AutoRun Thumb Drive?

Joerg wrote:

It appears that your client needs to learn how to write in a style that doesn't look like spammer-talk.

Yep ;-) ...Jim Thompson

Nope. It was a clean matter-of-factly email along the lines of "We have a matching problem with the RF circuit of the xyz unit. Can you help us on that? Schematic is attached".

The Smith chart had to come out for this one.

Joerg wrote:

So, listed under "Hairy". Did someone here once say that one of the kilobuck RF suites has an on-screen Smith Chart that takes input?

This may or may not relate to your problem: As the spammers change tactics over time, it is a good idea for you to occcasionally just flush the filter. The old memes sometimes deep-six stuff that is outside the bounds of what is currently spammy. Kinda like how JT disables his Usenet filters now & then to see what (if anything) he's been missing.

Maybe. But I'd rather apply the kilobucks towards top quality brewsky over the years and continue to use pencil, compass and paper :-)

Sure. But again: Why don't spam filters do the obvious? It's so trivial: Take a peek at the "Sent" folder and then quit blocking any domain that emails have been sent to. Because, obviously, those are established and desired communications channels and not spam.

The same could be applied towards Jim's filters. If you respond to a thread started by a guy from a blocked domain a message pops up "Should this address be white-listed?"

You're trying to avoid all effort on your own part. Most Bayesian-based filter "learners" are crap... except perhaps for E-mail filtering.

I update my white-list mostly by observing follow-ups and seeing a name I don't recognize... copy-paste into white-list.

Think more like a sequential programmer and you'll see how easy it is to write filters, particularly for exceptionally ignorant posters such as Nymbecile ;-) ...Jim Thompson

If I had a white-list function. Since TB is open source it's probably possible to write filters and such but it ain't that important to me. I'd rather design stuff, like an RF HV-generator right now :-)

Well, yeah, but some can be funny at times :-)

Joerg wrote:

As Jim just said, filters get applied in LAYERS. If you decide NOT to use a layer, that's on you. Bayesian filters only classify words

This is why the laziest guys make the best engineers: they want to get stuff done with the fewest brushstrokes. Someone actually has to to the first iteration, however. (Thunderbird[1] is Open Source; if you don't like how it handles, *you* could have a go at it.) . . [1] nFilter too IIRC.

Hasn't Jim just admitted to be one of them?

Yep, exactamente. But we are still better than some politicians who, when a problem manifests itself, just sit it out and wait until it goes away or is forgotten.

Ok, I am not at all a programmer. I can write a few lines of C or assembler if you twist my hand but I am not at all good at it. Python? Until about half a year ago I thought that was a large snake. And RAM is a Dodge pickup truck :-)

No. I'm a _thorough_ study ;-)

RAM is a Q45. Wanna drag ?:-) ...Jim Thompson

Same here. The pond pump was connected with an, ahem, how should I word it, "relaxed code version" of cabling. I pondered a proper and durable solution for eight years or so. A few months ago my wife decided that the pond was too much work and we dug it up. Problem solved :-)

I don't have a Dodge Hemi so first I'd have to borrow this car:

In the '60's I knew a guy who worked at Air Research (now Honeywell) who put an APU into a Volkswagen. He found that a prop was more efficient than "jet effect". ...Jim Thompson

Well, he used a GE T58. This ain't your ordinary APU. An extra 1350 horses do make a difference.

Or how about this one?

There's even a version for soccer moms:

If you have to have the side doors open for air flow, you lose structural rigidity... Chrysler products start out as crap... you want to chance it ?:-) ...Jim Thompson

=20

is=20

JT in drag? What a concept?

JCNR