Science is a rather restricted discipline that is omnipotent and is not suppose to be able to answer every single questions, particularly those that lies outside the domain of its scope. It is then necessary to define what science and what is the operating principle that defines science as science.
Scientific methodology defines what science is: It only treats physical phenomena that are measurable and comprehensible (GOD and consciousness are not included in this category). All physical phenomena are governed by physical laws, which are formulated in terms of physics theory or theorems. Physical laws (a Nature’s attribute) are expressed in terms of physical theories (our conceptual construct). They are formulated based on experimental observations and theoretical construction with the aid of mathematics. The theories formulated must be in principle testable so that one can be sure that they indeed correspond to the physical laws of the real world they try to describe. They must be predictive of some unknown phenomena that is then used as a check of the theory’s validity. To pass the QA procedure, a physical theory must be checked by experiment and observation before their validity is recognised. In other words, a so-called “physical theory” that fail to pass the experimental test, or that can’t even be tested in principle, can’t be qualified as a theory corresponds to our real world.
Physical laws are supposed to be UNIVERSAL: they are invariant in time and space, valid not only here and now but also elsewhere in the Universe and also in any moment in time, be it the past or the future. The laws would remain valid to whoever does the experiments, be it a Nobel prize winning physicist or a primary school teacher from Penang.
Any physical phenomenon governed by the same set of physical laws is repeatable when the same conditions pertaining to that phenomenon are satisfied. For example, when the temperature within the core of a star is larger than a specific temperature, say, 1 million degree, the hydrogen fuel within the stars will initiate nuclear fusion releasing tremendous amount of nuclear energy according to the physical laws that govern the nuclear physics of these reactions. So, any star in the Universe must do so no matter where it is located and when it existed.
For another example, if the physical laws say that ALL pure water boils at 100 degree at 1 atmospheric pressure, then every time you boil the pure water at the standard pressure it must always boil at 100 degree always, no matter when and where you boil it. Such an observation manifests that the laws governing the boiling of pure water (statistical physics and thermodynamics) is indeed universal.
A counter example: if someone claims to have produced cold nuclear fusion on a table top device using palladium, the same result should be reproducible by other people in the other end of the Universe using the necessary described condition. If it is not reproducible by many people with all the right condition it might means the break down of the universality of physical laws. However, most often this means those who made such claim gets screwed up in their experiment. They must be logically self-consistent and preserve causality. They don’t predict things in sharp contradiction, such as the time traveling. Theory that allows time traveling is necessarily a forbidden theory from logical point of view and cannot be correct because it opens up the possibility to violate causality. Hence it can’t be right to do time travel.
The physical laws are usually formulated using mathematics as a tool to penetrate into the logical structure of the theory. The employment of mathematics allows us to probe into the physical system as precise as possible and left little space for subjective interpretation. For example Newton’s law of gravity that governs the behaviour of the Earth-Moon system can very precisely tell us where the angular position of the moon, in the form of theta = (GM/r^3)t, where t is the time correspond to the angular position theta, G a constant, M the mass of the Earth (a constant) and r the distance between the Earth and the Moon.
As a related counter example, there are some kind of ‘theory’ (such as Ying-Yang, Universal Life Force Energy or reiki) doesn’t fulfill the precision and consistent criteria of such, leaving many space of subjective manipulation, resulting in non-unique and sometimes contradicting predictions. Such an unsatisfying feature in these theories is due the reason that they are not built upon any rigorous logical structure (in the mathematical sense), hence we can’t perform rigorous mathematical manipulation in them to obtain result that are logically persuasive.
Science must be falsifiable (at least in principle). This is one of the important criteria of science. One must also propose a way how his theory can be proven otherwise (example: God’s existence is an example of a theories that cannot be falsified) in order to call it a scientific theory. For example in Newton’s theory of gravity, the inertial of any mass is supposed to be independent of the type of material. Hence one can check to see if the inertia of many different objects of the same mass (say 1.00 kg) made of different material do differs. If it does that means the theory is falsified. If not then say that the theory is consistent with experimental observation. A theory that can’t be falsified even in principle can’t qualify to be a scientific theory, such as the existence of GOD).
Science has range of validity. It is valid only in the domain where it’s empirically proven to be valid. Such as classical mechanics that describe our ordinary scale world such as car collision, rocket launching and the dynamical properties of our solar system, is not valid in the quantum domain. Quantum mechanics is verified to be valid up to the scale as small as 10^(-19) m, whereas GR is valid up to the scale of as large of the size of the Universe.
Science is approximate truth; it approaches truth asymptotically. It is always ready to be supplanted by improved versions when better insight or technique or experimental data become available. Such as Newton’s theory of gravity being supplanted by GR as a generalization when dealing with cosmological structures.
Due to the very restrictive criteria imposed on scientific by its methodology, physics can only access the physical aspect of the Universe but not the non-physical things (example: soul, ‘qi’, consciousness).
Due to the employment of scientific methodology physics is very good at telling us how but not why (example). In comparison, philosophy attempts to address the question of why (but not how). Philosopher' methods are purely logical, sometimes intuitive but not empirical (Aristotle didn’t’ perform experiment in finding out if their understanding of nature fits the observation.). At times ancient natural philosophers failed to understand the natural law because not performing experiment to confirm their supposition. In this sense philosophy is not 'scientific', hence can not be directly be compared to science. These is simply an orange and apple comparison. Both addresses different aspects of our universe employing fundamentally distinct approaches.