I'm pulling these numbers partly from my experience and partly from my ass.
Average user at work receives ~ 10 spam messages per day, with our spam filters catching 80%, so the user receives 2 per day.
I figure you can classify spam messages into 3 types which require varying amount of effort to determine when to delete them.
First, you have obvious spam that can be deleted by reading the subject. I figure that 25% that get though spam filters are this type. Let's say they require 5 seconds to read/delete.
Second type requires the user opening the message and reading the contents. Let's say 70% are this type and require 15 seconds.
The last type are the ones that confuse users or they think they are legit. These are the messages that users will reply to, talk to their coworkers about, or forward to IS to be verified. This would be the remaining 5% and require 5 minutes (300 seconds) of time.
With those numbers, the users spend 27 seconds per message or 54 seconds per day.
Per year, that is 5.475 hours. If the average cost of a user (pay + benefits) is $30/hour then the annual cost for spam per user is $164.25/year or $0.225/spam.
Now, we are running spam software in this situation. Figure that the spam software cost $10/user/year in licensing, and an additional $10/user/year in hardware or administrative costs. I'm also assuming that the spam load on our mail servers is minimal enough that the costs involved there are insignificant.
That puts the total now at $184.25/year. However, without antispam software, that total would sky rocket to $821.25/year since the user would have to deal with 5x the spam. Of course, this may be high since there will probably a larger percentage of those 5-seconds-to-delete messages.
About the only interesting thing about these BS numbers is that the lower one ($184) is close to Ferris Research's estimate of $168/year; while the higher number ($821) is close to Nucleus Research's $874/year.
Damn....that's a good idea. There's been a few times where my wife and I have been watching something and we both can't figure out what someone just said. I trying skipping back with the Replay and still can't make it out. Next time I'll try turning on CC. lol
I don't see anything in the article that mentions the 30 second quick skip. They are talking about Commercial Advance (which jumps over commercials automatically when it detects them) and the send show feature.
I'm sure many RTV owners will agree with me that having the ability to send shows over the internet is pretty pointless (it can take hours to days to send one). That feature is probably rarely used my most. However, in the process, D&M better not remove the ability to stream shows from other RTVs on your LAN. That would sucks and break DVArchive (which is probably the biggest selling point of Replays).
Now Commercial Advance is a nice feature, but it only works on about 1/2 of the shows I watch, and on that half it doesn't work flawlessly. It wouldn't kill me to lose this features, but it is one of the reasons why I bought a Replay instead of Tivo.
Lastly there is the 30 second skip. There is no talk of removing this feature...and they better not. Being able to "shorten" show watching time is one of the big reasons for owning a PVR in the first place. This feature is a must!
And for those who don't know, you can do a X minute skip by typing the number of minutes on the keypad and pressing skip. 3 + Skip works well to skip over a lot of commercials.;)
I personally think this is a very important feature in cube design. I'm lucky now in that I have a 10x12 cube, with a U-shaped desk. I have my workstation on the part of the "U" facing the entrance, others have it facing the back wall of the cube.
It's nice to be able to greet people when they walk up to you. I hate having to "sneak" up on someone who's back is facing you, and then interrupt them with an "Excuse Me!"
Unfortunately, it's hard to configure most cubes like this due to size constraints....now if I could only do something about the loud-talker next to me.
My father's family ran a small printing business in Twin Lakes, WI (not too far from Lake Geneva). Back in the early 70's, these two guys from a small company came to my dad needing booklets to be printed for a new game. He and his brother decided against taking the risk of doing this large job and turned them down. It turns out that they were from TSR, trying to get D&D printed. Doh!
Now think about this. Only 2% are valid queries, while 98% are invalid. Let's say the average client (or recursive DNS server) makes 1 invalid or malformed query for every 1000 queries which seems like a good number. That means that only 0.002% of valid queries have to go through the root DNS servers. The seems like a great design if you ask me.
We finally put a spam filter in place last week (good timing, eh?).
Just over 30% of our incoming email is currently being tagged as spam with well under 0.1% being false positives.
One interesting thing I did find was that nearly 60% of the spam is directed towards only 10% of our users. These are probably the people who are using their work accounts for non-work related purposes.
Rackspace is horrible! I will no longer recommend them to people as they do nothing to stop the number of spammers they host.
Over 10% of the spam received by my company is from someone hosted at Rackspace. I have sent numerous messages to abuse@rackspace.com over the past month with no response.
It appears, I'm not the only one suffering from this (search the net-abuse ng's).
Do not support spammers. Do not support Rackspace!
We lost about 50% of the drives in our IBM 300GL PCs within a year and a half of purchasing them.
The drives were 10GB Quantums and the Phillips controller on them was frying. IBM's response to this was horrible. They knew the drives were bad, and were going to fail, yet they would only replace them after failing. Our account rep was no help what so ever in resolving this (she basically disappeared after selling us several hundred thousand dollars in equipment).
Eventualy, after some threats from our purchasing department, IBM sent up replacement drives for the rest of the PCs. I don't even want to think about how much time and money this whole ordeal costed us. Hell, we don't even blame IBM for the bad drives...how could they have known. However, we do blame them for the poor service we received in resolving the issue.
Becuase of this, we stopped buying any IBM equipment. All servers and workstations are now coming from different companies. Had IBM's support just treated us with a little respect back when this issue started, they would still have a (rather large) customer.
The acceptable-use policy states that the network is there only for educational purposes. Yes P2P *might* have some educational purposes, but probably not. Also, the university has receive over 150 notices of copyright violations this year. I'm highly suprised the legal department has stepped in to start shutting things down.
I work at a large university in an IT department (but no the one that manages the network).
Over 70% of our traffic is from Kazaa an other P2P clients. The university has had to triple it's available bandwith on the campus uplink costing mucho dinero. They have tried very hard to allow all traffic without restricting certain protocols or ports, but this year it was just too much. If I'm not mistaken, they are now rate limiting access to the dorms.
Why not use taffic shaping? Becuase we are currently handling over a gigabit traffic for most of the day...buying hardware to support that much traffic would be quite expensive. Second, P2P applications are a moving target, so new applications or savy users will be able to bypass the filters quite easily.
So now, the students will have to deal with slow access in the dorms if they are going to keep using P2P applications. At least the rest of us, with legitimate internet uses, have bandwidth available.
Slashdot Mirror
1992? In an early post, they found the code in a BSD newsgroup from 1984.
News Posting
Here what one of the searches comes up with.
They're not talking about crashing the OS. They are talking about applications running on the OS crashing.
Think about it...how is Dr. Watson going to catch a crash if the OS crashed too?
Wow, a post on Slashdot giving Linux help that isn't sarcastic or condescending. I applaud you.
Hrmm....did this test include the usability of help from peers?
I'm pulling these numbers partly from my experience and partly from my ass.
Average user at work receives ~ 10 spam messages per day, with our spam filters catching 80%, so the user receives 2 per day.
I figure you can classify spam messages into 3 types which require varying amount of effort to determine when to delete them.
First, you have obvious spam that can be deleted by reading the subject. I figure that 25% that get though spam filters are this type. Let's say they require 5 seconds to read/delete.
Second type requires the user opening the message and reading the contents. Let's say 70% are this type and require 15 seconds.
The last type are the ones that confuse users or they think they are legit. These are the messages that users will reply to, talk to their coworkers about, or forward to IS to be verified. This would be the remaining 5% and require 5 minutes (300 seconds) of time.
With those numbers, the users spend 27 seconds per message or 54 seconds per day.
Per year, that is 5.475 hours. If the average cost of a user (pay + benefits) is $30/hour then the annual cost for spam per user is $164.25/year or $0.225/spam.
Now, we are running spam software in this situation. Figure that the spam software cost $10/user/year in licensing, and an additional $10/user/year in hardware or administrative costs. I'm also assuming that the spam load on our mail servers is minimal enough that the costs involved there are insignificant.
That puts the total now at $184.25/year. However, without antispam software, that total would sky rocket to $821.25/year since the user would have to deal with 5x the spam. Of course, this may be high since there will probably a larger percentage of those 5-seconds-to-delete messages.
About the only interesting thing about these BS numbers is that the lower one ($184) is close to Ferris Research's estimate of $168/year; while the higher number ($821) is close to Nucleus Research's $874/year.
Why? Because they can't act as a bridge AND an access point at the same time.
There are very few products that do this, and Apple's is one of them.
Maybe not the DMCA, but possibly the lesser known AMCA.
Umm...that's why you leave your digital cable box on all of the time. I've never had a problem with my RTV and digital cable.
Damn....that's a good idea. There's been a few times where my wife and I have been watching something and we both can't figure out what someone just said. I trying skipping back with the Replay and still can't make it out. Next time I'll try turning on CC. lol
It's good to see others from the forums over here defending a great product. ;)
I don't see anything in the article that mentions the 30 second quick skip. They are talking about Commercial Advance (which jumps over commercials automatically when it detects them) and the send show feature.
;)
I'm sure many RTV owners will agree with me that having the ability to send shows over the internet is pretty pointless (it can take hours to days to send one). That feature is probably rarely used my most. However, in the process, D&M better not remove the ability to stream shows from other RTVs on your LAN. That would sucks and break DVArchive (which is probably the biggest selling point of Replays).
Now Commercial Advance is a nice feature, but it only works on about 1/2 of the shows I watch, and on that half it doesn't work flawlessly. It wouldn't kill me to lose this features, but it is one of the reasons why I bought a Replay instead of Tivo.
Lastly there is the 30 second skip. There is no talk of removing this feature...and they better not. Being able to "shorten" show watching time is one of the big reasons for owning a PVR in the first place. This feature is a must!
And for those who don't know, you can do a X minute skip by typing the number of minutes on the keypad and pressing skip. 3 + Skip works well to skip over a lot of commercials.
I personally think this is a very important feature in cube design. I'm lucky now in that I have a 10x12 cube, with a U-shaped desk. I have my workstation on the part of the "U" facing the entrance, others have it facing the back wall of the cube.
...now if I could only do something about the loud-talker next to me.
It's nice to be able to greet people when they walk up to you. I hate having to "sneak" up on someone who's back is facing you, and then interrupt them with an "Excuse Me!"
Unfortunately, it's hard to configure most cubes like this due to size constraints.
It's Linux. That's unpossible.
LOL...these are the type of people Windows admins have been putting up with for years, and now you *nix guys can start dealing with them.
"Hi, I was a desktop support tech, now I have been thrown into the job of managing our Windows network, how do I install that Active Directory thing?"
Windows has had the burnden of bad, inexperienced sysadmins for years, now Linux can share in the joy as it's more widely deployed.
Here's an interesting story....
My father's family ran a small printing business in Twin Lakes, WI (not too far from Lake Geneva). Back in the early 70's, these two guys from a small company came to my dad needing booklets to be printed for a new game. He and his brother decided against taking the risk of doing this large job and turned them down. It turns out that they were from TSR, trying to get D&D printed. Doh!
Now think about this. Only 2% are valid queries, while 98% are invalid. Let's say the average client (or recursive DNS server) makes 1 invalid or malformed query for every 1000 queries which seems like a good number. That means that only 0.002% of valid queries have to go through the root DNS servers. The seems like a great design if you ask me.
Hahaha....Thank you for one of the funnier posts today.
You may want to check out ASP.NET then. ;)
Why is this spam modded up?
This article is about Tim and GForge....and the interview makes it pretty obvious why you shouldn't rely on SourceForge, or VA for that matter.
...and it's becuase of guys like this that 80% of IT projects fail.
We finally put a spam filter in place last week (good timing, eh?).
Just over 30% of our incoming email is currently being tagged as spam with well under 0.1% being false positives.
One interesting thing I did find was that nearly 60% of the spam is directed towards only 10% of our users. These are probably the people who are using their work accounts for non-work related purposes.
Rackspace is horrible! I will no longer recommend them to people as they do nothing to stop the number of spammers they host.
c kspace.com.
Over 10% of the spam received by my company is from someone hosted at Rackspace. I have sent numerous messages to abuse@rackspace.com over the past month with no response.
It appears, I'm not the only one suffering from this (search the net-abuse ng's).
Do not support spammers. Do not support Rackspace!
Here are the current spammers at Rackspace: http://www.spamhaus.org/sbl/listings.lasso?isp=ra
We lost about 50% of the drives in our IBM 300GL PCs within a year and a half of purchasing them.
The drives were 10GB Quantums and the Phillips controller on them was frying. IBM's response to this was horrible. They knew the drives were bad, and were going to fail, yet they would only replace them after failing. Our account rep was no help what so ever in resolving this (she basically disappeared after selling us several hundred thousand dollars in equipment).
Eventualy, after some threats from our purchasing department, IBM sent up replacement drives for the rest of the PCs. I don't even want to think about how much time and money this whole ordeal costed us. Hell, we don't even blame IBM for the bad drives...how could they have known. However, we do blame them for the poor service we received in resolving the issue.
Becuase of this, we stopped buying any IBM equipment. All servers and workstations are now coming from different companies. Had IBM's support just treated us with a little respect back when this issue started, they would still have a (rather large) customer.
Also, to add this this point.
The acceptable-use policy states that the network is there only for educational purposes. Yes P2P *might* have some educational purposes, but probably not. Also, the university has receive over 150 notices of copyright violations this year. I'm highly suprised the legal department has stepped in to start shutting things down.
I work at a large university in an IT department (but no the one that manages the network).
Over 70% of our traffic is from Kazaa an other P2P clients. The university has had to triple it's available bandwith on the campus uplink costing mucho dinero. They have tried very hard to allow all traffic without restricting certain protocols or ports, but this year it was just too much. If I'm not mistaken, they are now rate limiting access to the dorms.
Why not use taffic shaping? Becuase we are currently handling over a gigabit traffic for most of the day...buying hardware to support that much traffic would be quite expensive. Second, P2P applications are a moving target, so new applications or savy users will be able to bypass the filters quite easily.
So now, the students will have to deal with slow access in the dorms if they are going to keep using P2P applications. At least the rest of us, with legitimate internet uses, have bandwidth available.