MPU-6050, Attempting to understand Krodal's code

Sorry but I don't follow the unit_8 bit,

working from the top down, I assume that all the definitions are just straight from the register map (defining all the variables in the first table?). then you have this function which I have never seen before:
// Declaring an union for the registers and the axis values.
// The byte order does not match the byte order of
// the compiler and AVR chip.
// The AVR chip (on the Arduino board) has the Low Byte
// at the lower address.
// But the MPU-6050 has a different order: High Byte at
// lower address, so that has to be corrected.
// The register part "reg" is only used internally,
// and are swapped in code.
typedef union accel_t_gyro_union
{
struct
{
uint8_t x_accel_h;
uint8_t x_accel_l;
uint8_t y_accel_h;
uint8_t y_accel_l;
uint8_t z_accel_h;
uint8_t z_accel_l;
uint8_t t_h;
uint8_t t_l;
uint8_t x_gyro_h;
uint8_t x_gyro_l;
uint8_t y_gyro_h;
uint8_t y_gyro_l;
uint8_t z_gyro_h;
uint8_t z_gyro_l;
} reg;
struct
{
int x_accel;
int y_accel;
int z_accel;
int temperature;
int x_gyro;
int y_gyro;
int z_gyro;
} value;
};

unfortunately the only bit of code i understand is the integer variables. I also do not know what the byte order is.

If you get bored of my VERY lacking knowledge I would understand as this may take a while to fully get.

Thanks again