Older blog entries for scandal (starting at number 61)

I had been having problems with my D-LINK DWL-650 802.11b card loosing link with my base station quite frequently. I also could not upload even a tiny amount of data without loosing link. I had been using the kernel orinoco_cs driver, but decided to try the linux-wlan-ng driver instead to see if it would be any better. Deep inside a README somewhere I found out that this is a known issue with these cards. Fortunately, the linux-wlan-ng driver has the ability to do the WEP encryption/decryption on the host cpu instead of on the card. Using this setting, I have not had a link failure so far! Unfortunately, Red Hat has refused in the past to include these drivers due to their "unsuitablity" for inclusion into the kernel. I hope that they will be included with Fedora at some point. While it works with a standalone install, none of the configuration is integrated with Fedora's GUI system config tools, which is a bummer.

According to this /. post about new features in the Linux 2.6.0 kernel:

  • The ext3 filesystem has gained indexed directory support, which offers considerable performance gains when used on filesystems with directories containing large numbers of files.
  • In order to use the htree feature, you need at least version 1.32 of e2fsprogs.
  • Existing filesystems can be converted using the command tune2fs -O dir_index /dev/hdXXX
  • The latest e2fsprogs can be found at http://prdownloads.sourceforge.net/e2fsprogs
This might be interesting for those of you with very large Maildir folders.


I've started following the mutt-dev mailing list again. So far I've managed to fix at least one bug per day, which is a good way to get back into the swing of things. My current goal is to dust off the Maildir header cache patch and address the numerous issues that were discovered. There still seems to be quite a bit of interest in this feature for those people who just refuse to archive old email. :-)


I ran across an article on The 10 Biggest Spam Myths in which it stated the following:

7. Never opt out.

The public's heard this so often, they accept it as gospel. A recent Bigfoot Interactive study found 58 percent of respondents believe unsubscribing from unwanted e-mail actually results in more unwanted e-mail. Bad as the spam problem is, sometimes good judgment and common sense can prevail. Educated (not just alarmed) consumers are less inclined to report as spammers known and trusted senders just to get off their lists.

I have to admit that I bought into the conventional wisdom on this topic. At work, I had been getting to nearly 300 spam messages per day, so I figured that I didn't really have that much to lose by trying an experiment to see what would happen if I started clicking on all the unsubscribe links in my spam. The spammers obviously already have my email address, and the rate of spam has been increasing over the past few months, so what did I have to lose? I've been keeping a spreadsheet with my daily spam counts for awhile now just to track my own experience. This helps me make some more concrete statements about how much spam I'm getting, rather than relying on my feeling about it. This helped me see the result of clicking though unsubscribe links in a measurable way.

Most of the web pages that handle unsubscribe links state that it will take up to 48 hours to process the request. Fair enough (although it seems to me that it would take only a few seconds to reap an address from a proper DB). So for about a week I clicked on every unsubscribe link I could find in my spam. To my surprise, only a small portion of those emails providing links got a 404. However, a lot of my spam is in Asian character sets like GB2312 or Big5 and I obviously could not figure out how to unsubscribe from those, being only an English speaker. After about two days, I did notice a dropoff about about 100-150 messages per day (roughly half the normal rate).

As many of the recent reports on the legislative attempts targetting spam have pointed out, using opt-out systems is extremely annoying to the end user. I spent a good deal of my morning time clicking through the remaining messages which actually provided URLs to unsubcribe. And there is still all that mail which I can't get rid of either because I can't read the email or the messages don't provide unsubscribe links... This made me question the usefulness of wasting my time unsubscribing even to those messages which do provide legitimate ways to unsubscribe.

The next part of my experiment was to stop clicking on unsubscribe links to see how that affected the rate of spam. I had previously noted that the rate of spam was increasing without my doing anything, and so far that appears to still be the case. Almost immediately my daily spam counts jumped by 50 messages per day when I stopped. Just yesterday I was back up to 211 messages. I suspect within another week or so I will be back up to right where I started at the beginning of this exercise.

Finally, I will just put in a plug for the SpamBayes Outlook Plugin without which my work email account would be completely useless. It is quite apparent to me that only technical means is going save email from being useless. Even if by some miracle the CAN-SPAM act does succeed in chasing spammers out of the US, I will still be deluged by an ever increasing amount of spam from the outside.

I got to wondering how other people are using the diary interest ratings, because for me the only useful values are not to rate someone at all, or give them something less than three (so they don't show up in the recentlog at all). Are there plans to be able to sort the recentlog page by highest-interest first? That would make the finer grained ratings a little more useful in my case.

15 Apr 2003 (updated 15 Apr 2003 at 15:58 UTC) »
11 Feb 2003 (updated 15 Apr 2003 at 15:59 UTC) »
6 Feb 2003 (updated 15 Apr 2003 at 15:59 UTC) »
1 Feb 2003 (updated 15 Apr 2003 at 15:59 UTC) »
26 Oct 2002 (updated 15 Apr 2003 at 16:00 UTC) »
24 Oct 2002 (updated 15 Apr 2003 at 16:02 UTC) »
24 Sep 2002 (updated 15 Apr 2003 at 16:03 UTC) »

52 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!