spf-discuss
[Top] [All Lists]

RE: Namespace

2004-01-30 07:12:13
But what are the odds of the draft making RFC status if it is not
changed? And what are the odds if it is changed to use _spf?

For most people experienced with DNS, and probably the IESG as well,
the current solutions looks somewhat unclean.

This is true, and one reason I proposed the change. I suspect that the
prefix will get added regardless of the venue. It is not likely that 
spf will be the only work item in the group. Spf core is almost complete
as-is. There are some issues with the text (inevitably) and there will
be feedback from developers, but the work items that are likely to take
most time are the issues of accreditation, notice and cryptographic
auth. These were all on the schedule in Aspen and the various industry
meetings.

The problem with approaching the IETF is that they have in recent years
shown very little respect for practical deployment issues. 

On DNSSEC I spent four years trying to get a minor change made to the
spec that would have allowed DNSSEC to be deployed in dotCOM. If the
change had been made DNSSEC would have been deployed two years ago when
we deployed the ATLAS infrastructure. This deadline was well known to
the people involved, it is quite easy to hide a project costing a half
million or so in a major upgrade costing twenty times as much.

Instead the greybeards decided to go for a system that would increase
the data volume in the DNS system by a factor of six overnight. Taking
the data from a gigabyte to six gigabytes. Due to the way that ATLAS
is designed that means you have to have three times the number of machines
and six times the amount of RAM. That is a huge cost. So DNSSEC has gone
back to being ignored.

The response I got to all this was 'dotcom is too big anyway, we don't
care if DNSSEC can't be deployed in dotcom'.


In spf terms it would be like us deciding to design spf in a way that 
made it impossible for exchange and lotus users to use and difficult for
sendmail users, then refuse to make a minor change to accomodate.

That anyway is the reason I am not at all happy with the proposal to 
take the spec to the IETF, and in particular to the same IETF area
that sabotaged DNSSEC. Microsoft, IBM and Sun can tell you similar 
stories.

The core constituency for this spec is actually the anti-spam filter
companies and the MTA providers.

                Phill

-------
Sender Permitted From: http://spf.pobox.com/
Archives at http://archives.listbox.com/spf-discuss/current/
Latest draft at http://spf.pobox.com/draft-mengwong-spf-02.9.5.txt
Wiki: http://spfwiki.infinitepenguins.net/pmwiki.php/SenderPermittedFrom/
To unsubscribe, change your address, or temporarily deactivate your 
subscription, 
please go to 
http://v2.listbox.com/member/?listname(_at_)©#«Mo\¯HÝÜîU;±¤Ö¤Íµø?¡


<Prev in Thread] Current Thread [Next in Thread>