Time is in seconds, length in meters, temperature in kelvin, etc. A unit of energy like a joule is then defined using these base units, so 1 joule is 1⋅kg⋅m^2⋅s^-2.
> The kilogram is the base unit of mass in the International System of Units (SI)
Arguably, an ugly wart, but one we are stuck with for historical reasons. The base units of the original metric system (metre and gram) were poorly proportioned for practical use, resulting in the two main scientific/engineering systems of metric units both choosing to prefix one base unit - the centimetre-gram-second (cgs) system chose to prefix the metre, the metre-kilogram-second (mks) system chose to prefix the gram, and eventually mks won out over cgs and evolved into SI.
Whatever warts SI has, they are nothing compared to the chaos of the Imperial/customary system
> The base units of the original metric system (metre and gram) were poorly proportioned for practical use
What is the dealbreaker here though? Because we have plenty of "poorly proportioned" SI units anyway; e.g. it would be much more practical to have megapascal, microfarad and megajoule as base units from an engineering pov (particle physicists might disagree;).
Pascal, farad, joule aren't base units, they are derived units.
Ideally, the base units should be prefixless. Except for kilogram, they all are.
Imagine a system exactly the same as SI, except instead of the kilogram, it had the kram, where 1 kram = 1 kilogram... then the gram would be the millikram, the milligram would become microkram, the microgram would become the nanokram, etc... if you were starting from scratch, without any historical baggage, wouldn't such a system be superior? But of course, we aren't starting without historical baggage – almost everybody knows what a kilogram is, kram is a word I just now made up.
I think some derived units being "poorly proportioned" is inevitable given the physics we have.
I understand what you mean-- I was just curious about why we could not just stick with gram-meter-second (since we have a bunch of "poorly proportioned" derived units anyway)...
Using the gram would not have removed the prefixes from all commonly used units.
In the beginning, the liter was a much more frequently used unit of volume than the cubic meter.
A liter was defined as the volume of a kilogram of water. In a system were the gram was the unit of mass, the corresponding unit of volume was the milliliter.
Which of the gram and the kilogram or which of the centimeter and meter were chosen as the units of mass and length did not matter much for mechanical units, in the way they were used in practice in the 19th century.
A definite choice of the base units has become important only after a bunch of new physical quantities have been defined for use in the theories of electricity, magnetism, heat and light, in the second half of the 19th century. When dealing with so many different physical quantities, not using unique base units would have caused too much confusion. While this necessity has been recognized, for many years 2 different choices for the base units were widespread, that based on meter-kilogram (used mostly by engineers) and that based on centimeter-gram (used mostly by theoreticians). Meter and kilogram were more typical for the sizes of practical machines, while centimeter and gram were more typical for the sizes of laboratory experiments.
If the base unit were gram, megapascals would be gigapascals, microfarads would be nanofarads, and megajoules gigajoules. Similarly a watt would be what's now a milliwatt and most "everyday" powers (except in electronics) would be kilowatts or megawatts.
