TL-DR: IAS is not about speed but about dynamic pressure. You can read it as "This is how fast I would need to fly at sea level standard conditions in order to get the same dynamic pressure"
As Jan Hudec pointed out already, indicated airspeed is computed from dynamic
pressure. That itself is computed from total (i.e. stagnation) and static pressure, and the difference between both.
For (lossless) incompressible flow (i.e. anything slower than half the speed of sound, you might remember Bernoulli's equation:
$\frac{1}{2} \rho_\infty v_\infty^2 + P_\infty = P_T$
What is measured in flight is $P_T$ (total pressure, at the stagnation point) and $P_\infty$ (static pressure of undisturbed flow, measured the sides of a pitot-static tube). The difference between both is called dynamic pressure ($q_\infty$), and allows you to compute velocity:
$v_\infty = \sqrt{ \frac{2q}{\rho_\infty}} $
...except you need to know the density. Density mostly follows the ideal gas law, which means that it varies with pressure and temperature, but it also changes with humidity.
In actual aviation, such effects can be taken into account for the purposes of navigation. You can work out the actual airspeed plus windspeed fairly well -- but for the purpose of designing an aircraft, dynamic pressure is actually more important than true airspeed, and also easier to compute. So when designing an aircraft and calculating loads, as well as as minimum and top speed, indicated airspeed is actually more useful than true airspeed, as it scales directly with dynamic pressure, and all aerodynamic loads on an aircraft scale directly with that.
That is why for example take-off, landing and stall speed are quoted as IAS. The important bit is not how fast you're moving but whether you're generating enough dynamic pressure to stay in the air (or take off, and so on...). When you're taking off in Lhasa you need the same IAS as in New York, but you need a higher TAS in Lhasa to compensate for the thinner air.
Now, for compressible flow (very roughly: going faster than half the speed of sound), the incompressible Bernoulli equation is not that accurate anymore. Prandtl-Glauert transformation can kinda make it work a bit further, but really it makes a lot more sense to stop thinking about velocity and use Mach number instead, and the isentropic state change equations. They result in this handy equation to work out velocity:
$v_\infty = M * a = \sqrt{\frac{\gamma-1}{2} \left(\left(\frac{P_T}{P_\infty}\right)^{\frac{\gamma-1}{\gamma}} -1\right)} \sqrt{\gamma R_s T_\infty} $
($\gamma \approx 1.4$ for air, $R_s$ is the specific gas constant, and $T_\infty$ is static temperature)
It's perfectly feasible to use this to work out the true airspeed (given some information on temperature and humidity, and calibration of the probe given that it doesn't quite see undisturbed flow) -- but except for navigation, IAS is still used when designing and flying an aircraft.
Reason: It's all about dynamic pressure. Someone who uses IAS is usually actually interested in dynamic pressure, not how fast the aircraft is moving. Using a velocity rather than a pressure is done out of habit (because for many decades, IAS was the only number available to pilots), and because it actually approaches true airspeed at lower Mach numbers and altitudes (i.e. for take-off and landing), so most pilots already have a "feeling" for those numbers. That means it makes sense to keep using IAS instead of straight-up pressure.
This is why even transonic aircraft have not just take-off, landing and stall speed quoted in IAS, but even top speed is limited not just in terms of Mach number but also IAS. This means: You're fine to fly at Mach 0.85 at 11km altitude but if you need to make an emergency descent you can't stay at M=0.85 all the way down to 2km because the aircraft was not built to deal with the dynamic pressure you'd get in the much thicker air at lower altitudes if you did not slow down.