[R-pkg-devel] tibbles are not data frames

[Jim Lemon]

I suppose that it is far too late to offer such a suggestion, but it
seems to me that the problem is in some measure the mechanism of
inheritance.

First, the tibble (although the name is incomprehensible, why not
something like "data.blob") is superior to the bog standard R
data.frame.

This may not be a good metaphor, but consider the problem of including
tigers in the mixed martial arts competitions. Tigers are much better
than the average (or perhaps all) MMA fighters at damaging their
opponents. However, they change the whole game. All of the usual
techniques are out the window if one encounters a tiger.

Suppose the tibble (or data.blob) did not inherit from the data.frame,
but had a different path of inheritance. Like the evolutionary
development of Felidae and Hominoidea, it would branch way back around
Mammalia. Then it would not fool the referees into letting it into the
MMA competition. If one wanted to use the improved functionality, it
would not be necessary to consider whether this thing that said it was
a data frame had too much hair and retractable claws. I can't say
whether this is an effective suggestion or even a good one, but I
thought it was worthwhile making.

Jim