spf-discuss
[Top] [All Lists]

Re: argh ... I wish people would RTFRFC

2004-09-07 23:05:57
At 07:10 PM 9/7/2004 -0700, you wrote:
So here we have

   20040907-21:41:31 mengwong(_at_)dumbo:~% dnstxt bluebottle.com
   v=spf2.0/pra a:209.223.237.194 a:209.223.237.195 mx -all

Setting aside the spf2 vs spf1 thing for now, I'm somewhat
dismayed to see people doing a:1.2.3.4 when the correct
syntax is ip4:1.2.3.4.

Should we change the spec to say that if the argument to an
"a" mechanism is obviously an IP address, we interpret it as
ip4 / ip6?

Here's one no vote to that question.  I don't think it's a good
idea to try and guess what people  meant when they they don't interpret the
spec correctly even if it may seem obvious. 

I'd say it's better to educate people how to use the syntax
as it is specified or perhaps there's a way to re-word the spec
that makes the intent clearer?

I suspect the confusion is between the SPF "a" mechanism
and the DNS A RR record.


And to think I was ridiculed in this discussion list in the last weeks for 
basically saying that:

1) Deployment would not be trivial nor swift out to the millions of domains

2) Warning about Murphy's Law and questioning the wisdom of complex syntax and 
macro capabilities

3) Saying the #1 priority is education and better web site for novices.

4) Considering ways (even alternative proposals) to automate the technical 
issues of deployment for anti-forgery

IMO, this is just the tip of the iceberg of "I told you so".

One of the basic laws known to people who actually publish commercial software 
marketed to masses is that people rarely RTFM.  Of course I am an "idiot", so 
my experience is not meaningful.