Matthew Palmer: Spam Filter Developers and Administrators: Dumber Than The Dumbest Person On The Internet
For the past couple of days, one of my e-mail addresses has been some
spammer's choice for forged From: lines in their spew. So, as
invalid addresses aren't high on a spammer's list of priorities, I get all
the bounces. The fact that MTAs, in this day and age, don't have SMTP session
recipient validation is unpleasant. However, with ISPs requiring all
outgoing e-mail to go through their own servers, I can kinda see where that
can break down. I still don't like it, but I'll live with it.
In amongst all the bounces, though, there's a lot of other, really
obnoxious, crap. So far, I've had five "please click this link / reply to
this e-mail so it'll go through" (AKA "please filter my spam for me")[1]. I've also had a
large number of e-mails saying that my e-mail was blocked or unwanted or
whatever, from spam filtering programs themselves.
What I haven't got any of, as far as I can determine, is any
e-mail from enraged recipients saying "stop sending me this crap!" or
anything of that nature.
The only conclusion I can reasonably draw from this data is that users know
that source addresses are forged and there's no point replying to them, but
the people whose job it is to write, maintain, and run spam software
don't. That's downright embarrassing. Not a single user was dumb
enough to assume that I really sent the e-mail, but IT "professionals" who
deal with spam for a living are.
If you are in any way involved in the production, sale, or use of
an anti-spam product that hasn't realised that the from addresses of spams
are universally forged, please shoot yourself in the head.
Really. I'm sick to death of people who should know better doing
the most stunningly stupid things regardless.
If you don't know that your software is spamming the rest of the world,
then you're still on the hook. What other dumb shit is your system doing
that you know nothing about? On the other hand, if you do know
that your spam filter is contributing to the noise, you're even
worse -- no spam has a real source address. If your software or
system spews crap because some clueless manager told you to do it, then you
need to grow some courage and ponder on the words of Napoleon:
A commander-in-chief cannot take as an excuse for his mistakes in warfare an order given by his sovereign or his minister, when the person giving the order is absent from the field of operations and is imperfectly aware or wholly unaware of the latest state of affairs. It follows that any commander-in-chief who undertakes to carry out a plan which he considers defective is at fault; he must put forward his reasons, insist on the plan being changed, and finally tender his resignation rather than be the instrument of his army's downfall.In other words, if you did it and you know you shouldn't have, it's still your fault, regardless of why you did it. Take some responsibility for your actions, for fucks sake.
[1. Every single one of which I was more than happy to confirm -- if you want other people to do a job for you, you have to deal with the fact that some of them might not do it in quite the way you expect. I encourage anyone else who thinks that an anti-spam system that requires the rest of the world to filter your inbox is stupid (even disregarding the likely problems of infinite loops if everyone had a challenge-response inbox) to do the same.
Out hiking and wading yesterday, and again today. Then the board meeting,
amending documents older than I am, then another walk to recuperate from
that (though it really went quite well). The trilliums are mostly down to
pink, still some white ones by the spring, quite gorgeous.
Unlike at Anna's, the overriding impression is lush and green and blooming.
Maybe that valley is just a very special place. Still, this spring feels
fragile, as if rain is much more important than it should be.
Oh, and nettle soup is yummy. I take back all my unkind thoughts about
stinging nettles over the years.
It turns out that I was insightful 
Practically ever since Jeff and I released
I was pleased to learn through
After months battling with the reverse cross build of gcc-4.3 to provide
After a fair amount of clean-up, a lot of waiting for review, and some
ruthless pruning, I and my colleagues at Solarflare finally got our net
driver accepted for Linux 2.6.26. Meanwhile a larger version of the driver
made it into Xen's Linux tree, along with dependent drivers to provide a
secure fast path to the hardware from domU. Somewhere down the road we're
going to have to resolve these two versions, but now I'm just pleased to
have a working driver in-tree.
There are some other changes I've worked on over the past few months:
I made an emergency release of
I have been knocking my head against the wall a while with this bug. It's supposed that this line should create a PID file in /var/run/gkrellmd.pid but there's no PID file created so the daemon cannot be stopped so there's a error while upgrading-removing
Recently I had to deal with POST requests on mod_perl 2 on a custom XML-RPC webservice server we have been developing and maintaining over the last year (we are actually moving it to REST some time in the middle future). It was time to move some of our base modules from a couple of CGIs to the power and muscle of mod_perl. So, while dealing with POST requests (like XML-RPC is actually specified) I faced the issue that I wasn t being able to get the POST data on the PerlResponseHandler phase. I was probably not reading it correctly, I got deeply into it and even read some libapreq source code. All I was getting was some nifty message like not supported . After reading some documentation (and, to my own shame, the Apache2::Request front page pod) I found out that the usual: