spamassassin-users April 2010 archive
Main Archive Page > Month Archives  > spamassassin-users archives
spamassassin-users: Re: FROM_STARTS_WITH_NUMS matches on text-to

Re: FROM_STARTS_WITH_NUMS matches on text-to-email

From: Ted Mittelstaedt <tedm_at_nospam>
Date: Mon Apr 12 2010 - 21:57:08 GMT
To: "'users@spamassassin.apache.org'" <users@spamassassin.apache.org>

On 4/12/2010 1:58 PM, Martin Gregorie wrote:
> On Mon, 2010-04-12 at 16:29 -0400, Jason Bertoch wrote:
>> I just received a FP report on a message sent from a phone via their
>> text-to-email gateway. FROM_STARTS_WITH_NUMS matched because the
>> sender's address is [10-digit phone number]@somecarrier.com.
>>
>> My initial instinct was to file a bug suggesting there be a check in the
>> rule to see if there are 10 and only 10 numbers. However, I quickly
>> remembered SA is international software with phone numbers being various
>> lengths around the globe. I wonder how difficult it would be to make
>> location specific exceptions based on RelayCountry?
>>
>> Thoughts or suggestions?
>>
> I had quite a bit to do with phone numbers en mass a while back. My
> initial reaction is that its not easy: not only do phone numbers vary in
> length between locales, but even such things as the 'international
> dialing' and non-local-call prefix vary from country to country. My
> guess is that determining by inspection whether the number is a phone
> number probably involves a plugin and a (large) set of numbering scheme
> templates. However, the domain name might help: obviously so for small,
> single country telcos, but while the globotelcos will certainly have to
> use national number structures in each country, do they also use
> different domains for each country they operate in? My guess is that
> some do and some don't.
>
> However, there is one fairly straight forward question that can be
> easily answered: has anybody ever seen an all-number mailbox/user id in
> circumstances where it *isn't* a phone number?
>

No, but I could set one up just to piss off someone...

Seriously, you shouldn't be asking that question. The fundamental flaw
here is in the assumption that an all-number mailbox user ID is
virtually certain to be spam. It is not. Clearly, the default score
assignment to that rule is too high.

Ted

>
> Martin
>
>