My dear Mr. Stone,

> Please read for comprehension.

I believe I did.  Apparently, though, either you didn't or I failed to
write sufficiently clearly for (your) comprehension.

> There's no good nor compelling reason for a kernfs alternative to
> PF_KEY to exist *at all*.

No compelling reason, no.  But good?  As good as anything in kernfs.
(How good that is is, of course, a subject for much opinionated heat
and comparatively little light.)

> Thre are quite good reasons not to want to pull in all the cruft that
> comes with kernfs, just to get one piece offunctionality they *do*
> care about,

This is entirely true.  However, it is a reason that kernfs should not
be required to get at the data (ie, a reason that kernfs should not be
the only interface) not a reason why it should be impossible to get at
the data via kernfs (ie, a reason that there should be no kernfs
interface).

I don't think I have ever been arguing that kernfs should be required
to get at the data.  What I don't understand is why you (and others) so
resist adding kernfs as an additional, optional, interface to getting
at the data.  (Not just in this instance, either; the "evil kernfs!
kernfs bad!" reaction I wrote of is much more general than this
particular instance.)

> Incorrectly reporting that sentiment as ``"evil kernfs! kernfs
> bad!"'', (as you did), borders on dishonesty.

The sentiment you write of is not what I was referring to.  Rather, I
was referring to your responding to a desire that kernfs be one
possible, optional, access path as if it were a desire that kernfs be a
required access path.  _That_ is the anti-kernfs reaction behind which
I have been unable to see any rational basis - indeed, not even any
real irrational basis (an example of the latter might be NIH syndrome).

> Which doesnt exactly encourage anyone to try explaining it all over
> again.

If you really can't see any difference between "why should kernfs be
unable to get at that data?" (which is what I thought I was asking) and
"why don't you require kernfs in order to get at that data?" (which you
have answered twice), then no, I see no point in your trying to explain
it any further to me, because we clearly are not speaking the same
language, however much it may sound like it.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B