What would be very interesting is a study of the bottlenecks. If you are considering a system with certain characteristics (processor/processors, memory, storage/raid etc) and you have X dollars, where is that best spent?
By paravirtualizing production machines, we found the greatest perfomance killer are complex bridging setups with virtual NICs on heavy traffic. This affects not only physical cpu load, more relevant for customers is some kind of 'feeled' performance;) where network latency caused by dropped packets counts as well. Another bottleneck is of a more physical nature. We noticed, that threaded app's on SMP VM's are remarkably slower if the VM uses cores on different physical cpu's. If vCPU's are pinned to pCPU cores, benchmarks are not very different from physical machines with comparable cpu's. Taking the focus only on virtualization, an overhead produced by a hypervisor is hard to measure. Thus spoken, we're using Xen. And yes, this can't be compared against the article about iis... we don't have any colorful charts;)
You're damn right, but think twice: How much of ICANN's power is nominal and how much is factical?
As long as the ICANN isn't really independed from local law and subordinated to the UNO, nothing will happen.
ICANN is a teethless lion.
Well, this friday i've desinfected two of our (linux)servers which have been infiltrated by abusing vulnurable CRM Software (customers installations). It doesn't matter if you jail this software and put it behind firewalls; these days it also doesn't matter what kind of architecture your server hardware is. It's way enough having a simple webserver with scripting capabilities and one single hole in the web software. The toolbox of todays crackers (or should i name them botnet consultants?) is huge enough to have success with simple trial and error. If the machines refuses to run x86 binaries, there are plenty of perl and/or php scripts doing the same stuff. Today was really frustrating since i found 3 Megs of well-designed tools and good code on a formerly known secure machine. The quality of the tools leads me to the thought that a) crackers are well organized and b) paid for their work. Another frustrating part is the communication with different abuse helpdesks to track down this crap. Not to mention that all ended up in romania... Sorry for sarcasm, but do you have *ANY* laws?
Oh... this is not my day, even slashdot's captcha offers me "punisher"... i ask myself, why always me??
It's not a usuable behaviour of bank representatives to request someones pin, so you would refuse to answer if someone asks for it. On websites, and even in their email folders, most people are really confused about questions and very unsure how to act. They don't have some 'common behavior' in mind because something like the netiquette is ancient and ignored by privates as well as by companies.
So, if many of you argue this problem as an educational one, you're only partly right. It's also a problem of absent cultural and ethical values in net-based communication which makes it really difficult for end-users to decide how to rank a message.
We are a small ISP, i assume no one of our customers would accept blocking ports and or IP ranges. The only thing left to deal with is the intense use of abuse mails. As you said, it wouldn't take long to block routings to wide areas of the net, maybe this will (mis)lead to some RFC for additional BGP features... Net Providers can't be the endusers nanny's, a gross wouldn't accept it. Also, giving the enduser a taste of security will prevent them from using *any* protection by themselves.
However, the traffic is paid anyway... this is what really count's, for comcast as well as for any other ISP.
> Failing both, anonymous/pseudonymous coding & releasing might be our only refuge.
I don't believe that offense patent infringement can be a refuge. To keep the values of the OSDL, you need to publish and release according to given rights, otherwise no one is able to trust in it. Additionally, i assume the use of illegally released software can't be the target of the OSDL.
This is exactly the point, Stallman talks about. Some european patent activists sometimes did similiar by requesting patents and releasing the software after successful patent assignation as open source under GPL.
Stallmans statement of setting open source generally as prior art is a nice vision. But, this reflects his job as a visionist;)
Do you think elections are completely unrelated to software patents? I'm not involved, but I ask myself, who holds the license of the software used in your great election machines? If it's closed and patented, are you able to trust the patent holder?
You cannot set your car/house/whatever in correalation to software. Software always depends on shareable and basic mankind knowledge. You might implement some tricky algorithm, but as any math done in you code You're able to get it into small and handy slices, with all of them more or less known to the public. If software patents are commonly used (and successfully targeted by laywers), one could get the Idea why not to set patents over simple math algorithms or even get highschools as licensee. Common knowledge is a much bigger market than the software industry. The whole patent lobbyism will expense the cost of public knowledge in common as there are no borders between knowledge of math or even knowledge of history. Patents are futile.
Just my 2 cents.
Slashdot Mirror
What would be very interesting is a study of the bottlenecks. If you are considering a system with certain characteristics (processor/processors, memory, storage/raid etc) and you have X dollars, where is that best spent? By paravirtualizing production machines, we found the greatest perfomance killer are complex bridging setups with virtual NICs on heavy traffic. This affects not only physical cpu load, more relevant for customers is some kind of 'feeled' performance ;) where network latency caused by dropped packets counts as well. Another bottleneck is of a more physical nature. We noticed, that threaded app's on SMP VM's are remarkably slower if the VM uses cores on different physical cpu's. If vCPU's are pinned to pCPU cores, benchmarks are not very different from physical machines with comparable cpu's. Taking the focus only on virtualization, an overhead produced by a hypervisor is hard to measure. Thus spoken, we're using Xen. And yes, this can't be compared against the article about iis ... we don't have any colorful charts ;)
You're damn right, but think twice: How much of ICANN's power is nominal and how much is factical? As long as the ICANN isn't really independed from local law and subordinated to the UNO, nothing will happen. ICANN is a teethless lion.
Well, this friday i've desinfected two of our (linux)servers which have been infiltrated by abusing vulnurable CRM Software (customers installations). It doesn't matter if you jail this software and put it behind firewalls; these days it also doesn't matter what kind of architecture your server hardware is. It's way enough having a simple webserver with scripting capabilities and one single hole in the web software. The toolbox of todays crackers (or should i name them botnet consultants?) is huge enough to have success with simple trial and error. If the machines refuses to run x86 binaries, there are plenty of perl and/or php scripts doing the same stuff. Today was really frustrating since i found 3 Megs of well-designed tools and good code on a formerly known secure machine. The quality of the tools leads me to the thought that a) crackers are well organized and b) paid for their work. Another frustrating part is the communication with different abuse helpdesks to track down this crap. Not to mention that all ended up in romania... Sorry for sarcasm, but do you have *ANY* laws?
... i ask myself, why always me??
Oh... this is not my day, even slashdot's captcha offers me "punisher"
It's not a usuable behaviour of bank representatives to request someones pin, so you would refuse to answer if someone asks for it. On websites, and even in their email folders, most people are really confused about questions and very unsure how to act. They don't have some 'common behavior' in mind because something like the netiquette is ancient and ignored by privates as well as by companies.
So, if many of you argue this problem as an educational one, you're only partly right. It's also a problem of absent cultural and ethical values in net-based communication which makes it really difficult for end-users to decide how to rank a message.
I think they would make more money by using Google's AdWord program instead of sueing them ;). The original article noted 1.6 Billion PI's a month...
We are a small ISP, i assume no one of our customers would accept blocking ports and or IP ranges. The only thing left to deal with is the intense use of abuse mails. As you said, it wouldn't take long to block routings to wide areas of the net, maybe this will (mis)lead to some RFC for additional BGP features... Net Providers can't be the endusers nanny's, a gross wouldn't accept it. Also, giving the enduser a taste of security will prevent them from using *any* protection by themselves. However, the traffic is paid anyway... this is what really count's, for comcast as well as for any other ISP.
> Failing both, anonymous/pseudonymous coding & releasing might be our only refuge. I don't believe that offense patent infringement can be a refuge. To keep the values of the OSDL, you need to publish and release according to given rights, otherwise no one is able to trust in it. Additionally, i assume the use of illegally released software can't be the target of the OSDL. This is exactly the point, Stallman talks about. Some european patent activists sometimes did similiar by requesting patents and releasing the software after successful patent assignation as open source under GPL. Stallmans statement of setting open source generally as prior art is a nice vision. But, this reflects his job as a visionist ;)
Do you think elections are completely unrelated to software patents? I'm not involved, but I ask myself, who holds the license of the software used in your great election machines? If it's closed and patented, are you able to trust the patent holder?
You cannot set your car/house/whatever in correalation to software. Software always depends on shareable and basic mankind knowledge. You might implement some tricky algorithm, but as any math done in you code You're able to get it into small and handy slices, with all of them more or less known to the public. If software patents are commonly used (and successfully targeted by laywers), one could get the Idea why not to set patents over simple math algorithms or even get highschools as licensee. Common knowledge is a much bigger market than the software industry. The whole patent lobbyism will expense the cost of public knowledge in common as there are no borders between knowledge of math or even knowledge of history. Patents are futile. Just my 2 cents.