conducted to determine a function of a femur length to overall height ratio. This and the tibia length ratio contribute to the "legginess" of a subject. A synopsis of this bit of research follows:

    Assumptions:
    This measurement can be defined from a function using overall height and a number where that number is a function of certain desires.

    The geometry lines and planes can be determined using the object or other rules.

    Meaningful, valid, and useable functions can be determined.

    Woodson's Skeletal Linkage System chart helped determine the location of the hip and knee articulations under the skin. The point of the hip joint is on a horizontal line with the depression between the curves in the silhouette of the hips , which appear tangent to the lateral surface of the great trochanter. Likewise the knee joints are horizontal with the points where the inner curves of the knee touch tangent with the heels close together , near the extents of the medial and lateral surfaces of the epicondyles. However, this does not illustrate the "Q-angle" [Luttgens, p.215] nor is it related to the bones' surfaces.

[ 50%ile Female Frontal from [Tilley].]
    Woodson's graphic [Woodson, p.708] was only marginally helpful. Tilley, by way of a lost noted web site, shows some more by adding dimensions to the geometry (see ). Still, a diagram with these dimensions, skin contours, and bone surfaces is in order. It is properly part of the documentation for an android or an androtic doll.

    Definitions used within this topic:
    Femur length (FemrLEN): Length of the femur from the distal surface of the medial condyle to the p roximal surface of the head extent. This is like the minimum length of a box to contain the bone.

    Overall (OA) height (OAHeight): The given or desired height of the android (or human).

    Aspect Ratio (AspectR for the scope of this entry): The object of this discussion. It is the FemrLEN/OAHeight. This is normally about a quarter of the individual's height, but can be longer in a long legged person; more specifically, a "leggy" female. This aspect ratio is modified by the appearance modifier index of the MODEL2.WQ1 (or its descendent) worksheet. An attractive female will have a greater ratio than a man or not-so-attractive woman.

    Desired Attributes of Subjects: Juvenile-to-young adult. Female. Caucasoid (May have some oriental, asian, indian, mediterranean, etc. This probably rules out aboriginal, mongoloid, negroid, etc.). Attractive. Small, slim-to-not heavy, perhaps "shapely." "Leggy" women mixed in with samples.

    Findings:
    SAMPLE
Avg. 0.262936
Stds. 0.015437
Count 36
Min. 0.243902
Max. 0.305263
Avg+3Std 0.309247
Avg+2½Std0.301529
Avg+2Std 0.293810
Avg+Std 0.278373

[    Overall height and estimated femur lengths.
]
    Graphic Interpretation:
    Clusters "tri-modal" around .25, .26, and .27. See the graph ( ) for patterns of distribution. This effect could be partly caused by the inherent rounding of the measurement process.

    The data points are sparse at the high end (the right half of the graph) and the lack of points to the left are not a normal distribution. This obviously skews the average high.

    CONCLUSIONS:
    A doll's, model's or android's femur length for this described desired subject may vary from 24.3902% to 30.9247%.

    This should be easy to determine because this is a simple function of a given value. This function being:

    The Appearance Index Modifier and the Base Android Type switches in Model 3.WQ1 help "fine tune" the desired AspectR.

     The Aspect ratio for normal young adult female should be near the mean; .263.

     The Aspect ration for attractive young adult female should be near the mean plus one standard deviation; .278.

     The Aspect ratio for "leggy" and attractive young adult female should be near the mean plus at least two standard deviations; .294.

    It is hard to believe that 3% could make so much apparent difference. Actually, with the lower leg bones (see the tibia discussion) the 6% is quite apparent. Because this same 6% must be complimented with must come from the other side (of the upper torso) too, this effect totals twelve percent (12%)!

    Masculine ratios could be applied equally to unattractive women. In fact, male parts would make a female android that is not attractive. (Why make something ugly??)

(The accompanying graph is not as clean as desirable because the graphic is a Lotus .PIC file. Quattro Pro has an apparent "bug" that causes the HPGL graphic files to finish with a color wipe that hides the image. WordPerfect uses HPGL files from other applications as fine as its native .WPG graphic file format.)

Refer to NPO-19265, volume 19, no. 9 of NASA Tech Briefs page 60. [NTB, Sept. 1995, p.60]

     Ferrofluids (n) are synthetic oils containing submicroscopic magnetic particles of approximately 1015 per drop. These magnetic particles respond to the control of a magnetic force. The response can be defined by formulation.

(A brochure from Ferrofluids Corporation seems to primarily sell motors and dampers. That brochure also contains the published report; by K. Raj and R. Moskowitz for more information.)

75-400 Gauss is the range of ferrofluids for stepper motors. How much is required for an android "muscle" is not determined as of June 8, 1995.

Viscosity ranges from 25 to 10,000 centipoise at 27°C. "Low" viscosity ranges from 50-200 cp while the "high" range is 500-10,000 cp. These products are highly resistive to evaporation. 4X10-7gm-cm²/sec At 175°C is typical. See K. Raj and R. Moskowitz's published paper, Commercial Application of Ferrofluids, Table 2, for more physical properties.

Ferrofluids have been commercially available for about two decade s ince about 1975. A single ferrofluid device may cost about 50% more than the conventional. However, they may last twice as long.

Unfortunately, the source information is in terms of motors, steppers, and dampers. Perhaps with gels (and whatever else is needed) another "muscle" technology may be near ready.

...Back to web site code.

Want to contribute, comment, etc.? Write me! ... Your suggestions are welcome.

as of July 1, 2001 ... Back to the Android Making Encyclopedia Page of/or to The Android Maker site's home page.