I stopped getting comment spam when I installed a version of my blog with a captcha requirement. Now I get the expected one comment every couple months.:)
The problem now is that there's a comment spammer going around who tries to spam my blog anyway. In the process this spammer throws up hundreds of bogus referer values and makes my referer logs virtually useless. Plus (s)he's wasting my bandwidth in the futile process of trying to comment spam.
Blocking is tough becaues the spammer uses so many domains and IPs from all over the place (likely due to open proxies).
Several bloggers are following the problem, but the best details are found here.
SP2 does not get along with the latest versions of Cygwin, at least if you're using Cygwin's SSH. It all still works if you have a slightly older version of Cygwin.
Yeah, this is what's keeping me off Thunderbird. I am still using Outlook, but with the cool extension Outclass which is an Outlook front-end for POPFile. It works really well. I know I could use POPFile's web interface, but it's so much nicer when it's integrated into the mail app. If Thunderbird can use Bayes for spam, open it up for other uses!
They then can detect detect things like if a card is used in more than one location, or
if more than once in n minutes.
This second one screwed me, a first-time visitor to NYC. We took the stairs down to the subway at a station somewhere near Times Square. I slipped my Metrocard through and entered, only to find out that in this particular station, you could only get to the other side of the tracks by going back up to the street, coming down another set of stairs, and reentering the gates. The card reader promptly informed us that we were reusing our cards too soon. It's not like I was trying to simultaneously use it halfway across the city or something. After an unpleasant conversation with a bitchy and hard-to-understand attendant, we were allowed to enter the correct platform. I think the Metrocards are too picky!
I don't have many companies to compare to, but Dell consumer support blows if you have more than a simple problem.
I bought an Inspiron 5100 last fall and it had a weird problem where the right CTRL key didn't work in combination with a few of the keys on the left side of the keyboard. They were good about replacing the keyboard and motherboard quickly, but neither fixed the problem. A replacement system had the same problem.
I found it almost impossible to convey this to overseas tech support via phone or carefully worded email. I got into loops where they wanted to replace the keyboard again, and also comments like "who uses those keys?" (CTRL+TAB in Mozilla anyone?)
Two months later (and probably a year off my lifespan), I had arranged to send the system back for a refund when I ended up fixing it myself with the latest BIOS, which they had a chance to install themselves when it was at the repair depot days earlier.
I'm sure Dell's consumer tech support is fine for telling someone to make sure the power is on, or plugging in the mouse, but they have no way to escalate an unsolved problem to someone who is truly interested in fixing it.
I just wonder if any other PC vendor is better? I may shell out for the Powerbook next time.
What happens when a spammer goes researching SPF records and finds that I have allowed both mydomain.net and smtp.myisp.net to send email on behalf of mydomain.net? All he has to do then is find a compromised machine on myisp.net and start spamming people using a mydomain.net address. Or does something prevent that?
I think the MSNBOT does obey robots.txt. I blogged back in mid June about rejecting MSNBOT, since it was sucking my bandwidth. 22,360 hits in half a month, using nearly 50MB. Google, on the other hand, hit 3600 pages for less than 13MB.
I set the robots.txt rejection on June 13th and my last hit from MSNBOT was on June 14th, so I think it's obeying the instruction.
I use a wiki to keep track of my to-do list. I document all of my projects on separate wiki pages.
I currently use Twiki, which is sort of a pain to set up, but has a lot of features. One plugin for it is called the Action Tracker, and it can be used to auto-generate a single to-do list from the action items on various wiki pages.
Sometimes I get lazy about the action lists, and a simple wiki page with a text to-do list works fine.
I didn't mean that WWII stuff would be in the archive mentioned in the article. I meant that it would be neat for a similar archive to be made available from 20th century material. Unfortunately, the majority of 20th century material is stuck in the U.S. Congress' perpetual copyright extension.
This is great. Imagine having tons of written history available on the net. It would give those Google guys a challenge.
Maybe there should be a Gutenberg Project for old newspapers and such. Lots of metadata for easy searching.
One of the things that drives me crazy about all the stupid copyright extensions is the amount of recent history that could be digitized. Just imagine the interesting things to be learned from minor accounts from World War II and other events. Right now it's just rotting away on paper and film.
Several DVDs we've gotten through Netflix have several minutes of trailers you can't skip. All they let you do is hold down the fast forward. I think Universal was the company that set up their DVDs that way.
Most of the cities/towns I've been in the southeast and midwest would not work well for bikes. There are very few biking lanes or paths, and the regular roads are congested with traffic. It just doesn't feel safe to ride with that traffic.
I wish my town was more accomodating to bikes...I'd be happy to ride some of the time.
What's to stop a bad guy from getting on a small boat in Canada/Mexico and motoring/sailing/etc. 100 miles down the coast or across a great lake and just getting off the boat in the U.S.? Why would they even need a passport? Seems to me if they really wanted in, they'd get in.
I hate spam too. So far today 64 made it past my server-level spam and virus filter (but were caught by the Bayes filter at the email client).
I think part of the problem is the conditioning of users that spam has caused. If I can't remember actually signing up at a site, I'm reluctant to click any link or reply to the message in the fear that it will simply confirm that my address works.
I can understand that a user might find themselves in that situation. As a sender, I'm not sure what else I can do on top of what I'm currently doing: unsubscribing any of the addresses in bounced messages, monitoring the AOL complaint feedback loop (do other ISPs have these?), and occasionally purging the older user accounts even if I haven't gotten bounces.
Sending email is a small part of my job, and I'm not an experienced mail server administrator. My goal in posting about this subject is mainly to learn more about the best practices for sending requested commercial email.
Slashdot Mirror
I stopped getting comment spam when I installed a version of my blog with a captcha requirement. Now I get the expected one comment every couple months. :)
The problem now is that there's a comment spammer going around who tries to spam my blog anyway. In the process this spammer throws up hundreds of bogus referer values and makes my referer logs virtually useless. Plus (s)he's wasting my bandwidth in the futile process of trying to comment spam.
Blocking is tough becaues the spammer uses so many domains and IPs from all over the place (likely due to open proxies).
Several bloggers are following the problem, but the best details are found here.
SpaceWeather has a spotting map. Should be easy to spot if you can find Orion and the Pleiades.
George is happy with the script. I wonder who will be Indy's CGI sidekick?
Now programmers will just have to call their operator "Isn't".
SP2 does not get along with the latest versions of Cygwin, at least if you're using Cygwin's SSH. It all still works if you have a slightly older version of Cygwin.
More info on my blog.
I checked a couple weeks ago and the problem still wasn't fixed, but maybe it is now?
Yeah, this is what's keeping me off Thunderbird. I am still using Outlook, but with the cool extension Outclass which is an Outlook front-end for POPFile. It works really well. I know I could use POPFile's web interface, but it's so much nicer when it's integrated into the mail app. If Thunderbird can use Bayes for spam, open it up for other uses!
He's probably right:
http://www.lessig.org/blog/archives/002153.shtml
Yeah, it was the unlimited day card. I'll know better next time.
This second one screwed me, a first-time visitor to NYC. We took the stairs down to the subway at a station somewhere near Times Square. I slipped my Metrocard through and entered, only to find out that in this particular station, you could only get to the other side of the tracks by going back up to the street, coming down another set of stairs, and reentering the gates. The card reader promptly informed us that we were reusing our cards too soon. It's not like I was trying to simultaneously use it halfway across the city or something. After an unpleasant conversation with a bitchy and hard-to-understand attendant, we were allowed to enter the correct platform. I think the Metrocards are too picky!
That just makes room for more copies of Willennium (if they have any left after sending it to the other states)!
I don't have many companies to compare to, but Dell consumer support blows if you have more than a simple problem.
I bought an Inspiron 5100 last fall and it had a weird problem where the right CTRL key didn't work in combination with a few of the keys on the left side of the keyboard. They were good about replacing the keyboard and motherboard quickly, but neither fixed the problem. A replacement system had the same problem.
I found it almost impossible to convey this to overseas tech support via phone or carefully worded email. I got into loops where they wanted to replace the keyboard again, and also comments like "who uses those keys?" (CTRL+TAB in Mozilla anyone?)
Two months later (and probably a year off my lifespan), I had arranged to send the system back for a refund when I ended up fixing it myself with the latest BIOS, which they had a chance to install themselves when it was at the repair depot days earlier.
I'm sure Dell's consumer tech support is fine for telling someone to make sure the power is on, or plugging in the mouse, but they have no way to escalate an unsolved problem to someone who is truly interested in fixing it.
I just wonder if any other PC vendor is better? I may shell out for the Powerbook next time.
I'm not sure I completely understand this yet.
What happens when a spammer goes researching SPF records and finds that I have allowed both mydomain.net and smtp.myisp.net to send email on behalf of mydomain.net? All he has to do then is find a compromised machine on myisp.net and start spamming people using a mydomain.net address. Or does something prevent that?
At least Fritz is not running for re-election, though his replacement is likely to be a Republican who'll be happy to follow Hatch. :(
Oh, I meant to write that the 3600 GoogleBot hits were for all of June, not half like the MSNBOT.
I think the MSNBOT does obey robots.txt. I blogged back in mid June about rejecting MSNBOT, since it was sucking my bandwidth. 22,360 hits in half a month, using nearly 50MB. Google, on the other hand, hit 3600 pages for less than 13MB.
I set the robots.txt rejection on June 13th and my last hit from MSNBOT was on June 14th, so I think it's obeying the instruction.
I use a wiki to keep track of my to-do list. I document all of my projects on separate wiki pages.
I currently use Twiki, which is sort of a pain to set up, but has a lot of features. One plugin for it is called the Action Tracker, and it can be used to auto-generate a single to-do list from the action items on various wiki pages.
Sometimes I get lazy about the action lists, and a simple wiki page with a text to-do list works fine.
I didn't mean that WWII stuff would be in the archive mentioned in the article. I meant that it would be neat for a similar archive to be made available from 20th century material. Unfortunately, the majority of 20th century material is stuck in the U.S. Congress' perpetual copyright extension.
This is great. Imagine having tons of written history available on the net. It would give those Google guys a challenge.
Maybe there should be a Gutenberg Project for old newspapers and such. Lots of metadata for easy searching.
One of the things that drives me crazy about all the stupid copyright extensions is the amount of recent history that could be digitized. Just imagine the interesting things to be learned from minor accounts from World War II and other events. Right now it's just rotting away on paper and film.
Support the Public Domain Enhancement Act!
Several DVDs we've gotten through Netflix have several minutes of trailers you can't skip. All they let you do is hold down the fast forward. I think Universal was the company that set up their DVDs that way.
I bitched about this stuff yesterday in my blog.
I'll still take DVDs over going to the movies, but the DVD experience could be better.
You're right in this case...you can use the Terraserver data...read their FAQ.
But I think another poster is correct in saying that some copyright their presentation of the data. See TopoZone for example.
As I discovered in March, you can use the Terraserver images copyright-free.
See the FAQ.
(I mistakenly believed you couldn't use the Terraserver images either when I wrote my blog in January)
Most of the cities/towns I've been in the southeast and midwest would not work well for bikes. There are very few biking lanes or paths, and the regular roads are congested with traffic. It just doesn't feel safe to ride with that traffic.
I wish my town was more accomodating to bikes...I'd be happy to ride some of the time.
What's to stop a bad guy from getting on a small boat in Canada/Mexico and motoring/sailing/etc. 100 miles down the coast or across a great lake and just getting off the boat in the U.S.? Why would they even need a passport? Seems to me if they really wanted in, they'd get in.
We get over 30 mpg in our 97 SL2 manual, and most of that is small-town driving.
Our 03 VUE with a CVT transmission only gets 23 mpg average in town but it's better than the 17 mpg of the pickup it replaced.
We love our Saturns!
I hate spam too. So far today 64 made it past my server-level spam and virus filter (but were caught by the Bayes filter at the email client).
I think part of the problem is the conditioning of users that spam has caused. If I can't remember actually signing up at a site, I'm reluctant to click any link or reply to the message in the fear that it will simply confirm that my address works.
I can understand that a user might find themselves in that situation. As a sender, I'm not sure what else I can do on top of what I'm currently doing: unsubscribing any of the addresses in bounced messages, monitoring the AOL complaint feedback loop (do other ISPs have these?), and occasionally purging the older user accounts even if I haven't gotten bounces.
Sending email is a small part of my job, and I'm not an experienced mail server administrator. My goal in posting about this subject is mainly to learn more about the best practices for sending requested commercial email.