Why? I mean, especially as they’re so easy to convert - 1 square meters is about 10 square feet.
I’m enjoying the fact that this thread is 15 years old and was revived on a lark.
There is no horse so dead that, even after 15 years desiccating in the merciless Sun, we can’t find enough life left somewhere in it to justify beating it for a few more rounds. 
Hey, lemme grab my stick and join right in. 
Huh?
Both Celsius and Fahrenheit have an unnatural offset in the scales, both are of comparable sizes (less than a factor of 2 different), and the vast majority of people never need to do calculations of any sort with temperature (and if you do, you almost certainly want to be using a zero-offset scale like kelvin or Rankin anyway). So the distinction between the two scales is not nearly as big as it is for most metric/American comparisons, and there’s little drawback to, for instance, using Fahrenheit for temperatures but everything else in metric.
Biologists and physiologists often perform calculations utilizing temperature because it is a critical parameter for any kind of metabolic activity, and because these are often stated in temperature differences or in a quasi-empirical formula with an arbitrary correction factor, they generally use thermometer temperature rather than absolute temperature. Food scientists (particularly bakers) may also do temperature-based calculations for similar reasons.
The scale of using Fahrenheit versus Celsius is kind of irrelevant because either the difference in scale isn’t critical or you just take measurements down to a suitable decimal place (especially since most of those measurements are digital, anyway), Temperature is fundamentally the measurement of a statistical property of a system so there is often a lot of variability from measurement to measurement in a dynamic system down near the limit of accuracy. Older literature in these fields sometimes uses the Imperial measurements but virtually anything you do in chemistry, biology, physiology, and associated fields today will be in SI because it is way easier to do calculations in a system of thermodynamically completely consistent units versus the gangbang of Imperial units with things like “BTU/hr-ft2”.
From a practical standpoint for the layperson, there is no benefit in precision of using Fahrenheit versus Celsius other than familiarity; a person cannot detect the difference between single degrees on the Fahrenheit scale, and the Celsius scale is actually more intuitive in terms of human comfort and safety, where 0 ºC is the temperature at which water freezes, and every jump 5 ºC is essentially a step in terms of comfort level, comparable to decade jumps on the Fahrenheit scale. It is actually really easy to learn to mentally convert from those scales by realizing that at ‘room temperature’ 25 ºC = 77 ºF, and every jump of 5 on the Celsius scale is close to 10 on the Fahrenheit scale (or you can memorize 30 ºC = 86 ºF, 35 ºC = 95 ºF, 40 ºC = 104 ºF, 45 ºC = 113 ºF, and the same on the downscale) and the increments between are basically two degrees on the Fahrenheit scale for every unit of Celsius. After you’ve done this for a few days it becomes second nature to just convert between and get within a degree of difference either way.
Stranger
I’m a staunch user of the English system and will defend it to everyone except a physicist. They are too hard headed. One of the misconceptions I have heard from metric users is the belief that we use all the length measurements together. Like 3 miles, 46 yards, 5 feet ,11 and 1/3 inches. But we don’t.
Miles vs kilometers: Not that much different. I give the edge to miles for large distances. And for travel we don’t break the miles down into yards or feet, we use decimal miles - 6.5 miles. Just like the metric people. And running a 4 minute mile was one of mankind’s goals and still the mark of a fast runner. I have no idea how this converts to metric. Who cares?
Yards vs meters: no difference. Pick one. The only thing we use yards for is sports. Golf, football, and shooting.and that includes artillery. We rarely break it down further.We still should use it for running events.
Cubic yards vs meters. No difference. The only place we use cubic yards is in construction for concrete. Even gravel and dirt is sold by weight.
Feet vs whatever. Feet wins. People are perfectly scaled in feet. A short person is 5 feet tall. A tall person is 6 feet tall. Don’t give me no 1.82 meters. Room sizes in houses are ideal for feet.
Inches vs millimeters. Huge win for inches. If I hold something in my hand It should be be 3 or 4. Not 75 or 100. For constructing any thing from machines to commercial buildings we use decimal inches, same as metric.
For most machining the “thou”, one thousandth of an inch, .001”, is the perfect unit of accuracy. Nothing in metric comes close. A millimeter is 40 thou, far to large. Tenth of a millimeter? 4 thou, still way too large. Hundredth of a millimeter? Sigh. Too small. One turn of a micrometer spindle is 25 thousandths. Four quick turns - tenth of an inch. Yet we call it a “micrometer”, how ironic.
The only people that use feet and inches are carpenters working in wood.
Not quite.
I recently learned from a Doper in the architecture / building construction biz that current US guidance is to dimension everything about a building in millimeters. Building an NFL stadium or airport terminal? Millimeters. Building a toolshack? Millimeters.
The math is easy, no fractions or decimals, and the tolerances work out fine. What’s really wrong with your building being 157,140 mm in length? Avoiding ambiguity or units confusion is valuable.
Your list reads like a set of arbitrary complaints and rationalizations as to why Imperial units are somehow intrinsically ‘better’ by virtual of scale and familiarity even though the scales of centimeters and meters are roughly comparable to inches and feet, not orders of magnitude in difference.
Most civil structures in the United States are using combinations of feet and inches, as does virtually all architectural engineering. Geotechnical and geological engineering uses a combination of length/area/volume units depending on the field they are applied to. Most analysis in geophysics uses SI units because of how cross-disciplinary it is but obviously any analysis or reference used in civil and architectural applications is converted to the appropriate set of reference units.
SI units aren’t intrinsically better in any specific sense of basic meaurement but because they are consistent between length/time/mass/energy/chemical/thermodynamic/electromagnetic/radiological metrics they don’t require a lot of arbitrary conversions that frequently create errors in analysis and calculations, and make the handling of significant figures and tolerances easy to perform without ambiguity, which is preferable from the standpoint of metrology. The standard centimeter-gram-second (CGS) or meter-kilogram-second (MKS) scales are generally appropriate for all ‘human-scale’ systems, and conversions using prefixes and factors of 10 or 103 are easily checked by simple order-of-magnitude verifications.
Stranger
How tall are you?