keep in mind that this is a thread about Macs, and therefore literally rife with Mac users.
If I have learned anything at all from Slashdot, it is that threads which draw the Mac users are going to have the funniest posts of any seen on Slashdot. Mind you the humor isn't always intended or even something they are aware of, which at least in this case, makes the hilarity all that much more impressive and grandiose.
That thing is cool as hell. I personally don't have any real use for it, but I still have this urge to buy about 40 of them and then just roll around in them.
My dad is travelling and living out of his car right now and writing a book. He could really make use of a lot of the features of this thing.
I have seen many a Honda Civic with a paint job that changes color as you walk around it.
Supposedly that type of paint is the most exspensive kind. Has many small metal flakes of the different colors in there.
All I know is that I can't imagine wanting that on a car, but I have seen so many with it.
Although I suppose that maybe, much like the people of lore in this article, these fellows think the changing colors makes the car magic, and therefore able to actually be fast.
I never had a console when I was growing up due to being poor (we didn't even have a tv, so even if I for some reason did have a console, it wouldn't matter much).
I would always go to my friends' houses and want to play their consoles, but for a game like Zelda, it is hard to just play for a bit and get good at it - so as a result, I never had the full Nintendo experience.
As a late 20 something now, I have a PS2 and a gameboy advance sp. I have had dreamcast and a nintendo 64 as well.
The GameBoy Advance SP is just frickin amazing - I can now play all the old games like Mario whatever and now Zelda as well and relive my missed youth!
Here in Bermuda the GBA SP goes for about $160 and each game is about $50, but even at the higher costs and lower selection, I am still having an absolute blast.
No telemarketers, but my DSL costs $250 a month, but there are no taxes, but everything costs more, but the weather is great, but there are tourists, but you can play golf, but I suck at golf.... it goes on and on, but in the end, this place rules.
yeah, I knew that it bypassed out SMTP server and we have gone through at least once running the "fix" from Norton on all of the machines (some of my users have machines so slow that in order to get any real work done, they turn off Norton since it is a system hog - if you have a slow system you notice it more - would be nice if we could just upgrade everyone's sytems, but apparently that costs money, which we aren't spending yet). The fix file didn't find any machines on the network that had the virus (we had a client sending us mail that was infected, so we started asking people to check to see if they were infected - one swore that we were the culprit, so that led to the full network check).
I will look at our firewall now to see what it has in terms of checking outbound traffic. Thanks for the tip.
I always avoid the phone, I hate it. even moreso, I hate having to answer the phone and then it isn't even anyone I know and they want me to listen to them. and it is always obvious that they don't like what they are doing.
ugh - everything about it is awful.
but since I have moved, for whatever reason, there is no telemarketing here at all. probably b/c there is either a law against it, there is not enough return on investment, or the place is so small (about 65K people) that people would start to recognize the people on the phone ("Carl?! Is that you? Why you callin' me?!")
I am the programmer and IT person at a financial firm for awhile until I spin off into my own company here doing similar things for other companies on a consulting basis.
We only have 16 or so users that are in the office and maybe another 4 or 5 that use our resources, but are pretty much never here.
Even with those, I have seen a fairly large increase in the number of our clients with the virus and then our virus scanning software reporting it getting sent to us. Fortunately so far we seem to be clean of it, but I have added some filter EventSinks on our Exchange server to block out a wider range of attachment types.
This particular one is annoying since it has 4 types of attachments that we can't universally block and get away with (.txt,.htm,.html, and.eml).
I have fingers crossed that our anti-virus software on the Exchange server will keep up with it.
yeah, I see that now - but as I said in another post, when I spoke with them about it, they never mentioned it and I never found it when I looked on the server.
I looked a few mins ago and saw that SA is on the server, but it is Version 2.43, and the first SA version to use Bayes was 2.50 - I currently use 2.60, so I don't really want to switch to their version.
I just checked the server and found spamassassin version 2.43 on there. There is no Bayes in SA until versions 2.5 and after.
I currently run 2.60 (they don't seem to have updated it in a long while now) and am too pleased with that to go back to an older version.
I will write them again and see if they are willing to upgrade the one on my server - the worst they can say is no. Well, I suppose the worst they could say was that they are cancelling my account:)
Well hot damn, didn't know it was already on the server. I looked all over the place on the server and didn't find it at all and when I wrote to them to get the okay to run the Perl script, they never mentioned it (I wasn't sure how psyched they would be if it was cpu intensive. But since they kill anything that runs over 30 seconds it doesn't matter too much).
I have a bunch of e-mail addresses and they forward around depending on where I am (less so now that I don't have a cell phone - I had filters that would forward messages to my phone as text messages if they were from certain people during certain times).
I'll have to look into the spamd - I wonder how long that has been there - I swear it wasn't when I started. Thanks!
I have seen all of the local client software and I personally have never bothered with it.
I always felt that the whole point of spam being annoying was that it wasted bandwidth. It gets sent to my server, and then I have to download it all from my server, and then it gets sorted away from my eyes in my client.
It is fairly trivial if you get enough regular mail for it to matter, and you are on a fast connection.
But I can't tell you how annoying it is to be on a slow dial-up connection and download 50 messages and then see that they all got filtered into the spam folder and that there were no "real" messages. While there is a nice feeling of seeing them all get caught, it is annoying to have to wait for a download (and pay for it) and then get no return on the investment.
That is why I always try to have the spam blocking on the server side. Although I now spend most of my time using ssh into my server and that way it isn't downloading all of the mail until I want to see something.
Perhaps if I combine the fact that I have SA on the server, and then if I also had a client side option, I would get everything properly blocked that way (the only reason stuff gets through my server setup right now is if the server is under a high load, then my SA script will time out and the mail gets through).
I assume they are just simplifying the idea for the masses - but the article says that if "word one" shows up 60 times in spam, but twice in ham, then another message that has "word one" show up 40 times is more likely to be spam.
But technically I don't think that is how the actual process works - well, the same way - but they use sub sections. If you only look at words, then you are really shooting yourself in the foot with bayesian analysis.
Instead it seems it would be much better to look at a block of characters at a time. I know from my own experience playing with Markov matricies that 3 and 4 chars work fairly well in creating English text - so I would imagine that analysis would be similar. I don't know how much the size of the chunk looked at matters as much as the fact that it looks at every character and as long as the chunk is smaller than the entire text message ideally.
So you would have a hash (dictionary if you use VB) and look at the first N characters of your e-mail and then put that in the hash and increment that value by one (in Perl: $myHash{$chunk}++). Then you move the chunk forward one character and put that into the hash and increment it up. That is the learning process. Then the analysis process would be to move the chunk over the text and add up the hits and then use the final score as a way to determine how it should be dealt with.
The way SA does it is the Bayes analysis is just one part of the whole - so if the Bayes says it really is likely spam, but 5 other things say it is likely not, then depending on what you have set the Bayes trigger to worth, it could actually be decided as ham.
Anyway, I think the whole process is neat and I love playing with Markov Matricies.
The company that I am at now is a financial services one - dealing largely with hedge funds.
The term "asset" shows up in 90% of our mail - it is amusing how many issues the companies in our sector have with poorly written filters that think "asset" is a bad word.
I suggested that we start referring to the same concept as "fuckerbabies" but it hasn't caught on yet.
Our company is small, so it works well for us. We have 16 people and I would guess under 200 clients. I took our Global Contacts list and put all of the clients in there into the whitelist.
Then I sat and watched the e-mail coming in for about a week (takes a surprisingly small amount of time with the mail load that we have). I manually sorted the mail out into spam/non-spam - mainly looking for mistakes that SA made.
I had it set to 7.5 as a trigger - I wanted it to err on the side of letting spam through instead of marking client mail (even then, our mail currently just marks the mail so that end users can filter it in Outlook easily - I wrote the script to also allow other server side options of killing the mail).
Anytime that I saw a mistake, I would add that person to the whitelist (I usually try to add whole domains to a whitelist if they are a domain that we deal with a lot and can be trusted) - but the e-mails that are from yahoo.com, hotmail.com, and all of the other services that people use for home mail, I have to do those on an individual basis.
After a week things settled down a lot.
I am still working out a few extra things that will help out a lot. We have spam coming in to addresses that don't exist here (some used to, others never did) - so if we can block those the instant they hit the server, then the less bandwidth is wasted on them. The server was confiugred to send out a message letting people know that the address that they sent to didn't exist. But if that is sent to a spammer, then they too usually don't exist on their account anymore, so then that gets bounced back with a message as well. A series of those is annoying. I disabled that on our server, but still allow out of office and confirmation e-mails.
I then setup an ASP page that lets me add white/black list users from a webpage - that way if I leave as an admin, whoever takes my place doesn't need to know about SpamAssassain and can just use that web page to handle issues if they come up (assuming they are minor false hits either way).
I also wrote scripts to learn from the mail everyday and then clean it out - for the first little bit it was all manual, but now it is automated. Those scripts just generally learn from what it already filtered out as spam and ham, which isn't terribly excellent - ideally the missed messages in either one would be manually sorted out, but I prefer to have as much hands off time as possible with the mail.
I have since modded the config files so that bayes_90 and bayes_99 score much higher since I have never seen those be wrong on spam. I also lower the required hits to 4.5 (4 was too low for us).
I had to disable the rbl checking since that doesn't work under Win32 and I also disabled auto learning since that slows things down and I was already scripting forced learning in anyway.
On my personal shared server at pair.com, that has been great for me - I have my whitelist established and I can tweak the rules. The only time mail ever gets through is when the server is under a high load - then the spamassassin perl script will time out and let the mail through. This wouldn't happen if I either could run the spamc/d, or if I had a dedicated server - I will likely get one of those eventually there at pair.com (I am very pleased with them), but they are a bit expensive now until I can justify it with a business making money off of what it serves intead of just my own personal playground.
My own personal account is on a shared server at pair.com, and I run SpamAssassin (the perl script, can't put the spamc/d on there since I'm not root). I have written on here before how I have saved myself a lot of hassle over the last few months by installing SA. I now stop 100+ messages a day (usually more like 140 now). My stats tell me that since Feb, I've stopped over 15K Spam messages. Hot damn.
Where I currently work now we have Exchange and I wanted SpamAssassin on there, but we weren't getting the money approved to put it on. So I hacked in SpamAssassin via an Exchange 2000/2003 EventSink. If you want the code for it, feel free to grab it from http://www.cardboardutopia.com/ExchangeSpamFilter. zip
But do note that if you have many users on your machine, you aren't going to want to use this - an EventSink on Exchange runs in serial, so SpamAssassain's Perl script (the spamc/d doesn't work under Win32) will get executed on every incoming mail, and it will have to wait until it is done before it gets the next one.
We process about 2000-5000 incoming messages a day and it does okay, but we have a very light load.
IM logging isn't daunting. what is really daunting is all thems computers out there. I think they should just get rid of all thems comptuers. problem solved.
I also think that they should log everything that the person ever says while in the office, and ideally, they should write out everything that they think as well, and that too should be logged.
and yes, I'm making note of this, and will save it for 3 years.
$100K pretax dollars is more than I make in pretax dollars. Anyone that makes more than me obviously does not deserve it. Therefore, they should be replaced with robots. QED
Coffee beans have something in them that actually increases your hunger levels. Caffeine overrides that mechanism, its ability to blunt hunger cravings is greater than the other's ability to make us hungry.
So by having a decaffinated coffee, you are bypassing all of the positive sides of caffeine (increased metabolism, increase FFA into the bloodstream so you can burn more fat with the now increased metabolism, blunted hunger, keeps you awake, etc), and instead you get something that just makes you more hungry.
I love my coffee, although since moving, I haven't had a cup - used to have at least 3 a day. I'm glad for it, it lowers my tolerance of it, meaning that I will have increased response next time I do have some coffee.
Slashdot Mirror
I hope you are happy in shattering my hopes of ever reaching twice as far with my two arms.
Now I can't reach twice as far until I get four arms.
keep in mind that this is a thread about Macs, and therefore literally rife with Mac users.
If I have learned anything at all from Slashdot, it is that threads which draw the Mac users are going to have the funniest posts of any seen on Slashdot.
Mind you the humor isn't always intended or even something they are aware of, which at least in this case, makes the hilarity all that much more impressive and grandiose.
I can't really imagine him as the voice in a cartoon.
Does this mean that he is moving into the kids movie realm now?
I know that he says that he rejected the following he got from wuss movies that he started in as beefcake type roles (Legends of the Fall and such).
Perhaps he is now rejecting the following he got from films like Seven and Fight Club.
Or maybe he just likes Linux.
That thing is cool as hell. I personally don't have any real use for it, but I still have this urge to buy about 40 of them and then just roll around in them.
My dad is travelling and living out of his car right now and writing a book. He could really make use of a lot of the features of this thing.
I have seen many a Honda Civic with a paint job that changes color as you walk around it.
Supposedly that type of paint is the most exspensive kind. Has many small metal flakes of the different colors in there.
All I know is that I can't imagine wanting that on a car, but I have seen so many with it.
Although I suppose that maybe, much like the people of lore in this article, these fellows think the changing colors makes the car magic, and therefore able to actually be fast.
damn cursed monkey paw
I never had a console when I was growing up due to being poor (we didn't even have a tv, so even if I for some reason did have a console, it wouldn't matter much).
I would always go to my friends' houses and want to play their consoles, but for a game like Zelda, it is hard to just play for a bit and get good at it - so as a result, I never had the full Nintendo experience.
As a late 20 something now, I have a PS2 and a gameboy advance sp. I have had dreamcast and a nintendo 64 as well.
The GameBoy Advance SP is just frickin amazing - I can now play all the old games like Mario whatever and now Zelda as well and relive my missed youth!
Here in Bermuda the GBA SP goes for about $160 and each game is about $50, but even at the higher costs and lower selection, I am still having an absolute blast.
I've moved to Bermuda with my fiancee.
No telemarketers, but my DSL costs $250 a month, but there are no taxes, but everything costs more, but the weather is great, but there are tourists, but you can play golf, but I suck at golf.... it goes on and on, but in the end, this place rules.
yeah, I knew that it bypassed out SMTP server and we have gone through at least once running the "fix" from Norton on all of the machines (some of my users have machines so slow that in order to get any real work done, they turn off Norton since it is a system hog - if you have a slow system you notice it more - would be nice if we could just upgrade everyone's sytems, but apparently that costs money, which we aren't spending yet).
The fix file didn't find any machines on the network that had the virus (we had a client sending us mail that was infected, so we started asking people to check to see if they were infected - one swore that we were the culprit, so that led to the full network check).
I will look at our firewall now to see what it has in terms of checking outbound traffic.
Thanks for the tip.
I always avoid the phone, I hate it.
even moreso, I hate having to answer the phone and then it isn't even anyone I know and they want me to listen to them.
and it is always obvious that they don't like what they are doing.
ugh - everything about it is awful.
but since I have moved, for whatever reason, there is no telemarketing here at all. probably b/c there is either a law against it, there is not enough return on investment, or the place is so small (about 65K people) that people would start to recognize the people on the phone ("Carl?! Is that you? Why you callin' me?!")
doh, need my morning coffee - meant zip - those files I wrote are the ones it scans.
back to being retarded...
I am the programmer and IT person at a financial firm for awhile until I spin off into my own company here doing similar things for other companies on a consulting basis.
.htm, .html, and .eml).
We only have 16 or so users that are in the office and maybe another 4 or 5 that use our resources, but are pretty much never here.
Even with those, I have seen a fairly large increase in the number of our clients with the virus and then our virus scanning software reporting it getting sent to us.
Fortunately so far we seem to be clean of it, but I have added some filter EventSinks on our Exchange server to block out a wider range of attachment types.
This particular one is annoying since it has 4 types of attachments that we can't universally block and get away with (.txt,
I have fingers crossed that our anti-virus software on the Exchange server will keep up with it.
yeah, I see that now - but as I said in another post, when I spoke with them about it, they never mentioned it and I never found it when I looked on the server.
I looked a few mins ago and saw that SA is on the server, but it is Version 2.43, and the first SA version to use Bayes was 2.50 - I currently use 2.60, so I don't really want to switch to their version.
I just checked the server and found spamassassin version 2.43 on there.
:)
There is no Bayes in SA until versions 2.5 and after.
I currently run 2.60 (they don't seem to have updated it in a long while now) and am too pleased with that to go back to an older version.
I will write them again and see if they are willing to upgrade the one on my server - the worst they can say is no.
Well, I suppose the worst they could say was that they are cancelling my account
Well hot damn, didn't know it was already on the server. I looked all over the place on the server and didn't find it at all and when I wrote to them to get the okay to run the Perl script, they never mentioned it (I wasn't sure how psyched they would be if it was cpu intensive. But since they kill anything that runs over 30 seconds it doesn't matter too much).
I have a bunch of e-mail addresses and they forward around depending on where I am (less so now that I don't have a cell phone - I had filters that would forward messages to my phone as text messages if they were from certain people during certain times).
I'll have to look into the spamd - I wonder how long that has been there - I swear it wasn't when I started.
Thanks!
I have never used SpamBayes, but just in general, if you are only training it on the spam, then it isn't going to get much smarter.
You have to train it regularly on both spam and ham so that it works correctly on *your* mail.
They come with a general rule set and then can get much smarter over time with training.
I personally use spamassassain and really am pleased with how much it has helped me.
I have seen all of the local client software and I personally have never bothered with it.
I always felt that the whole point of spam being annoying was that it wasted bandwidth. It gets sent to my server, and then I have to download it all from my server, and then it gets sorted away from my eyes in my client.
It is fairly trivial if you get enough regular mail for it to matter, and you are on a fast connection.
But I can't tell you how annoying it is to be on a slow dial-up connection and download 50 messages and then see that they all got filtered into the spam folder and that there were no "real" messages.
While there is a nice feeling of seeing them all get caught, it is annoying to have to wait for a download (and pay for it) and then get no return on the investment.
That is why I always try to have the spam blocking on the server side. Although I now spend most of my time using ssh into my server and that way it isn't downloading all of the mail until I want to see something.
Perhaps if I combine the fact that I have SA on the server, and then if I also had a client side option, I would get everything properly blocked that way (the only reason stuff gets through my server setup right now is if the server is under a high load, then my SA script will time out and the mail gets through).
I assume they are just simplifying the idea for the masses - but the article says that if "word one" shows up 60 times in spam, but twice in ham, then another message that has "word one" show up 40 times is more likely to be spam.
But technically I don't think that is how the actual process works - well, the same way - but they use sub sections.
If you only look at words, then you are really shooting yourself in the foot with bayesian analysis.
Instead it seems it would be much better to look at a block of characters at a time.
I know from my own experience playing with Markov matricies that 3 and 4 chars work fairly well in creating English text - so I would imagine that analysis would be similar.
I don't know how much the size of the chunk looked at matters as much as the fact that it looks at every character and as long as the chunk is smaller than the entire text message ideally.
So you would have a hash (dictionary if you use VB) and look at the first N characters of your e-mail and then put that in the hash and increment that value by one (in Perl: $myHash{$chunk}++).
Then you move the chunk forward one character and put that into the hash and increment it up.
That is the learning process.
Then the analysis process would be to move the chunk over the text and add up the hits and then use the final score as a way to determine how it should be dealt with.
The way SA does it is the Bayes analysis is just one part of the whole - so if the Bayes says it really is likely spam, but 5 other things say it is likely not, then depending on what you have set the Bayes trigger to worth, it could actually be decided as ham.
Anyway, I think the whole process is neat and I love playing with Markov Matricies.
The company that I am at now is a financial services one - dealing largely with hedge funds.
The term "asset" shows up in 90% of our mail - it is amusing how many issues the companies in our sector have with poorly written filters that think "asset" is a bad word.
I suggested that we start referring to the same concept as "fuckerbabies" but it hasn't caught on yet.
Our company is small, so it works well for us. We have 16 people and I would guess under 200 clients.
I took our Global Contacts list and put all of the clients in there into the whitelist.
Then I sat and watched the e-mail coming in for about a week (takes a surprisingly small amount of time with the mail load that we have).
I manually sorted the mail out into spam/non-spam - mainly looking for mistakes that SA made.
I had it set to 7.5 as a trigger - I wanted it to err on the side of letting spam through instead of marking client mail (even then, our mail currently just marks the mail so that end users can filter it in Outlook easily - I wrote the script to also allow other server side options of killing the mail).
Anytime that I saw a mistake, I would add that person to the whitelist (I usually try to add whole domains to a whitelist if they are a domain that we deal with a lot and can be trusted) - but the e-mails that are from yahoo.com, hotmail.com, and all of the other services that people use for home mail, I have to do those on an individual basis.
After a week things settled down a lot.
I am still working out a few extra things that will help out a lot. We have spam coming in to addresses that don't exist here (some used to, others never did) - so if we can block those the instant they hit the server, then the less bandwidth is wasted on them.
The server was confiugred to send out a message letting people know that the address that they sent to didn't exist.
But if that is sent to a spammer, then they too usually don't exist on their account anymore, so then that gets bounced back with a message as well.
A series of those is annoying.
I disabled that on our server, but still allow out of office and confirmation e-mails.
I then setup an ASP page that lets me add white/black list users from a webpage - that way if I leave as an admin, whoever takes my place doesn't need to know about SpamAssassain and can just use that web page to handle issues if they come up (assuming they are minor false hits either way).
I also wrote scripts to learn from the mail everyday and then clean it out - for the first little bit it was all manual, but now it is automated.
Those scripts just generally learn from what it already filtered out as spam and ham, which isn't terribly excellent - ideally the missed messages in either one would be manually sorted out, but I prefer to have as much hands off time as possible with the mail.
I have since modded the config files so that bayes_90 and bayes_99 score much higher since I have never seen those be wrong on spam.
I also lower the required hits to 4.5 (4 was too low for us).
I had to disable the rbl checking since that doesn't work under Win32 and I also disabled auto learning since that slows things down and I was already scripting forced learning in anyway.
On my personal shared server at pair.com, that has been great for me - I have my whitelist established and I can tweak the rules.
The only time mail ever gets through is when the server is under a high load - then the spamassassin perl script will time out and let the mail through.
This wouldn't happen if I either could run the spamc/d, or if I had a dedicated server - I will likely get one of those eventually there at pair.com (I am very pleased with them), but they are a bit expensive now until I can justify it with a business making money off of what it serves intead of just my own personal playground.
My own personal account is on a shared server at pair.com, and I run SpamAssassin (the perl script, can't put the spamc/d on there since I'm not root).
. zip
I have written on here before how I have saved myself a lot of hassle over the last few months by installing SA. I now stop 100+ messages a day (usually more like 140 now).
My stats tell me that since Feb, I've stopped over 15K Spam messages. Hot damn.
Where I currently work now we have Exchange and I wanted SpamAssassin on there, but we weren't getting the money approved to put it on.
So I hacked in SpamAssassin via an Exchange 2000/2003 EventSink.
If you want the code for it, feel free to grab it from http://www.cardboardutopia.com/ExchangeSpamFilter
But do note that if you have many users on your machine, you aren't going to want to use this - an EventSink on Exchange runs in serial, so SpamAssassain's Perl script (the spamc/d doesn't work under Win32) will get executed on every incoming mail, and it will have to wait until it is done before it gets the next one.
We process about 2000-5000 incoming messages a day and it does okay, but we have a very light load.
IM logging isn't daunting.
what is really daunting is all thems computers out there.
I think they should just get rid of all thems comptuers.
problem solved.
I also think that they should log everything that the person ever says while in the office, and ideally, they should write out everything that they think as well, and that too should be logged.
and yes, I'm making note of this, and will save it for 3 years.
I like your logic - thumbs up!
$100K pretax dollars is more than I make in pretax dollars.
Anyone that makes more than me obviously does not deserve it.
Therefore, they should be replaced with robots.
QED
By far the best proof I've ever seen.
*golf clap*
Coffee beans have something in them that actually increases your hunger levels. Caffeine overrides that mechanism, its ability to blunt hunger cravings is greater than the other's ability to make us hungry.
So by having a decaffinated coffee, you are bypassing all of the positive sides of caffeine (increased metabolism, increase FFA into the bloodstream so you can burn more fat with the now increased metabolism, blunted hunger, keeps you awake, etc), and instead you get something that just makes you more hungry.
I love my coffee, although since moving, I haven't had a cup - used to have at least 3 a day.
I'm glad for it, it lowers my tolerance of it, meaning that I will have increased response next time I do have some coffee.
sure seems to like the name Enoch.