On Mon, 17 Mar 2008, Peter N. M. Hansteen wrote:
>"Dave Anderson" <[EMAIL PROTECTED]> writes:
>
>> that, when discussing queueing, the pf FAQ mentions only CBQ and PRIQ
>> while man pf.conf(5) also defines HFSC.
>
>It's probably a matter of coming up with an example configuration that
>is simple enough to present well within the probable reader's
>attention span and fits document's format, corresponds reasonably well
>to a situation a prospective reader would regognize, and with the
>characteristics to demonstrate what makes HFSC stand out as the better
>algorithm for that particular application, in an example that is
>sophisticated enough to demonstrate a reasonably complete set of
>significant parameters while keeping the reader's focused on the
>relative strengths of the algorithm rather than getting lost in
>potentially confusing detail.
>
>That just about sums up why writing up something along those lines is
>still on my list of things to do, rather than written and published
>already. I will keep trying, the general task juggling allowing.
That's all true, and anyone who would depend on the FAQ without also
looking at the man pages is probably better off not trying to use HFSC.
It's just that my sense of the quality of OpenBSD documentation is
offended by the fact that HFSC isn't mentioned at all in text which
appears to discuss all of the options.
>I realize this probably sounds terribly condescending, that's not what
>I intended. It's just that some subjects are in fact very hard to
>write well.
Not a problem.
Dave
--
Dave Anderson
<[EMAIL PROTECTED]>
