I work for a company that sets up LAN's in hotels. We frequently work with both the managers/owners of the hotel, and the local ISP. If the ISP tells us one thing, and the owner another thing, the BEST and FASTEST way to get it sorted out is a 3 way call. The ISP's hate that sort of thing, but shit gets done fast.
Thats true, however consider how big some of these companies are, and then consider if 2/3/4 of these big companies decided to ALL abandon flat-rate pricing at the same time. They wouldn't be losing cutomers becuase they would have nowhere to go. Given how are government/lobbiest system works, I bet they could even get away with it.
Re:Can AJAX finally bring us "push technology"
on
Ajax in Action
·
· Score: 1
Polling is ok for somethings, really not good for others.
What if you had a number of people in a chat room written in ajax. You would have to poll very frequently, every few seconds or so, for each person in that chat room. If your chat room gets any number of users, you going to be pushing a lot of traffic, most of which would be avoided if we could do server side interrupts.
What if you have a call center, and you have a thing on your page that is showing incoming calls, this is another case where you would have to poll very often, on the order of every second or two. If your call center has more than just a few people, once again, it adds up pretty quick, and once again, this could be avoided with sever side interrupts.
Sure we could write a non browser application to do this, but then we would have to screw around with supporting all sorts of different operating systems, and also it would completely eliminate the old "log in the call center software real quick and see how things are going and say hi from some random computer"
Re:Can AJAX finally bring us "push technology"
on
Ajax in Action
·
· Score: 1
I AGREE!
I have this book, and I am writing ajax for production use a work right now. If there was an easy way to set up server side interrupts, we could have our pages do much cooler things, and it would actually probably be much easier to write.
Basically what this means is that along with doing async requests, ajax would have to open a socket, either a plain socket, or sslv2/sslv3 or whatever, and then the client side scripts on the server could write things to this socket.
The tricky part would be how the flow control happens on the client side. With normal sockets usually the both the server AND the client fork/make another thread. So on the server side, one thread would listen for incoming requests, and one would write stuff to the socket. This is no problem for most server side scripting languages. On the client side, one would render pages, do javascript/dom stuff, etc, the other would listen for commands on that socket.
Right now there is no such thing as fork()/pthread_create() in ajax. Implementing this DIRECTLY would basically mean writing an entire new langague. They would have to hack up a simple small api for reading that socket, defining the end of a message, defining a function to call when we recieve a message, defining how that funtion call interacts with the flow of the rest of the javascript, etc.
In my experiance, adding something like this is NOT trivial. When you start changing the flow of execution of your script, lots and lots and lots of things break. Then again, I have not personally ripped appart any browsers lately.
I didn't know ANY javascript before 48 hours ago, right now I am writing ajax for production use. You dont need to use/understand XML to use ajax, and javascript is simple as fuck to use. I assume you already know html, so really, stop bitching, start reading.
We can barely get robots that we can fully control to do anything more than a shuffle, the only running robot I have ever seen can only do so on a flat surface with no obsticles and no wind. I sure as hell wouldn't want any balance "help" from a computer.
My Freespire installation seemed to go smoothly until I was presented with a login and password scenario. I thought maybe I had downloaded the wrong thing. As determined as I was, I went online to checkout what to do, and was taught I was supposed to enter "root" and login and something else as password (don't remember anymore). To be honest, I don't remember what happened after that, but it was enough to make me give up.
Have you ever installed windows?
This is not an issue with linux being too hard. This is an issue with you not understanding how to install an operating system. Most(if not all) linux distro's ask for a root password.
Lets see what windows does.
Quite frankly, the one thing we're up to arses in is apps programers, and, ironically, the one thing in the computer field we're desperately short of right now is computer scientists.
I would like to say that I agree. I am currently taking a four year CS program, and I am really tired of programming. Personally I dont find my programming assignments difficult at all, and therefor do not find them interesting.
However, I would like to say that CS programs have more than one other route to choose. As you pointed out they can focus more on the theory side, and graduate more researchers in the field. Another option which is almost never considered, is to teach practical things either in network administration, or more detailed information about particular applications that are widely used.
I have no intention of being a code monkey for all my life, I personally would like to get into network administration, but have a real CS background. Personally I think it is kind of sad that my fellow students wouldn't know what a web server, or a mail server, or a router was if it bit them in the ass(on any operating system even). Also if your wondering, yes I do live for CS.
Some more information for you.. this is an blurb from the iptables man page
----------------
owner This module attempts to match various characteristics of the packet creator, for locally-generated packets. It is only valid in the OUTPUT chain, and even this some packets (such as ICMP ping responses) may have no owner, and hence never match.
--uid-owner userid
Matches if the packet was created by a process with the given effective user id. --gid-owner groupid
Matches if the packet was created by a process with the given effective group id. --pid-owner processid
Matches if the packet was created by a process with the given process id. --sid-owner sessionid
Matches if the packet was created by a process in the given session group.
------
You can filter network traffic based off of the same system that you can use to filter access to files. Even more fun is the ability to filter network traffic based off of a process id.
Partially true. If you take your strongest hash and just increase the number of bits of the result, assuming that someone can crack that hash, it will simply take longer to compute a collision. This would probably increase in a linear fashion.
Howoever, If you use more than one algrithm, it becomes harder to find a collision that fits both systems AND has the correct file size. This would probably increase in a exponential fashion(read: impossible).
"Most computer users today have at least a general understanding of how Windows works, this means that for a mid-sized business a dedicated IT position is not necessary."
As someone who works in technical support, I have to disagree with you. I talk to "managers" and "technical staff" on a regular basis who can not distinguish cat5 from a phone chord.
There is a lot more to a network than what operating system your clients use. There are cables, switches, power, servers, phones, security problems, applications, isps, routers, firewalls, backup systems, etc.
Yes, for a medium size buisness you will have to deal with all of these. If you have a network in your building, or buildings, that you need to work, you need someone who knows what the hell they are doing to take care of it. Usually it is WAY more expensive to have a network go down for a few hours to a couple days, than to hire some someone who isn't brain dead to take care of the system.
Extreme Option A) Everyone does everything at night, and sleeps during normal light hours. That means everyone has to have lighting turned on during there wake hours at all times.
Extreme Option B) Everyone does everything only at day, and goes to sleep when it gets dark out. This means hardly any lighting would be needed at all.
Which one uses less energy? Duh!? The question is about the actual amount of energy saved, and the practicality of doing so.
I was just thinking the same thing. Well... we used some sort of resource to create the electricty that is causing heat, why not use the heat to create some more electricity.
Certainly this wouldn't be something that you do at home. However for situations like huge data centers with racks that are putting off 50-90K BTUS, times however many racks there are, that adds up to an ass load of heat. Why not pipe all that heat off to a generator sitting on the roof?
"Bullshit when it applies to Linux. Having two highly competitive desktop platforms like GNOME and KDE results in both getting better faster."
So what if some company wants to write some application for linux, what do they choose?
If they stick with one then what if eventually there is a neccesity to use the other. What if they one they choose gets phased out? What if there are forks? What if another peice of software that would be useful to them only ran on the other?
If they choose both systems then that takes extra time to develop, extra maintance, leads to a larger footprint, etc.
You might say just use the widgets provided with XFree or Xorg, those will be always be available. Then the problem is your application doesn't look like anything else, users have to learn the look and feel and interface to your program.
"And it's nonsense to say that Linux programs don't usually use the same layout and menus"
If you really believe that, then have grandma try using a linux desktop system of your choice and note all the the gui related questions.
"Anybody switching from the Windows 2000 GUI to the XP GUI is going to have MAJOR problems with figuring out where everything is on the Start menu"
Like you said, many people simply switch back to classic view then everything is back in its old place. You answered your own criticism.
As far as competition and improvement goes between the two desktops, I would gladly wait an extra 4 milliseconds for stuff to open as a trade off for more serious linux applications.
"Anybody who says Windows is easier to use than Linux is simply wrong."
Well everyone has there own needs and preferences, its pretty hard to get scientific data on which is "easier". I use linux every day, I love linux, I love all the options it gives me. But we NEED a standard desktop if we want to get anywhere.
Slashdot Mirror
Im sure that works well for most file sharing trafic. However, how does it work against file sharing traffic over a proxy or a vpn?
You sir are full of shit.
1) If you count how many people who claim to work at best who deny/confirm the piracy of software, you are the wwaaay minority
2) Your account appears to be new, and your ONLY comments on slashdot have been in this thread.
Personally, I would guess your some sort of manager/PR person for bestbuy who is trying to save face in the IT world.
Actually I meant more like this
http://en.wikipedia.org/wiki/Declaration_of_war
DING DING DING DING DING. WINNER!
I work for a company that sets up LAN's in hotels. We frequently work with both the managers/owners of the hotel, and the local ISP. If the ISP tells us one thing, and the owner another thing, the BEST and FASTEST way to get it sorted out is a 3 way call. The ISP's hate that sort of thing, but shit gets done fast.
We are not at "war", in the same way that Korea was not a "war". It is a "military action". To be at "war" takes a vote from congress.
Nothing that was ever worth anything to anybody was ever ...
written without using the enter key?
The fact that the masses attend this stuff without shouting down the "radical
leaders" means they co-sign it.
That statement is a generalization and is absurd.
Thats true, however consider how big some of these companies are, and then consider if 2/3/4 of these big companies decided to ALL abandon flat-rate pricing at the same time. They wouldn't be losing cutomers becuase they would have nowhere to go. Given how are government/lobbiest system works, I bet they could even get away with it.
Polling is ok for somethings, really not good for others.
What if you had a number of people in a chat room written in ajax. You would have to poll very frequently, every few seconds or so, for each person in that chat room. If your chat room gets any number of users, you going to be pushing a lot of traffic, most of which would be avoided if we could do server side interrupts.
What if you have a call center, and you have a thing on your page that is showing incoming calls, this is another case where you would have to poll very often, on the order of every second or two. If your call center has more than just a few people, once again, it adds up pretty quick, and once again, this could be avoided with sever side interrupts.
Sure we could write a non browser application to do this, but then we would have to screw around with supporting all sorts of different operating systems, and also it would completely eliminate the old "log in the call center software real quick and see how things are going and say hi from some random computer"
I AGREE!
I have this book, and I am writing ajax for production use a work right now. If there was an easy way to set up server side interrupts, we could have our pages do much cooler things, and it would actually probably be much easier to write.
Basically what this means is that along with doing async requests, ajax would have to open a socket, either a plain socket, or sslv2/sslv3 or whatever, and then the client side scripts on the server could write things to this socket.
The tricky part would be how the flow control happens on the client side. With normal sockets usually the both the server AND the client fork/make another thread. So on the server side, one thread would listen for incoming requests, and one would write stuff to the socket. This is no problem for most server side scripting languages. On the client side, one would render pages, do javascript/dom stuff, etc, the other would listen for commands on that socket.
Right now there is no such thing as fork()/pthread_create() in ajax. Implementing this DIRECTLY would basically mean writing an entire new langague. They would have to hack up a simple small api for reading that socket, defining the end of a message, defining a function to call when we recieve a message, defining how that funtion call interacts with the flow of the rest of the javascript, etc.
In my experiance, adding something like this is NOT trivial. When you start changing the flow of execution of your script, lots and lots and lots of things break. Then again, I have not personally ripped appart any browsers lately.
I didn't know ANY javascript before 48 hours ago, right now I am writing ajax for production use. You dont need to use/understand XML to use ajax, and javascript is simple as fuck to use. I assume you already know html, so really, stop bitching, start reading.
http://www.w3schools.com/js/default.asp
http://en.wikipedia.org/wiki/AJAX
We can barely get robots that we can fully control to do anything more than a shuffle, the only running robot I have ever seen can only do so on a flat surface with no obsticles and no wind. I sure as hell wouldn't want any balance "help" from a computer.
This has been said so many times already. Do not call lists do NOT apply to buisness numbers.
My Freespire installation seemed to go smoothly until I was presented with a login and password scenario. I thought maybe I had downloaded the wrong thing. As determined as I was, I went online to checkout what to do, and was taught I was supposed to enter "root" and login and something else as password (don't remember anymore). To be honest, I don't remember what happened after that, but it was enough to make me give up.
Have you ever installed windows?
This is not an issue with linux being too hard. This is an issue with you not understanding how to install an operating system. Most(if not all) linux distro's ask for a root password. Lets see what windows does.
Quite frankly, the one thing we're up to arses in is apps programers, and, ironically, the one thing in the computer field we're desperately short of right now is computer scientists.
I would like to say that I agree. I am currently taking a four year CS program, and I am really tired of programming. Personally I dont find my programming assignments difficult at all, and therefor do not find them interesting.
However, I would like to say that CS programs have more than one other route to choose. As you pointed out they can focus more on the theory side, and graduate more researchers in the field. Another option which is almost never considered, is to teach practical things either in network administration, or more detailed information about particular applications that are widely used.
I have no intention of being a code monkey for all my life, I personally would like to get into network administration, but have a real CS background. Personally I think it is kind of sad that my fellow students wouldn't know what a web server, or a mail server, or a router was if it bit them in the ass(on any operating system even). Also if your wondering, yes I do live for CS.
I believe that ut2004 is completely multithreaded
Some more information for you.. this is an blurb from the iptables man page
----------------
owner
This module attempts to match various characteristics of the packet creator, for locally-generated packets. It is only valid in the OUTPUT chain, and even this some packets (such as ICMP ping responses) may have no owner, and hence never match.
--uid-owner userid
Matches if the packet was created by a process with the given effective user id.
--gid-owner groupid
Matches if the packet was created by a process with the given effective group id.
--pid-owner processid
Matches if the packet was created by a process with the given process id.
--sid-owner sessionid
Matches if the packet was created by a process in the given session group.
------
You can filter network traffic based off of the same system that you can use to filter access to files. Even more fun is the ability to filter network traffic based off of a process id.
Partially true. If you take your strongest hash and just increase the number of bits of the result, assuming that someone can crack that hash, it will simply take longer to compute a collision. This would probably increase in a linear fashion.
Howoever, If you use more than one algrithm, it becomes harder to find a collision that fits both systems AND has the correct file size. This would probably increase in a exponential fashion(read: impossible).
Well said!
Wow.
"Most computer users today have at least a general understanding of how Windows works, this means that for a mid-sized business a dedicated IT position is not necessary."
As someone who works in technical support, I have to disagree with you. I talk to "managers" and "technical staff" on a regular basis who can not distinguish cat5 from a phone chord.
There is a lot more to a network than what operating system your clients use. There are cables, switches, power, servers, phones, security problems, applications, isps, routers, firewalls, backup systems, etc.
Yes, for a medium size buisness you will have to deal with all of these. If you have a network in your building, or buildings, that you need to work, you need someone who knows what the hell they are doing to take care of it. Usually it is WAY more expensive to have a network go down for a few hours to a couple days, than to hire some someone who isn't brain dead to take care of the system.
Actually it is logical.
Extreme Option A) Everyone does everything at night, and sleeps during normal light hours. That means everyone has to have lighting turned on during there wake hours at all times.
Extreme Option B) Everyone does everything only at day, and goes to sleep when it gets dark out. This means hardly any lighting would be needed at all.
Which one uses less energy? Duh!? The question is about the actual amount of energy saved, and the practicality of doing so.
PLEASE MOD PARENT UP
I was just thinking the same thing. Well... we used some sort of resource to create the electricty that is causing heat, why not use the heat to create some more electricity.
Certainly this wouldn't be something that you do at home. However for situations like huge data centers with racks that are putting off 50-90K BTUS, times however many racks there are, that adds up to an ass load of heat. Why not pipe all that heat off to a generator sitting on the roof?
"Bullshit when it applies to Linux. Having two highly competitive desktop platforms like GNOME and KDE results in both getting better faster."
So what if some company wants to write some application for linux, what do they choose?
If they stick with one then what if eventually there is a neccesity to use the other. What if they one they choose gets phased out? What if there are forks? What if another peice of software that would be useful to them only ran on the other?
If they choose both systems then that takes extra time to develop, extra maintance, leads to a larger footprint, etc.
You might say just use the widgets provided with XFree or Xorg, those will be always be available. Then the problem is your application doesn't look like anything else, users have to learn the look and feel and interface to your program.
"And it's nonsense to say that Linux programs don't usually use the same layout and menus"
If you really believe that, then have grandma try using a linux desktop system of your choice and note all the the gui related questions.
"Anybody switching from the Windows 2000 GUI to the XP GUI is going to have MAJOR problems with figuring out where everything is on the Start menu"
Like you said, many people simply switch back to classic view then everything is back in its old place. You answered your own criticism.
As far as competition and improvement goes between the two desktops, I would gladly wait an extra 4 milliseconds for stuff to open as a trade off for more serious linux applications.
"Anybody who says Windows is easier to use than Linux is simply wrong."
Well everyone has there own needs and preferences, its pretty hard to get scientific data on which is "easier". I use linux every day, I love linux, I love all the options it gives me. But we NEED a standard desktop if we want to get anywhere.
I want to point out one thing. There are two ways for government to get our money.
Taxes and Fees.
Yes we have high taxes, but our fees are pretty low. For example it doesn't cost us 250 bucks to register a vehicle.