In Perspectives on Statistical Thermodynamics, Yoshitsugu Oono, it is written that
Do not conclude, however, that since the second law is invincible, information thermodynamics is unimportant. It is possible for a system we are interested in to perform something that ordinary thermodynamics does not allow, and to pay the price somewhere else. For example, it may be possible to go beyond the limit of thermodynamics for the time being by shifting the average value of fluctuations, although the price must be paid (dearly) later. There must be such cases in molecular machines. It is argued that cell senescence is due to the accumulation of defects and wastes, but cell death could have a more active implication of carrying away the debt caused by utilizing information. Thus, entropic apoptosis and entropic altruism are conceivable.
but I don't understand the following two points.
- Why is cell death associated with "debt caused by utilizing information"? Do cells die for increasing entropy to compensate the entropy decrease which caused by their using information (but what kind of information)?
- Why can we conclude "Thus, entropic apoptosis and entropic altruism are conceivable."? Does this mean, to let other cells decrease the entropy by consuming information, some cells kill themselves to increase entropy?