And where do you suppose we'll be "mining" antimatter? It's the same problem we have today with hydrogen fuel cells: where do we get the f'ing hydrogen? (answer: the extremely energy (and money) expensive process of hydrolysis. we get it from water, or methane)
The issue is not the server, per se, but the components that can only be run from that old version. I have a few of those still around (toshiba pbx management engine: you give it its own VM and never fuck with any part of it! Shut down the VM when it's not being used.)
if he built things right in the first place
WRONG. Obviously you aren't a programmer, nor do you know any. Functions get changed, renamed, deprecated, and removed. No matter how well you write your java craplet, changes in the JRE will eventually break it. I have a desktop full of various versions simply because apps can't work with newer versions. The same is true of perl, python, and php applications on Linux.
(Cisco knows this and that's why they bought Meraki)
So they could nickel-and-dime you in the cloud as well.
Meraki is ok. But their hardware is way too expensive, and the never ending cloud management fees can't be ignored. For an enterprise that can't afford a huge IT staff, the stuff is perfect, if costly.
There are, indeed, such "white box" switches out there. They aren't 100% open as Broadcom isn't about to release the SDK for their switch chips. (and having worked with the mess, you. don't. want. it.) And it's Broadcom's chips at the heart of almost every manufacturers switches. (even Cisco and HP)
Cisco tried for years to modularize IOS. It was a mess and never left the lab. Instead, they turned into an application running on a linux platform. The ASA... linux running a single application that is the entire firewall. (linux is easier to run on random x86 systems than porting your homegrown, inhouse OS) For IOS, we ended up with NX-OS (switches), and IOS-XE (routers) -- and IOS-XR [QNX based] on the bigiron routers. [ https://networkingnerd.net/tag... ]
Negative. This is simple nickel-and-dime price rouging. They sell you the same piece of hardware with a different model number on it, crippling it in software and licensing. For example, the ages old ASA 5510/20/40/50. They're all the same hardware. They get slightly faster, and better processors, and more DIMM slots, but it's the same software on the same motherboard. On the 10, the four gigabit nics are limited to 10/100; if you buy the expensive security plus license, then *two* of them can be set to gig speed. On the 20, with a 400MHz faster celeron, all four run at full speed. Per "user" limits are just as lame, and in fact, take a great deal of work to track.
It's marketed to the enterprise as a means of "in-place upgrading". You only need a firewall with 100 user, 10k connection, 100Mbps right now, but might 10x that in a few years. The bullshit is they're selling you hardware that's orders of magnitude faster and more capable at a profitable price with the base license. The $10k box is identical to the $1k box; the only difference is on paper. It's the same software on the same hardware, any difference in price is nonsense; you've already paid for the hardware capabilities, and software development.
Security updates/fixes can be had for free. It can be a bit of work to get them honor it, but every security advisory has a clause at the bottom to contact TAC for non-contract customers. (In my experience, it's faster to search the internet for it, than deal with out-of-contract interactions with TAC.)
Depends on what you make them do, and the version of code used to do it. There are some parts of IOS they simply don't test. (I've had QA engineers say so.) The biggest issue I've had -- and they've never fully fixed it -- is DHCP on a NAT outside interface. "interface nat" doesn't work. And then there's always some little memory leak or corruption issue somewhere that will crash the thing at some point. (even if it takes years.)
- Cisco 7401's: cache design fault causes random reboots. - PIX 501: poor choice of power connector, and poorer choice of internal power components led to the first few thousand having to be "fixed" -... bad caps (industry wide issue for a few years) -... bad RAM across the entire spectrum of products
And that's not counting the thousands of "one off" failures enterprises experience all the time. (that's why you buy Smartnet!) I've had the SRAM (packet memory) in a VIP "go bad". Shit breaks. Things have design flaws. Software ALWAYS has bugs. Cisco is huge. Their products are increasing complex, so issues are going to be common and "a big deal".
I keep mine disconnected because the idiots at Visio give you zero control over system updates. There's no way to check for an update. There's no way to say no to an update. You'll be watching TV and *poof* the TV reboots and takes for bloody ever to return to action.
Gravity isn't doing all the work. 99% of the arresting power is in the deep bed of gravel. Non-destructive? Have you ever driven a 20ton truck and trailer into one? Those traps can deal out damage. (bent suspension parts, flat tires, snapped couplings, etc.) Far less damage that would otherwise occur, for sure. And it sure as hell does imped traffic -- rubber necking f'ing fools. Plus, getting out of that pit is quite a spectacle.
They don't work for airports because these sorts of traps would damage the landing gear, and potentially the airframe. It would also close any runway with a plane in the trap, because now you have no safe area for the next plane, and it'll take a fleet of heavy machines to get damn plane out of the pit.
If you knew anything about "those other vehicles", you'd know cars and planes behave very different. Just watch a VW take a corner hard; one corner has its nose in the dirt and the opposite corner is wagging in the air. You can't do that with a plane; the 100+ft wings would be destroyed. Also, anyone who's had even half an hour of track education knows maximum braking occurs IN A STRAIGHT LINE, maximum acceleration occurs IN A STRAIGHT LINE. The more you turn, the less of either you can do.
Actually, they did the best they could with what they had, and what they were allowed to do. As they clearly said in the episode, putting a bomb in an actual flying plane at altitude is absolutely not allowed -- no one will fly it, and the FAA won't let it in US airspace. (and they don't have the budget to blow up a fully functional 747.)
In almost every documented case, the people blown out of the plane are either not strapped in properly, or their seat went with them. In every case I'm aware of where some nut does get a bomb on board, it doesn't rip the plane in half; it makes a small hole and the plane lands safely minus the bomber (who goes out their new hole) and maybe a nearby passenger or crew member.
Oh, it's been tried. It just doesn't f'ing work. The amount of explosives required, and their precise positioning, is not something anyone is likely to ever be able to carry out. (getting enough idiots to do it isn't the issue. getting the multiple bricks of C4 on the plane, into place, and detonated at the same time...)
Yes, airport backscatter xray machines can tell the difference between a laptop battery and a block of C4. Batteries have lots of metallic plates inside them. Youtube is full of videos of people taking them apart.
No, C4 does not look like a battery to their imaging technology. No, an iPad does not have the volume to house (or be) enough C4 to make much of a hole. (kill a few people, sure. bring down the plane, absolutely not.)
From history. This is far from a new idea. It used to be a lot easier as laptops were bigger and their batteries were bigger. Security screeners the world over have been trained to look for explosives in everything for many, MANY decades. The ability to do any serious damage, however, is still just as limited. (hint: blowing out a window won't destroy the plane or "suck anyone out".) You'd have much better luck climbing into the avionics bay and blowing up some of the computers. Sure, you'd kill a few people, etc. but the plane isn't going to fall out of the sky Lost style.
Plus, iGadgets are very tiny devices these days. How much boom-juice can you get in one? About 1% of what you'd need to kill a plane.
Those aren't errors in the GPS, but the data it's working with. If it doesn't know about a road, it won't tell you drive down it. If it doesn't know about time restrictions, or construction, or accidents, etc., etc. If we're talking about Waze, then I to have say that's not "GPS"; it's more of a "well, no one else is driving here, so go here!" system.
We validate what the GPS is telling us to do, but we don't ignore it's instructions and plan our own path. If one can't turn left, they pass the turn and wait for the GPS to figure things out. If you can't get in the correct lane in time, again, no panic, just keep driving until the GPS recalculates.
My favorite was the old Lexus GPS. It knows there are roads, and draws lines on the screen, but "has no data" on them so will not navigate over them. (and fills the screen with endless warnings so you can't see any of the little grey lines)
Nope. That's exactly the thing. We trust the computer and will blindly follow it over a cliff. The passenger, being human, is just as fallible as the driver, so we don't trust it. The computer will NEVER forget to tell you about a turn; the passenger most definitely can (and will.)
Exactly. They really had to research that the parts of the brain that handle navigation aren't active when we aren't navigating?!? That's part of the reason we use a GPS in the first place... so we don't have to think about it. (the key reason being, we don't have a clue where the hell we're going.)
By design, they're supposed to be unique. Manufacturers aren't supposed to "recycle" an OUI, but I've heard some lesser known Chinese companies have. The likelihood of having a collision is nearly zero. Now, if you start "randomly" generating your own MACs, the probability of collisions goes way up. (30 years and counting, I've never seen two NICs with the same MAC -- well, that I hadn't messed with, or were broken (all 0's))
Assuming a built-in-address is unique is a safe bet. Assuming a made up one is unique is going to be a problem eventually.
Except for the whole using water towers BS. That'll be even more impossible to get past locals than converting a large portion of a mountain top into a lake. Hell, they had a fit over a telescope that would've used a few acres; such a thing will need a few HUNDRED acres.
Slashdot Mirror
And where do you suppose we'll be "mining" antimatter? It's the same problem we have today with hydrogen fuel cells: where do we get the f'ing hydrogen? (answer: the extremely energy (and money) expensive process of hydrolysis. we get it from water, or methane)
The issue is not the server, per se, but the components that can only be run from that old version. I have a few of those still around (toshiba pbx management engine: you give it its own VM and never fuck with any part of it! Shut down the VM when it's not being used.)
WRONG. Obviously you aren't a programmer, nor do you know any. Functions get changed, renamed, deprecated, and removed. No matter how well you write your java craplet, changes in the JRE will eventually break it. I have a desktop full of various versions simply because apps can't work with newer versions. The same is true of perl, python, and php applications on Linux.
So they could nickel-and-dime you in the cloud as well.
Meraki is ok. But their hardware is way too expensive, and the never ending cloud management fees can't be ignored. For an enterprise that can't afford a huge IT staff, the stuff is perfect, if costly.
There are, indeed, such "white box" switches out there. They aren't 100% open as Broadcom isn't about to release the SDK for their switch chips. (and having worked with the mess, you. don't. want. it.) And it's Broadcom's chips at the heart of almost every manufacturers switches. (even Cisco and HP)
Cisco tried for years to modularize IOS. It was a mess and never left the lab. Instead, they turned into an application running on a linux platform. The ASA... linux running a single application that is the entire firewall. (linux is easier to run on random x86 systems than porting your homegrown, inhouse OS) For IOS, we ended up with NX-OS (switches), and IOS-XE (routers) -- and IOS-XR [QNX based] on the bigiron routers. [ https://networkingnerd.net/tag... ]
Negative. This is simple nickel-and-dime price rouging. They sell you the same piece of hardware with a different model number on it, crippling it in software and licensing. For example, the ages old ASA 5510/20/40/50. They're all the same hardware. They get slightly faster, and better processors, and more DIMM slots, but it's the same software on the same motherboard. On the 10, the four gigabit nics are limited to 10/100; if you buy the expensive security plus license, then *two* of them can be set to gig speed. On the 20, with a 400MHz faster celeron, all four run at full speed. Per "user" limits are just as lame, and in fact, take a great deal of work to track.
It's marketed to the enterprise as a means of "in-place upgrading". You only need a firewall with 100 user, 10k connection, 100Mbps right now, but might 10x that in a few years. The bullshit is they're selling you hardware that's orders of magnitude faster and more capable at a profitable price with the base license. The $10k box is identical to the $1k box; the only difference is on paper. It's the same software on the same hardware, any difference in price is nonsense; you've already paid for the hardware capabilities, and software development.
Security updates/fixes can be had for free. It can be a bit of work to get them honor it, but every security advisory has a clause at the bottom to contact TAC for non-contract customers. (In my experience, it's faster to search the internet for it, than deal with out-of-contract interactions with TAC.)
Depends on what you make them do, and the version of code used to do it. There are some parts of IOS they simply don't test. (I've had QA engineers say so.) The biggest issue I've had -- and they've never fully fixed it -- is DHCP on a NAT outside interface. "interface nat" doesn't work. And then there's always some little memory leak or corruption issue somewhere that will crash the thing at some point. (even if it takes years.)
- Cisco 7401's: cache design fault causes random reboots. ... bad caps (industry wide issue for a few years) ... bad RAM across the entire spectrum of products
- PIX 501: poor choice of power connector, and poorer choice of internal power components led to the first few thousand having to be "fixed"
-
-
And that's not counting the thousands of "one off" failures enterprises experience all the time. (that's why you buy Smartnet!) I've had the SRAM (packet memory) in a VIP "go bad". Shit breaks. Things have design flaws. Software ALWAYS has bugs. Cisco is huge. Their products are increasing complex, so issues are going to be common and "a big deal".
I keep mine disconnected because the idiots at Visio give you zero control over system updates. There's no way to check for an update. There's no way to say no to an update. You'll be watching TV and *poof* the TV reboots and takes for bloody ever to return to action.
Gravity isn't doing all the work. 99% of the arresting power is in the deep bed of gravel. Non-destructive? Have you ever driven a 20ton truck and trailer into one? Those traps can deal out damage. (bent suspension parts, flat tires, snapped couplings, etc.) Far less damage that would otherwise occur, for sure. And it sure as hell does imped traffic -- rubber necking f'ing fools. Plus, getting out of that pit is quite a spectacle.
They don't work for airports because these sorts of traps would damage the landing gear, and potentially the airframe. It would also close any runway with a plane in the trap, because now you have no safe area for the next plane, and it'll take a fleet of heavy machines to get damn plane out of the pit.
If you knew anything about "those other vehicles", you'd know cars and planes behave very different. Just watch a VW take a corner hard; one corner has its nose in the dirt and the opposite corner is wagging in the air. You can't do that with a plane; the 100+ft wings would be destroyed. Also, anyone who's had even half an hour of track education knows maximum braking occurs IN A STRAIGHT LINE, maximum acceleration occurs IN A STRAIGHT LINE. The more you turn, the less of either you can do.
Actually, they did the best they could with what they had, and what they were allowed to do. As they clearly said in the episode, putting a bomb in an actual flying plane at altitude is absolutely not allowed -- no one will fly it, and the FAA won't let it in US airspace. (and they don't have the budget to blow up a fully functional 747.)
In almost every documented case, the people blown out of the plane are either not strapped in properly, or their seat went with them. In every case I'm aware of where some nut does get a bomb on board, it doesn't rip the plane in half; it makes a small hole and the plane lands safely minus the bomber (who goes out their new hole) and maybe a nearby passenger or crew member.
Oh, it's been tried. It just doesn't f'ing work. The amount of explosives required, and their precise positioning, is not something anyone is likely to ever be able to carry out. (getting enough idiots to do it isn't the issue. getting the multiple bricks of C4 on the plane, into place, and detonated at the same time...)
Yes, airport backscatter xray machines can tell the difference between a laptop battery and a block of C4. Batteries have lots of metallic plates inside them. Youtube is full of videos of people taking them apart.
No, C4 does not look like a battery to their imaging technology. No, an iPad does not have the volume to house (or be) enough C4 to make much of a hole. (kill a few people, sure. bring down the plane, absolutely not.)
From history. This is far from a new idea. It used to be a lot easier as laptops were bigger and their batteries were bigger. Security screeners the world over have been trained to look for explosives in everything for many, MANY decades. The ability to do any serious damage, however, is still just as limited. (hint: blowing out a window won't destroy the plane or "suck anyone out".) You'd have much better luck climbing into the avionics bay and blowing up some of the computers. Sure, you'd kill a few people, etc. but the plane isn't going to fall out of the sky Lost style.
Plus, iGadgets are very tiny devices these days. How much boom-juice can you get in one? About 1% of what you'd need to kill a plane.
Those aren't errors in the GPS, but the data it's working with. If it doesn't know about a road, it won't tell you drive down it. If it doesn't know about time restrictions, or construction, or accidents, etc., etc. If we're talking about Waze, then I to have say that's not "GPS"; it's more of a "well, no one else is driving here, so go here!" system.
We validate what the GPS is telling us to do, but we don't ignore it's instructions and plan our own path. If one can't turn left, they pass the turn and wait for the GPS to figure things out. If you can't get in the correct lane in time, again, no panic, just keep driving until the GPS recalculates.
My favorite was the old Lexus GPS. It knows there are roads, and draws lines on the screen, but "has no data" on them so will not navigate over them. (and fills the screen with endless warnings so you can't see any of the little grey lines)
Nope. That's exactly the thing. We trust the computer and will blindly follow it over a cliff. The passenger, being human, is just as fallible as the driver, so we don't trust it. The computer will NEVER forget to tell you about a turn; the passenger most definitely can (and will.)
Exactly. They really had to research that the parts of the brain that handle navigation aren't active when we aren't navigating?!? That's part of the reason we use a GPS in the first place... so we don't have to think about it. (the key reason being, we don't have a clue where the hell we're going.)
By their records, it's "available" to everyone.
Verizon Business (VZB) and Verizon FiOS (FiOS) are different businesses. VZB's fiber does not connect to FiOS customers.
By design, they're supposed to be unique. Manufacturers aren't supposed to "recycle" an OUI, but I've heard some lesser known Chinese companies have. The likelihood of having a collision is nearly zero. Now, if you start "randomly" generating your own MACs, the probability of collisions goes way up. (30 years and counting, I've never seen two NICs with the same MAC -- well, that I hadn't messed with, or were broken (all 0's))
Assuming a built-in-address is unique is a safe bet. Assuming a made up one is unique is going to be a problem eventually.
On a side note, with that much surface area, you'll be losing A LOT of water to evaporation! (and conversely collecting a lot from rain)
Except for the whole using water towers BS. That'll be even more impossible to get past locals than converting a large portion of a mountain top into a lake. Hell, they had a fit over a telescope that would've used a few acres; such a thing will need a few HUNDRED acres.
Give me Elon Musk's billions and I'll give you much better solutions.