> Now, it is time to look into code yet. And you will find that coding style
> is amusingly uniform and naming conventions are so strict as it is possible.
I didn't look at the networking code specifically, I was talking in
general. Yes, the coding style in the kernel is almost uniformly excellent,
though not quite uniform.
[...]
> BTW the talk was not about identifiers. But if you are stick on
> improving identifiers, then take into account historical aspect.
> F.e. nothing in the world would force me to initiate rename "struct sk_buff"
> or file skbuff.h or change of name for skb_put(). These names were given
> them by their parents and nobody have rights to change them.
> Actually, even parents lost this right, these names lives by their own
> life for long time.
Very true. But if it is too hard to figure out what they do and how the
whole hangs together (the names _can_ be very strong clues), there should
be some dictionary at hand. As a russian writing code in English you know
that :-)
Somebody suggested renaming them, and #define'ing the old names to the new
ones for a transition period, say during 2.5, and then wasting the old
names in 2.7 for 2.8. What is your opinion on that? I can understand that
as you work on this, you've very familiar with the current names, but
please do consider the plight of newcomers that try to work their way into
it. If it is gratuitously too hard, you're turning away possible future
valuable developers, and reducing the number of eyes that look it over and
find bugs.
-- Horst von Brand vonbrand@sleipnir.valparaiso.cl Casilla 9G, Viņa del Mar, Chile +56 32 672616- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/