* ppktp 80ghz bw limits resolution?
* CGF best way to determine wavelength? alternative methods?
* estimate if 40 ps includes the main contribution is electrical noise based on slope-rms calculation.
* Fig1: include amplifier in the warm since its impt.
* include another diagram for snspd biasing.
* make some model for a tail - e.g. gaussian + exp decay, tail changes w wl significantly...?
* include fibre length
* upper bound the jitter from contributions?
* heuristics is compatible with the data within stats... red chi sq.
* does the ratio change between the gaussian and the tail, and changes according to the wavelength? this source allows u to char the features of a v fast diode and make conclusions.
* graphics line thicknesses.
* optical axis make it same thickness.
* symbols for fiber couplers.