Wow, I had no idea that the economics of international trade were so simple. So all we need to do is match our employment taxes with that of competing nations, and we win? No, of course not. Economics is complicated. So complicated that even the most complex modeling routinely fails to accurately predict things. This is just the kind of overly reductionist pseudo-analysis that drives conversations into ideological corners that have little bearing on reality. Anyone peddling the 'it's this simple' explanation is a salesman. A snake-oil salesman, and a pretty poor one at that.
FTA: "driving one requires the ability to correctly predict what's going to happen far out ahead."...or you just say 'fuckit' and tailgate passenger cars at = 1 car-length, cut in front of passenger cars at the same distance, pass other trucks on a 4lane highway with a speed delta of ~1mph and probably wonder why most cars do whatever they can to get as far away from you as possible on the stretch of I-40 between Gallup and Albuquerque, NM.
It's not so much the USB port as the frame that surrounds it that will dictate durability. If the frame around the port properly stabilizes the plug so as to take all stress off of the actual contacts, then the USB-C ports should be quite durable. I've noticed issues with *some* micro-usb ports, and those are the ones that were very poorly stabilized by the frame of the devices (I'm looking at you, 2nd gen nexus7!).
Most likely, the mandate for the latest major versions of packages comes from STIG. The STIG is a guideline, and more often than not, requires an explanation as to why you have chosen not to comply. I've been able to get around the broken STIG assumption that only the latest major version of a package contains the required security fixes by writing up a vulnerability assessment that includes the patch-level of the package from the vendor (RedHat in my case), and the changelog of that patch that almost always includes the CVE for which the fix was provided.
The intent of the STIG issue is to close a security hole, *NOT* to force you to dangerously track the very latest major/minor version of a package. If you can show that you are closing security issues by suing the latest version from your vendor, then you are complying with the goal of the STIG.
I love all the spur-of-the-moment-statisticians come outta the woodwork when someone suggests being a little over-protective of their children. Yes, you and your 'n' of 1, with a statistical significance of absolute '0' will provide all the insight anyone will ever need. 'back when I was a kid... and I turned out just fine!' damn, it must be true!
Fact is, there are three things you need to balance here, not just the probability of something bad(TM) happening: 1. probability of bad thing happening 2. morbidity/mortality of said bad thing (a badness factor, say) 3. the cost of providing various levels of protection against said badness.
I can't prevent all possible bad outcomes from driving a car, but I'm willing to spend a certain amount of money to limit the badness of driving myself and my kids in my car. I pay for a safe car, snow tires, car seats, etc. and I make sure we always wear our seatbelts. Also, I don't drive like an asshole.
oh, gawd here we go again...
1. bacteria coat just about every surface of the planet. washing your hands with anti-bacterial soap to help reduce the transmission of disesase will *NOT* ever change that fact... ever.
2. The levels of triclosan in anti-bacterial soap is so low as to have a minimal affect. Keep in mind that most bacteria are killed and/or washed off the skin with proper soap+water hand-washing.
3. ignorant, knee-jerk, extremist ranting (ie. anti-bacterial soap will end our exposure to bacteria and thus ruin our immune systems) has a far greater potential to adversely affect the public health than anti-bacterial soap. Everyone knows everything on the internet is true. Just look at the anti-vaccine sheep...
you rant about the public not caring about science, and prove your own lack of this very care in the process. nice one.
The DSL to your house rides copper only to the nearest DSLAM, which might be in the vault just a few blocks away. From there it rides to the CO via fiber (most likely) where it joins the rest of the ATM network. From there it can ride (via ATM) to any ISP in the world, but typically only to ISPs in your state. The ATM links that your cheap-ass consumer DSL rides are highly controlled for QOS and you can guarantee that you are on the lowest tier of service. Your ISP brings their DSL virtual circuits in on an ATM line that is probably also smaller than the total aggregate of DSL virtual circuits that ride in on it. So, yes, your DSL can be 'over subscribed' as it is mostly a logical circuit and not a physical one. You get what you paid for, most of the time, and not more.
This is a mid-sized ISP in Salt Lake City, UT with great upstream connections, a solid IT staff, and promiscuous use of FOSS (mostly debian) on their systems. They encourage the use of linux, servers, and hosting. The prices are great, and the base level bandwidth (GB/mo) is plenty for most people. If you should want to turn into a hog, that's fine, they just charge you more for it (GB/mo). The best is that they do not count any bandwidth used at night or weekends, so if you wanna be a hog and not pay for it, you just limit your download times and yer all set. oh, and *everyone* gets a static IP. I pay 2bucks/mo for their secondary DNS service.
I pay a bit more for my DSL service than I would with qwest's ISP (who do they use now anyway?), and am HAPPY to do so, even if the price went up.
your right that it's not the parity calculations soaking CPU time that cost you, however, it's not the read latencies that get you. You calculate parity on write, from data you hold in memory. What kills you is every block written to the logical drive must first be written to memory, parity calculated, and only then written to disk. It isn't the CPU usage that is the problem here, but the latency you incur from those (albeit fast) calculations and memory operations. TO contrast that, RAID-1 requires only that each block be written to 2 physical drives instead of just one. The only added latency here is that you must wait for both drives to complete the write and you will always wait for the slower one.
I don't think you should allow anything but the actual story dictate how you post. It shouldn't matter at all who submitted and what URL they leave. If you like the story, post it. If people can't handle that, well, they've got some growing up todo. There are lots of bright, young people on/., but there are certain things that you only acquire with age. I would let moderators mod-down any OT discussions, especially if they get out of hand. I think the bottom line with/. is that the people decide what to talk about. I can scan through a couple hundred posts in no time... anyway.
Slashdot Mirror
Wow, I had no idea that the economics of international trade were so simple. So all we need to do is match our employment taxes with that of competing nations, and we win? No, of course not. Economics is complicated. So complicated that even the most complex modeling routinely fails to accurately predict things. This is just the kind of overly reductionist pseudo-analysis that drives conversations into ideological corners that have little bearing on reality. Anyone peddling the 'it's this simple' explanation is a salesman. A snake-oil salesman, and a pretty poor one at that.
FTA: "driving one requires the ability to correctly predict what's going to happen far out ahead." ...or you just say 'fuckit' and tailgate passenger cars at = 1 car-length, cut in front of passenger cars at the same distance, pass other trucks on a 4lane highway with a speed delta of ~1mph and probably wonder why most cars do whatever they can to get as far away from you as possible on the stretch of I-40 between Gallup and Albuquerque, NM.
It's not so much the USB port as the frame that surrounds it that will dictate durability. If the frame around the port properly stabilizes the plug so as to take all stress off of the actual contacts, then the USB-C ports should be quite durable. I've noticed issues with *some* micro-usb ports, and those are the ones that were very poorly stabilized by the frame of the devices (I'm looking at you, 2nd gen nexus7!).
Most likely, the mandate for the latest major versions of packages comes from STIG. The STIG is a guideline, and more often than not, requires an explanation as to why you have chosen not to comply. I've been able to get around the broken STIG assumption that only the latest major version of a package contains the required security fixes by writing up a vulnerability assessment that includes the patch-level of the package from the vendor (RedHat in my case), and the changelog of that patch that almost always includes the CVE for which the fix was provided.
The intent of the STIG issue is to close a security hole, *NOT* to force you to dangerously track the very latest major/minor version of a package. If you can show that you are closing security issues by suing the latest version from your vendor, then you are complying with the goal of the STIG.
I love all the spur-of-the-moment-statisticians come outta the woodwork when someone suggests being a little over-protective of their children. Yes, you and your 'n' of 1, with a statistical significance of absolute '0' will provide all the insight anyone will ever need. 'back when I was a kid... and I turned out just fine!' damn, it must be true!
Fact is, there are three things you need to balance here, not just the probability of something bad(TM) happening:
1. probability of bad thing happening
2. morbidity/mortality of said bad thing (a badness factor, say)
3. the cost of providing various levels of protection against said badness.
I can't prevent all possible bad outcomes from driving a car, but I'm willing to spend a certain amount of money to limit the badness of driving myself and my kids in my car. I pay for a safe car, snow tires, car seats, etc. and I make sure we always wear our seatbelts. Also, I don't drive like an asshole.
oh, gawd here we go again... 1. bacteria coat just about every surface of the planet. washing your hands with anti-bacterial soap to help reduce the transmission of disesase will *NOT* ever change that fact... ever. 2. The levels of triclosan in anti-bacterial soap is so low as to have a minimal affect. Keep in mind that most bacteria are killed and/or washed off the skin with proper soap+water hand-washing. 3. ignorant, knee-jerk, extremist ranting (ie. anti-bacterial soap will end our exposure to bacteria and thus ruin our immune systems) has a far greater potential to adversely affect the public health than anti-bacterial soap. Everyone knows everything on the internet is true. Just look at the anti-vaccine sheep... you rant about the public not caring about science, and prove your own lack of this very care in the process. nice one.
The DSL to your house rides copper only to the nearest DSLAM, which might be in the vault just a few blocks away. From there it rides to the CO via fiber (most likely) where it joins the rest of the ATM network. From there it can ride (via ATM) to any ISP in the world, but typically only to ISPs in your state. The ATM links that your cheap-ass consumer DSL rides are highly controlled for QOS and you can guarantee that you are on the lowest tier of service. Your ISP brings their DSL virtual circuits in on an ATM line that is probably also smaller than the total aggregate of DSL virtual circuits that ride in on it. So, yes, your DSL can be 'over subscribed' as it is mostly a logical circuit and not a physical one. You get what you paid for, most of the time, and not more.
This is a mid-sized ISP in Salt Lake City, UT with great upstream connections, a solid IT staff, and promiscuous use of FOSS (mostly debian) on their systems. They encourage the use of linux, servers, and hosting. The prices are great, and the base level bandwidth (GB/mo) is plenty for most people. If you should want to turn into a hog, that's fine, they just charge you more for it (GB/mo). The best is that they do not count any bandwidth used at night or weekends, so if you wanna be a hog and not pay for it, you just limit your download times and yer all set. oh, and *everyone* gets a static IP. I pay 2bucks/mo for their secondary DNS service.
I pay a bit more for my DSL service than I would with qwest's ISP (who do they use now anyway?), and am HAPPY to do so, even if the price went up.
on the relative slowness of RAID-5:
your right that it's not the parity calculations soaking CPU time that cost you, however, it's not the read latencies that get you. You calculate parity on write, from data you hold in memory. What kills you is every block written to the logical drive must first be written to memory, parity calculated, and only then written to disk. It isn't the CPU usage that is the problem here, but the latency you incur from those (albeit fast) calculations and memory operations. TO contrast that, RAID-1 requires only that each block be written to 2 physical drives instead of just one. The only added latency here is that you must wait for both drives to complete the write and you will always wait for the slower one.
I don't think you should allow anything but the actual story dictate how you post. It shouldn't matter at all who submitted and what URL they leave. If you like the story, post it. If people can't handle that, well, they've got some growing up todo. There are lots of bright, young people on /., but there are certain things that you only acquire with age. I would let moderators mod-down any OT discussions, especially if they get out of hand. I think the bottom line with /. is that the people decide what to talk about. I can scan through a couple hundred posts in no time... anyway.