How do you calculate the gravity field of a single photon and what are the   
   criteria?   
      
   There *is* one answer that is published, but I'm going to skip it and try to   
   start from first principles.   
      
   This is what I resolved:   
   * the field should be treated as a quantum field on a classical curved   
   background,   
   * the curved classical background should be that which *already* includes the   
   gravity of the quantum state,   
   * the stress tensor to be used in Einstein's equation is the vacuum   
     expectation value, where "vacuum" means "zero field, in that classical   
   curved background".   
   * the classical background is a solution to Einstein's equation with that   
   stress tensor.   
      
   So, there's an iterative process that goes like this:   
   * set the metric g = eta, the Minkowski metric;   
   * repeat {   
   *    let Omega_g := |g> be the corresponding   
   classicalized stress tensor;   
   *    update g by solving Einstein's equations G(g) = kappa T_g for g;   
   * } until (g converges);   
      
   This is *not* merely the "RHS = vev" prescription, because the first "v" in   
   "vev" is now g-dependent.   
      
   This is *not* a perturbation expansion of a quantum state, because the state   
   space itself is being updated.   
   Any two quantum state spaces that disagree on which motions are inertial lie   
   in inequivalent state space sectors.   
      
   So it's actually shifting the state space across a sequence of   
   inequivalent state spaces to converge onto the state space which   
   corresponds to the gravity field of the quantum state.   
      
   The quantum state psi_g is simply carried along for the ride and remains   
   the same - as a function of g; it is only g that is being updated.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)