Re: [Gnu-arch-users] TLA on Cygwin

[Karl Waclawek]

> Given all of this, my question is, how objectionable would eliminating the
> path name redundancy from arch be?  Something a la
> http://lists.fifthvision.net/pipermail/arch-users/2003-July/030199.html
> perhaps?  I realize it is essentially changing the arch "protocol", but it
> seems far easier than hacking tla or ArX to use the Windows API for all file
> work and would improve the portability of the software even to Posixly
> incorrect operating systems of which Windows is not the only one.

As a newbie who hasn't even done much with Arch, maybe I am not reading
this article from July right. It seems to indicate a quadratic growth
in path length. Example (if I am reading it right):
Assuming each direcory/file name has components of 3 characters separated
by two dashes each. For simplicity (counting one set of dashes too many)
let's therefore assume we have 5 characters per name component. Let's also
disregard the slashes separating the names. With that we get the
following growth pattern (formula: 5 * 1/2 * depth * (depth+1))

== depth ==    == length ==
    10             275
    20            1050
    30            2325
    40            4100
    50            6375

If that is right, the question (for a newbie) is: How deep will Arch 
generate directory structures in a worst case scenario? Anything
to worry about?

Karl