[ExI] Avoiding Spam

[Kelly Anderson]

Back in 2003 I started a one man company Habit3 Software that sorted email
in order of importance. It used the same technique as SpamAssassin, but
added a few hundred rules looking for positive attributes like "this email
came from someone I've previously sent email to" and the like. I even added
"roles" so that the rules could be applied differently depending upon
whether you were working, or doing family, or some other activity that you
regularly engaged in. It worked relatively well, though it was
computationally intensive. I wrote it in C# because I like C#, but later
found out it was INCREDIBLY bad at efficient string processing. Anyway, I
got talked into working on something else for a long while. Does anyone
know if anyone else implemented an email sorter like this in the meantime?

I've thought about the idea of recycling some of my old programming ideas
now that AI does heavy lifting in the coding arena... I'm still not
convinced that it would be worth my time though.

-Kelly


On Thu, Oct 17, 2024 at 1:57 PM John Klos via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> Hi,
>
> > Personally, I don't get much spam, and it's easily spotted and removed.
> I
> > just thought it might be worth reviving this old idea I had a few years
> ago
> > but didn't do anything with (because I don't really have a spam problem).
>
> This solution has been around for ages, even in spite of Google making a
> big mess of it and having security issues because Google's too big and
> dumb to follow standards, even standards that've existed for decades
> longer than they have.
>
> However, some spammers have picked up on this and enough email address
> scrapers just remove the + and anything after it. It still helps, though.
>
> More than a decade ago I started doing something like this, but I create
> an actual address for each use. If I start getting spam at, say,
> john.coffeeshoponthecorner at whatever.com, which I'd've created and used
> only in one specific place ever, I know that that coffee shop was
> compromised or they shared my data. Large corporations like Oracle, Adobe,
> Amazon, Avid, Google, Microsoft and eBay have "shared", whether knowingly
> or via compromise, my email address with spammers. It has helped to reduce
> spam tremendously because I can turn off addresses when companies go out
> of business, become shitty, or simply don't care about security.
>
> > I guess the existing measures are good enough for most people. And I
> suppose
> > if you're on the alert for spam, you're also more likely to catch other
> bad
> > stuff that's far worse.
>
> Having run email servers since the '90s, I have some strong feeling about
> and some good experience with dealing with spam. The biggest problem is
> the webification of the Internet. You can get a domain and hosting,
> registered anonymousyly, hosted anonymously, with working reverse DNS,
> with anonymouns WHOIS information, with TLS certificates, DMARC and SPF,
> with practically no work at all, and so long as companies like Cloudflare,
> OVH and DigitalOcean will take your money and let you do whatever you
> want, with no real consequences and with complete anonymity, we'll have
> spam forever. They profit from this, and Google profits by people being
> afraid to have their email anywhere except the same cesspool where
> everyone else is, too.
>
> The Venn diagram of extropians and technical people has plenty of overlap,
> but spam and email is probably not a real interest of most people here.
> But if anyone is interested in spam and email stuff, particularly
> self-hosting email, I'm writing up some articles and would love feedback.
> If you're interested, contact me off-list :)
>
> John
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20241020/631648ef/attachment.htm>