Disqus - Latest Comments for Spam

Re: carlo.log → Amazement

Spam — Thu, 30 Jan 2003 00:50:28 -0000

Clean code is code that is easily understandable to one proficient in the language, as well as being aesthetically pleasing. This can mean that good algorithms are used, the right level of abstraction is used, there is little or no redundant code, the problem is well-modeled, etc.

Re: carlo.log → Nearing GLUE 1.1

Spam — Tue, 18 Feb 2003 11:46:47 -0000

Looking at the unbbcode() function, it seems there must be some way to parameterize the function instead of having one case per tag.

If nothing else, a simple list of code,tag pairs could be used for most of the simple transformations, like bold, italic, abbreviations, et cetera. Then the oddball ones like img and list could be done in special cases. That would cut down on the redundant code in there.

Re: carlo.log → Nearing GLUE 1.1

Spam — Wed, 19 Feb 2003 05:37:28 -0000

I'm not saying that it's horrible code, merely that there's redundancies that can be "factored out", if you will.

I'll post the code to the GLUE development forum. I've actually got something similar, if slightly more complicated, already working in a project that I'm doing. It doesn't generalize to cover all the tags that unbbcode() handles, but it does cover most of the simple cases.

Re: carlo.log → Nearing GLUE 1.1

Spam — Fri, 21 Feb 2003 05:35:43 -0000

Abbreviation works for me. Mozilla 1.2.1 on i386 Linux.

Re: carlo.log → Mar 12th 2003, 08:48 GMT

Spam — Fri, 14 Mar 2003 22:14:58 -0000

I don't think Bush had anything to do with this; it was just a couple Representatives who have control over the House's menu.

Re: carlo.log → Mar 15th 2003, 20:11 GMT

Spam — Sun, 16 Mar 2003 01:08:46 -0000

Bah; they can't even generate the LALR languages, much less the more fun ones, like the Turing-decidable and Turing-recognizable ones. I mean, what kind of deity doesn't even have a stack?

Re: carlo.log → Jun 27th 2003, 09:54 GMT

Spam — Fri, 27 Jun 2003 19:42:31 -0000

That's a good argument for getting rid of all HTML email. If you have to send text, then Bayesian filtering can work on it. If your mail client is set to never display HTML messages, you'll never see such spams. They'll just show up as a blank message with a couple attachments, and that's clearly spam.

Re: carlo.log → Jun 27th 2003, 09:54 GMT

Spam — Sat, 28 Jun 2003 00:58:50 -0000

@5: The way spambayes handles things now is:
Slurp the message into memory, ignoring binary-encoded attachmentsRemove all HTML tags from the message. In essence, perform the following: s///g; (spambayes is Python, but it uses PCRE)Tokenize the message. Much voodoo here, and this is where most new spambayes development is taking place. Using the token database, take the 15 top-weighted tokens (weight: distance from 50%) and classify the message

The presence of HTML comments gets ignored since HTML gets stripped. You don't get tokens for, say, number of comments, but a string like VIAGRA will get transformed to "VIAGRA" before the tokenizer gets to it. It's a partial solution to the problem, but in my experience, it works very well. I haven't had a false positive in months. In fact, the only false positive I've gotten this year was when a friend of mine forwarded me an (unintentionally) amusing porn spam that he got.

Re: carlo.log → You’ve got mail

Spam — Wed, 20 Aug 2003 20:32:32 -0000

@6: A further problem is that stupid virus filters send notification back to the poor guy in the From: field, when From: is trivially forgeable. In fact, the only thing not trivially forgeable in an SMTP session is the IP address of the connected client.

The virus filters where I work send the message on to the recipient after cleaning; this prevents our servers from spamming the world with bounces to people who happen to know someone who is dumb enough to run this thing.

Re: carlo.log → Sep 15th 2003, 19:18 GMT

Spam — Mon, 15 Sep 2003 23:33:55 -0000

The worst part is that you watch it five or six times waiting for the ending, and then it's stuck in your head for days.

Re: carlo.log → Oct 15th 2003, 22:10 GMT

Spam — Thu, 16 Oct 2003 10:34:43 -0000

No install? I'll have to check that out, then. I just hope Firebird fits on my little USB keychain disk with all the other crap on there. I've been thinking about making the switch, but plain old Mozilla has been working just fine for me. This may well tip the scales in Firebird's favor.

Re: carlo.log → Oct 24th 2003, 13:21 GMT

Spam — Sat, 25 Oct 2003 01:21:12 -0000

I got the same message, but on Monday. This spammer's a persistent little shithead.

Re: carlo.log → Oct 24th 2003, 13:21 GMT

Spam — Sat, 25 Oct 2003 03:24:04 -0000

I've never had any luck with direct oral saliva application. I think I'm just not flexible enough.

Re: carlo.log → Jan 29th 2004, 17:28 GMT

Spam — Thu, 29 Jan 2004 23:26:32 -0000

Could you write a script to go through the referrer log, find all those with more than X hits, and add them to a blacklist automagically? (I assume you have the skills; I just don't know if GLUE likes that sort of thing.)

It's a pain in the ass, but it beats having to scan the logs manually every time your DB usage shoots through the roof.

Re: carlo.log → Jan 29th 2004, 09:53 GMT

Spam — Fri, 30 Jan 2004 00:37:18 -0000

@10:

[Quote] Besides the reason that most 'big' games don't make it to the Mac on release is purely one of market decisions by the publishers anyway. [Quote]

Before the publisher even gets near it, the programmers can ensure that porting to the Mac (or any non-Windows platform) is a Herculean effort by doing one simple thing: using DirectX.

If your game makes heavy use of DirectX, particularly D3D, you'll basically have to rewrite the thing to port it. If you use OpenGL for the renderer, then the game may be portable.

There are other ways to shoot yourself in the foot, like assuming a little-endian architecture, using lots of inline assembly, etc. The most common way to kill portability, though, is D3D.

Re: carlo.log → Feb 10th 2004, 18:54 GMT

Spam — Wed, 11 Feb 2004 00:42:11 -0000

Part of why LOTR translated so well to the big screen is that so much of the books was devoted to describing the world. Where Tolkien spent 20 pages describing the lay of the land, the colors of the foliage, the sights in the distance, etc., Peter Jackson could just put in a ten-second shot of it.

The Ender's Game books are much heavier on plot movement and dialogue, so there's going to be lots and lots of stuff that doesn't make it in.

Re: carlo.log → Apr 15th 2004, 20:06 GMT

Spam — Fri, 16 Apr 2004 07:53:47 -0000

"In other words, we want to prove that P(1) is true, as well as P(2), P(3), P(4), and so on."

From Introduction to the Theory of Computation by Michael Sipser.

Now if you'll pardon me, I have homework to do.

Re: carlo.log → Apr 23rd 2004, 10:21 GMT

Spam — Sat, 24 Apr 2004 00:25:52 -0000

It's probably registry-related. Every random program that comes along throws a bunch of crap into the Windows registry and leaves it there forever, even after an uninstall. When the registry gets larger, access times get slower*. Hence, stuff like explorer.exe and pals, which constantly access the registry, get slower and slower as well.

The only good way I know of to fix it is to reinstall. Microsoft has a program called RegClean that will clean out unnecessary keys from your registry. The only problem here is that keys needed by competitors' programs are often considered "unnecessary." RegClean consistently broke Netscape 4, for example.

I really have to wonder how the registry is stored on disk. If I were doing it, I'd use a b-tree. Sure, accessing a deeply-buried key would be slow, but having one large, deep subtree wouldn't slow down accesses to the rest of the registry.

* No, I don't know why. I can't find out what the on-disk format of the Windows registry is.

Re: carlo.log → Jul 1st 2004, 08:43 GMT

Spam — Sat, 03 Jul 2004 00:29:36 -0000

So they can get nipple rings and never lose their keys again.