Detectors of Internally Reflected Cherenkov light (DIRC) are powerful devices for charged particle identification (PID). Indeed, the primary detector to separate kaons and pions up to few GeV/c in the barrel region of the BABAR experiment was based on the DIRC technology and performed extremely well over almost a decade of operation. In the first part of this talk, we will review the DIRC principles (charged particles emit Cherenkov light when crossing fused silica bars; part of the photons are trapped by total internal reflection and propagate in the radiators until a camera where they are detected by photon detectors) and the associated experimental challenges. We will focus on the BABAR ring-imaging Cherenkov detector, the DIRC. After presenting its design, we will summarize its performances such as the experience gained by operating this apparatus at the SLAC PEP-II asymmetric B-Factory, which delivered more than 500/fb of integrated luminosity with a peak luminosity at 1.2 10^34 /cm^2/s. Then, we will describe the R&D program which aims at designing a new generation DIRC detector, able to work at much higher luminosity (and background) while keeping excellent PID performances. This effort, which started during the last years of the BABAR running, has led to a design for the Focusing DIRC detector (FDIRC), which is currently being tested at SLAC in a cosmic ray telescope. The main difference between the BABAR DIRC and the FDIRC is a completely redesigned photon camera: small blocks of fused silica (the same material as the DIRC bars in which charged particles generate the Cherenkov light) including two mirrors and instrumented with fast multi-anode photomultipliers now replace the huge tank filled with ultra-pure water used in the BABAR design. We will also briefly present a concept of DIRC-based detector for PID in the endcap region of a HEP detector, originally designed for the SuperB project, now terminated due to lack of funding.