What you may need is loadbalancing and multiple servers. Granted, it's a function of how much mail on the whole you have to filter, some form of loadbalancing will be needed.
Round robin dns'ing, a load balancing machine, a firewall that can do the likes (bigIP, yuck, i hate them).
Your question is geared towards SMTP, but it's generally a network service question and how to handle X amount of traffic with Y resources.
It is productive though... to MS. It's the same reason any company would point out the great things about their product and the supposed downfalls of their own product.
Creates a foul taste for the competition.
Look at the switch campaign. It doesn't necessarily say PC's are bad, but that Mac's are easier. It doesn't mention the overhead of learning a new interface or other stuff.
MS is just more targeted for the bad rep they have. Nothign to see here. Move along.
I see a significant relation between this and metro cards. They are a piece of plastic, thinner than a credit card, but has a magnetic strip.
I have what's called a "weekly" metrocard. Unlimited trips for the entire week. The way to stop people from swiping others through, is by limiting the amount of time before you can use it in that spot. You can go to other spots w/o a problem serially. Just not the same one before 20 minutes are up.
The problem is, when this thing misreads/miswrites. It would give an error, to see the clerk and won't let me through. The clerk will usually find something like, "You just used this 4 minutes ago." It's up to the discression of the clerk to either let you through, computer error, or to wait 20 minutes.
What happens when this happens with these little devices? Neat in theory, but there's something that humans do better than any machines. Communicate and understand, in full duplex, a transaction.
It depends on the scale of the rewrite. Rewriting your OS would suck. Rewriting a single library and fixing things via refactoring methods.. it's not that bad.
Well, in terms of code level observations, yeah, it does what he says. In terms of applications, such as email clients, etc.. as he gave as an example, they do.
I wouldn't be surprised if you saw like, a 1% increase if you used straight tcp/ip.
tcp over tcp incures an x% size of packet increase. Reminds me of that stupid commerial. "We saved a nickel for a transaction" The go on and on about this stupid nickel and then they talk about how they do "20 million in transactions a day."
But hey, if you have the bandwidth, knock yourself out.:) If it works, it works.
I kinda liked fitday when i was trying to gain weight. It's also free. You don't get to export/import data, but as a temporal solution, it works.
As for tracking who you are, you can lie on most of the info. Worst of it possibly is, they'll track an anonymous person's diet. Is your intake that sacred?:)
And how many times did you not see the tiny, tiny cell phone in someone's hand while both are by the ear yammerin'.
Or better yet, the new headphones/mic's. When someone is not facing you directly.
Trust me. Once these things become popular, it will be a social norm to guess, he's talking to someone via a device, not into random space. I hate it myself.. fookin' confusin'.
Yes, 802.11 is a little more insecure due to one facet.
Take 2 computers, link them by ethernet cable, lock it up pretty well, and poof you have a mostly secure network.
Only thing stopping you from getting on my home network right now, is the fact you don't have a cable plugged into my switch at home. I also have a good firewall on my dsl line.
Now, if i were to put the switch on the sidewalk, anyone could just walk up, and jack in. They'd have access behind my firewall and to my dsl line. That is what wireless is like: putting an invisible switch whever you happen to be, within certain distance of an access point. So it's harder to secure by the fact that you don't need a wire to connect, but just be in proximity.. and unless you have shielding around your AP and computers that use the AP's, you are more open.
I'm gonna get flamed or called some random name for this.. but here's my guess.
Using the bios is prolly slower. If the cpu can access the card across the bus directly, I'd wager that it's faster than going to the bios, which doesn't need to be the fastest thing in the world to access directly.
Why isn't there a chip on the board that's fast and generic would be my next question, unless they use the bus in place of the bios and skip the whole "standards" idea.
I thought you were focusing on the client. My bad.
Point is, switching uses multiple towers while it figures itself out. If too many people are switching at the same time, either because a lot of people people step over the boarder of two towers at the same time, or thousands and THOUSANDS of peopel are flying over the two and rediculous speeds, the two towers will be overloaded.
I'm not talkiong about the client having problems, but the towers.
If a tower can take NNN amount of clients, but it takes X seconds for a tower to realize than it's been switched away. If everyone was on one tower and switch the next, during those X seconds of figuring out, hey me (the old tower) isn't being used anymore. Effectively, you are using two towers. (And in darkness bind them).
Anyway, if everyone moves around at speeds where you'd use multiple towers in the above senario, you run out of towers really quickly.
Clients who can connect will connect, but it's the connection saturation you are worried about.
1)The paranoia is NOT justified, look at the Sept 11th events, tons of people on cellphones on the planes with no problems. If an airplanes electronics are accepting super low power interference from ISM band devices they should be fixed because they will have real problems if they get too close to radar installations.
Maybe, maybe not. A flying plane will switch between towers very quickly. Can you speak up for how well it works at speeds of 200mph+? At walking speeds, you don't get a lot of switchovers.
ipsec has the advantages of encrypting entire lines with a good setup. ipsec has the disadvantage of having to be setup (exchanging keys, etc etc). you do excahnge keys via non-network like a good boy, right?
ssh has the advantage of having very little setup and is uber portable. problem is, you can't encrypt an entire line easily, like you might with stunnel, or the contender, ipsec.
But why contend? You are comparing two things that don't really compete directly. They come about it from different directions. One achieves security by encrypting telnet (and adding a bunch of features), while the other achieves security by encrypting the entire line.
Granted, using both might be overkill, but hey, it depends on how ipsec is used with the OS and when the decripted data is viewable. Anyone get tcpdump to work on the ipsec layer yet?
Slashdot Mirror
What you may need is loadbalancing and multiple servers. Granted, it's a function of how much mail on the whole you have to filter, some form of loadbalancing will be needed.
Round robin dns'ing, a load balancing machine, a firewall that can do the likes (bigIP, yuck, i hate them).
Your question is geared towards SMTP, but it's generally a network service question and how to handle X amount of traffic with Y resources.
I have. I suggest you do the same.
It is productive though... to MS. It's the same reason any company would point out the great things about their product and the supposed downfalls of their own product.
Creates a foul taste for the competition.
Look at the switch campaign. It doesn't necessarily say PC's are bad, but that Mac's are easier. It doesn't mention the overhead of learning a new interface or other stuff.
MS is just more targeted for the bad rep they have. Nothign to see here. Move along.
I see a significant relation between this and metro cards. They are a piece of plastic, thinner than a credit card, but has a magnetic strip.
I have what's called a "weekly" metrocard. Unlimited trips for the entire week. The way to stop people from swiping others through, is by limiting the amount of time before you can use it in that spot. You can go to other spots w/o a problem serially. Just not the same one before 20 minutes are up.
The problem is, when this thing misreads/miswrites. It would give an error, to see the clerk and won't let me through. The clerk will usually find something like, "You just used this 4 minutes ago." It's up to the discression of the clerk to either let you through, computer error, or to wait 20 minutes.
What happens when this happens with these little devices? Neat in theory, but there's something that humans do better than any machines. Communicate and understand, in full duplex, a transaction.
It depends on the scale of the rewrite. Rewriting your OS would suck. Rewriting a single library and fixing things via refactoring methods.. it's not that bad.
-s
Which doesn't work in some instances (for moz). There's a bug open for nytimes.com on it.
Can't you turn off popups by disabling javascript, or mucking through the source of moz and removing the javascript function for window.open?
Well, in terms of code level observations, yeah, it does what he says. In terms of applications, such as email clients, etc.. as he gave as an example, they do.
-s
GNOME or KDE on Redhat. Definitely. [/joke]
Yeah, I'll karma-burn for this one.
I was thinking more of Limbaugh. :)
I wouldn't be surprised if you saw like, a 1% increase if you used straight tcp/ip.
:) If it works, it works.
tcp over tcp incures an x% size of packet increase. Reminds me of that stupid commerial. "We saved a nickel for a transaction" The go on and on about this stupid nickel and then they talk about how they do "20 million in transactions a day."
But hey, if you have the bandwidth, knock yourself out.
http://www.fitday.com
:)
I kinda liked fitday when i was trying to gain weight. It's also free. You don't get to export/import data, but as a temporal solution, it works.
As for tracking who you are, you can lie on most of the info. Worst of it possibly is, they'll track an anonymous person's diet. Is your intake that sacred?
Btw, last time i checked was a year ago. YMMV.
scp and maybe rcp defies that for sure. it keeps the unix permissions. You maybe be thinking of the UMASK thing,but that can be ignored too.
You prolly mean oldest app or longest running app. :)
At least now we won't have to write the core things thrice or more, once for the pc, once for the pda and once for the cell phone...
Granted, UI's have to be different.
And how many times did you not see the tiny, tiny cell phone in someone's hand while both are by the ear yammerin'.
Or better yet, the new headphones/mic's. When someone is not facing you directly.
Trust me. Once these things become popular, it will be a social norm to guess, he's talking to someone via a device, not into random space. I hate it myself.. fookin' confusin'.
Yes, 802.11 is a little more insecure due to one facet.
Take 2 computers, link them by ethernet cable, lock it up pretty well, and poof you have a mostly secure network.
Only thing stopping you from getting on my home network right now, is the fact you don't have a cable plugged into my switch at home. I also have a good firewall on my dsl line.
Now, if i were to put the switch on the sidewalk, anyone could just walk up, and jack in. They'd have access behind my firewall and to my dsl line. That is what wireless is like: putting an invisible switch whever you happen to be, within certain distance of an access point. So it's harder to secure by the fact that you don't need a wire to connect, but just be in proximity.. and unless you have shielding around your AP and computers that use the AP's, you are more open.
What about road rage? :)
I'm gonna get flamed or called some random name for this.. but here's my guess.
Using the bios is prolly slower. If the cpu can access the card across the bus directly, I'd wager that it's faster than going to the bios, which doesn't need to be the fastest thing in the world to access directly.
Why isn't there a chip on the board that's fast and generic would be my next question, unless they use the bus in place of the bios and skip the whole "standards" idea.
I thought you were focusing on the client. My bad.
Point is, switching uses multiple towers while it figures itself out. If too many people are switching at the same time, either because a lot of people people step over the boarder of two towers at the same time, or thousands and THOUSANDS of peopel are flying over the two and rediculous speeds, the two towers will be overloaded.
I'm not talkiong about the client having problems, but the towers.
If a tower can take NNN amount of clients, but it takes X seconds for a tower to realize than it's been switched away. If everyone was on one tower and switch the next, during those X seconds of figuring out, hey me (the old tower) isn't being used anymore. Effectively, you are using two towers. (And in darkness bind them).
Anyway, if everyone moves around at speeds where you'd use multiple towers in the above senario, you run out of towers really quickly.
Clients who can connect will connect, but it's the connection saturation you are worried about.
Maybe, maybe not. A flying plane will switch between towers very quickly. Can you speak up for how well it works at speeds of 200mph+? At walking speeds, you don't get a lot of switchovers.
Just food for thought.
Analogous to opt-out, eh? :)
ipsec has the advantages of encrypting entire lines with a good setup. ipsec has the disadvantage of having to be setup (exchanging keys, etc etc). you do excahnge keys via non-network like a good boy, right?
ssh has the advantage of having very little setup and is uber portable. problem is, you can't encrypt an entire line easily, like you might with stunnel, or the contender, ipsec.
But why contend? You are comparing two things that don't really compete directly. They come about it from different directions. One achieves security by encrypting telnet (and adding a bunch of features), while the other achieves security by encrypting the entire line.
Granted, using both might be overkill, but hey, it depends on how ipsec is used with the OS and when the decripted data is viewable. Anyone get tcpdump to work on the ipsec layer yet?
Between this story, the Open Internet story, spammers sueing Journalists and what not...
I may have to start a betting pool.. and maybe get some popcorn as the hilarity ensues on, "Internet Deathmatch".