Andrew Kohlsmith wrote:
>>Let me get this straight -- we have ignorant and the willfully abusive
>>people in these countries creating or abetting spam for others to deal
>>with, and *we're* supposed to be concerned about public relations?
>>
>I don't think you're getting it.
>
>If North America (I'm from Canada) didn't have what I would consider a
>stranglehold on .com/net/org, we would be using .someserver.us, or
>.someserver.ca. And I'm also willing to bet that the .us TLD would not
>bother themselves about going after open relays in the .us TLD. Further to
>that, if someone were to put .us into this specific test I bet you'd be
>squaking about it not being right, since you have so many emails coming in
>and going out to .us TLDs.
>
>Don't get me wrong, I'd *love* for the TLDs to have the job of making sure
>that any servers with DNS MX records were not open relays (including the
>.com/net/org), but that is unfortunately a fairy-tale at this point in time.
>
I never made the claim that it's the TLD controlling organizations' job
to deal with spam or open relays. However, as De Gaulle was supposed to
have said, one must break some eggs to make an omelet. I don't care
whose job it is -- reduce the amount of spam, guys, we'll be happy to
take your TLD out of the list. Moreover, the list can be changed or the
weight set to zero. But I've had at least one yea on this subject, and I
bet a test against a decent-size spam corpus would yield decent results.
>>Seriously -- are we now going to pitch the default rules against finding
>>Big5 encoding in the Subject line because it might trample the feelings
>>of some people? Maybe they'll figure out that their get-rich-quick
>>
>
>Seriously -- are we now going to pitch the default rules against finding
>English words in the subject line because it might trample the feelings of
>some people?
>
I'm not sure what your point is here. I'm looking at the following rule --
header CHARSET_FARAWAY eval:check_for_faraway_charset()
describe CHARSET_FARAWAY Character set indicates a foreign language
These are eval rules that look for things like BIG5, Korean, Japanese,
and other character encodings in the subject line. What rule are you
looking at? I don't believe there is a rule that ranks English-language
words as spam. In fact, the rules as they exist now (see
is_charset_ok_for_locales() in Locales.pm, which forms the core decision
of ok/not okay for the above test) ALWAYS assumes USASCII and ISO8859
are valid charsets, even for Asian locations! An Asian user of
SpamAssassin would have a fair amount of customization to do if he
wanted to "return the favor" as it were and eliminate western European
charsets. So SpamAssassin *already* discriminates against Asian users
who might want to use it. So you see, I have precedent in the existing
code.
>>schemes are unwelcome and stop sending them. It catches spam for me,
>>that's all I can say. The point of SpamAssassin is, to some degree, to
>>disseminate the collected experience of its contributors. And that
>>includes unpleasant and possibly politically incorrect experience.
>>
>This is true, but are you absolutely sure that your experiences are the same
>as those for people using SA with very different email types? I'm all for
>political incorrectness, but not when it's obviously harming entire languages
>and there are literally hundreds of other tests which are correctly
>spamfile'ing the email.
>
Obviously, I can't make that claim. But as I said -- there's precedent
within SpamAssassin for this sort of discrimination. And so long as it
works for me, I have no problem with it. I have a feeling that I'm not
the only one -- there's probably more than a few of us who have little
truck with Asia and wouldn't mind viewing e-mail from those countries
with some built-in suspicion. I have postmaster@, root@, and webmaster@
all aliased to my personal account -- and that account has appeared on
Usenet and mailing list discussions, with the concomitant increase in
spam that always attends such stuff. Because of all this, I get a fair
amount of spam, a decent percentage of which is from the Asian countries
listed in the original regexp. (Actually, I expanded it from "cn|kr|sg"
to the list found in EvalTests.pm in $CCTLDS_WITH_LOTS_OF_OPEN_RELAYS.)
As they say, Works For Me.
BTW, I'm not trying to be an ass, either :-).
--
http://www.pricegrabber.com | Dog is my co-pilot.
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
