Matthew Miller writes:
On Sat, Dec 08, 2018 at 11:58:18AM -0500, Sam Varshavchik wrote:
> >But who knows ? Maybe there is a memo.
> I'd be shocked if there was any public discussion of this, anywhere.
Prepare to be shocked! We've talked about this quite a bit over the years.
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.o...
These scripts are old, fragile, hard to maintain, and actually not working
and well-understood. They have real bugs and problems. Moving to a single
code path is a big improvement.
This is a sleight of hand. This is not about removing redundant code paths,
but preserving backwards compatibility.
> This is just more of the same: replacing simple, working,
> well-understood infrastructure with an over-engineered hairball and
> intentionally obfuscated documentation, making it possible to
> generate revenue from providing actual, usable knowledge needed to
> effectively use the replacement frankentools.
Whatever flaws we've got, this allegation is so far off of the mark that
it's not even in the ballpark. Never attribute to malice that which can be
explained by *this stuff is hard*.
I'm trying to understand what prevents having a shell script, or two, that
takes a single parameter, and turns on or off the specified network
interface. And then calling those scripts "ifup" and "ifdown".
Why does this have to stop working, as it does now?