RE: Last Call: IP Payload Compression ...

[Robert Friend <[e-mail]>]

Hi Bernard,

The 2 LZS CCP methods and MPPC can be negotiated in a stateless mannner
(LZS with "0 histories").  I think you are confusing the algorithms with
the protocols.  In CCP statefulness is an option, based on the
algorithm.  Thus implementation dependent.  For example, there is no
mention of sequence numbers in the CCP base rfc, this is also
implementation dependent.

I don't think a generic statement is warranted here.

Regards,
_____________________________________________________________

Robert C. Friend              Hi/fn
Applications Engineering      5973 Avenida Encinas, Suite 110
voice: [phone]         Carlsbad, CA 92008
FAX:   [phone]         email: [e-mail]


> -----Original Message-----
> From:	Bernard Aboba [SMTP:[e-mail]]
> Sent:	Tuesday, February 03, 1998 10:04 PM
> To:	William Allen Simpson; Bob Monsour
> Cc:	[e-mail]; [e-mail]
> Subject:	Re: Last Call: IP Payload Compression ...
> 
> Bob Monsour wrote:
> 
> > I would expect that in
> >the absence of IPSec, L2TP and MobileIP would make use of PPP
> compression.
> >If IPSec is present to protect those connections, then it would make
> sense
> >to use IPPCP.
> >
> 
> Most PPP compression methods are stateful, while IPPCP is stateless.
> This
> means that IPPCP is more robust against packet loss for media (like
> the
> Internet) where ordering is not guaranteed, and substantial packet
> loss can
> occur. As a result, L2TP and MobileIP *should* use IPPCP, instead of
> PPP
> compression. It makes sense for the draft to encourage this.