Development of optical systems is toward smaller sizes, more new functions and improved device performance. However, traditional optical systems face many fundamental limits. For example, the Abbe’s diffraction limit sets a minimal value at which the optical microscopy, telescopy and lithography resolution can be achieved. On the other hand, the Snell’s law states that one can only change the shape or materials of optical elements to realize different optical functionalities. Recently, metasurfaces based on artificially structured thin films with unique optical properties are thought to be a promising alternative to traditional optical materials. The thickness of these structures is far smaller than light wavelength, allowing the miniaturization and integration of several optical components and systems together. But there is still not a universal theory available for the understanding and design of metasurface-based devices. Luo et al. [1] systematically brought forward a new concept of metasurface wave (M-wave), which is quite different from the traditional bulk electromagnetic wave. The M-wave is a special electromagnetic excitation, either propagating along the surface or tunneling through the surface with abrupt phase and amplitude changes, which provides a general theory to describe the principles of ‘‘planar nanophotonics’’. They used the M-wave to extend the traditional laws of diffraction, refraction, reflection and absorption. In particular, they revealed that the M-wave has three unique optical properties: extremely short wavelength, locally tunable phase shift and tunable chromatic dispersion. Based on these three characteristics, several important theories and laws to further elaborate the M-wave theory have been presented, such as the metasurface-assisted diffraction theory (MDT), metasurface-assisted law of refraction and reflection (MLRR), metasurface-assisted law of polarization conversion (MLPC) and metasurface-assisted absorption theory (MAT). These theories would pave the way in the design of novel electromagnetic devices with various improved performance [2–6]. On the basis of the extremely short wavelength property and the MDT which emerged through the M-wave, the authors proposed a plasmonic lithography technique. It was shown that a sub-22 nm resolution can be achieved by using a light source at 365 nm. These results demonstrated that both the Abbe’s diffraction limit and the near-field diffraction limit can be surpassed. Unlike the super-resolved fluorescence microscopy (The Nobel Prize in Chemistry 2014), which only applies well to fluorescence samples, this new technique does not rely on any fluorescent materials and thus has no limitation to applications, in principle. With the locally tunable phase shift property and the MLRR, Luo et al. [1] demonstrated a set of flat optical devices, including lenses, prisms, axicons and spiral phase plates. The author extended the super-resolution ability to far-field imaging systems. It was shown that a 10-m metasurface-based telescope can achieve the performance of a normal 16-m one. This is also a big improvement and contributes to revise the text book of Optics. For instance, Born and Wolf [2] have written in their Principle of Optics that the primary diffraction-limited images of an objective lens would not be observed in more detail by the following eyepiece. On the contrary, the new results show that deep sub-diffraction features absent in the diffraction-limited images of objective lens can be optically restored in the relayed system by taking advantage of the exotic properties M. Hong (&) Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore e-mail: elehmh@nus.edu.sg