But in practice, there is no meaningful difference between "QoS" and "Net Neutrality" as there is no meaningful definition of what constitutes a "type" of traffic.
For example, NetFlix uses a traffic based on SilverLight's streaming video capability. Comcast only has to change this ever so slightly so it's a different "type" and then they can claim QoS. There is no meaningful limit to the rules that can be used to determine QoS, thus there is no meaningful distinction between QoS and "Net Neutrality".
In case you hadn't noticed, Credit Card companies secure your credit card using techniques very similar to A/V vendors' products. They do heuristic scanning of transactions, looking for consumer spending patterns and throwing red flags when they change significantly. You can wax poetic all you want about "smart cards" but the system is big enough that we'll probably *never* be without similar methods for protecting your bank account
In my laptop, I have an SSD. Upgrading the HDD cost about as much as a new laptop and cost significantly less. I've been able to buy 2+ years of time on my old laptop with an upgrade at significantly less cost.
So the numbers make sense, here!
We host a heavily database-driven app. Use of an SSD reduces latency by at *least* 95% in our testing. It's a no-brainer. Even if we replaced the SSDs every single year, we'd still come out way ahead. SSDs are where it's at for perfromance!
I don't understand why a spine is necessary. If the market is crying out for XP, why not just simply migrate XP to an annual license of $20 and let people keep their beloved 512 MB P IVs going for as long as their bits keep shuffling?
Microsoft had an awesome opportunity with WinXP and they just threw it away...
I would argue, it's not that he's *right*, it's that he's right in a way that still doesn't help.
Wife: "Honey, do you know where my keys are?"
Husband: "Pretty sure they're right where you left them!".
Technically correct, completely unhelpful.
And that's how QMail is. (or was, I stopped using it years ago) Qmail is a nightmare I'd rather forget. Sure, it is/was very secure, modularly written with strong privilege separation, built-in clustering support, etc. But because of the horribly restrictive license it was under, you had to download it and apply a half dozen patches in order to get it to do what you wanted. Worse, the patches were often somewhat in conflict, so you end up doing patches in a specific order to get what you want, and/or manually editing files. And then, when you are done, you have something that not only has its own init system, it actually conflicts with the standard init system so you have to pretty much disable every other service on the said server.
But you know what seems like the most asshole part of all? Updates are infrequent at best, and the license doesn't allow you to distribute updated versions. It just isn't going to get any better. Everywhere else, you run a single command (EG: yum install postfix) but to get qmail going takes weeks.
I'm a big fan of DJB's code quality, and the new Crypto being released as LGPL means I would actually get behind using his code. I'm just glad that heartbleed means that the critical security infrastructure will finally get somewhere near the attention it deserves...
Need to do a semi-repetitive task that's not quite annoying enough to script on a bunch of servers? TMUX to the rescue! You can open 10 windows to 10 different servers, and synchronize what you type so it shows up on all screens, or click on an individual screen to run just that command there.
but the pure electric car isn't going to be ready until a) massive updates to the power grid b) swappable batteries c) battery tech that lets cars go 500-1000 miles on a charge.
Why the boolean logic?
In case you hadn't noticed, pure electric cars are stomping the ever loving crap out of the luxury/performance car market. So long as the cars are selling at a growing pace, they are here to stay and are ready for the people who continue to buy them.
And as long as this happens, manufacturers will make continuing improvements to the cars they make.
A) The power grid is constantly being worked on. As people buy more cars, the grid will be upgraded to match demand.
B) Swappable batteries might be one of those improvements. But they don't seem to be required, at least not yet.
C) 1000 miles on a charge? Show me any common car that gets anything like that range.
Lots of people expect the world to change all of a sudden. But it doesn't really. Instead, continuing incremental changes gradually make the world into a different place. Those incremental changes have rather drastically changed how people interact in just the 30 or so years that I can personally remember.
Your points are valid, in a sense. But do you really think that people are going to stop trusting Open Source technologies? What isn't part of the conversation is just how terribly horrible OpenSSL actually is. It's a readability nightmare. The patch makes my eyes bleed, makes me weep gently to myself as I rock myself in an attempt to succor the horrific nightmare that code of this quality is what drives most Internet "security".
I so sorely wish more consideration was given towards NACL as a replacement for OpenSSL. It's clean, elegant, readable. Bugs will be shallower because readers might have *some idea* what is going on. And with an LGPL license, it should be quite embeddable.
IMHO, OpenSSL should be toss summarily as soon as possible. Beneath its horrific API and code lurk untold numbers of nascent, undiscovered holes no doubt already being exploited by our good friends at the NSA.
Writing security code is *hard*, folks. Making it hard to read only makes it impossible to debug...
^^^ This. I bought an officially branded Comcast MODEM on ebay for $21 after shipping. It paid for itself in just 3 months, and hasn't given me a lick of trouble in YEARS.
You also could have clicked the link I provided,l and perhaps typed the word "ethernet"... There are lots of models covering many different use cases. My unit works great over wifi, which is great since I have no Ethernet in my bedroom...
Google TV isn't a failure, it's just not the only success.
I have a Google TV stick and I love it! It is just a tablet that uses my TV as its screen and a wireless keyboard as its input. It is about the size of a thumb drive. It cost $40 on Amazon (.search for mk808b to get the exact model I'm watching Hulu+ on as I write this)
See my post history for details: this is quite successful. I have no idea what Google would want to improve...
Play store: no problem. How else do you think I installed Netflix, CBS.com, Hulu, uTorrent, and all the other apps?
Seriously, just imagine a tablet running on your TV using a mouse/remote instead of a touch screen. That's what I use every day. (And what is currently playing Sherlock Holmes a la Hulu; my wife loves that show)
Seriously: a major airplane "disappears" despite evidence that it wasn't really crashed. Everybody's wondering who dunnit and how, and whether or not it will become another impromptu bomb.
There's a *lot* you can carry on a 777. $50 mil is a lot, but the amount of damage such a plane could do with a little direction makes $50 mil look like peanuts. And it's pretty clear that anybody with the skills to make it disappear as completely as it did is capable of more than just a little direction.
The problem here is that the product has no specific point to it - it exists *solely to produce vendor lock in*. Since it's little more than a re-badged Android TV stick there's really nothing special at all about it. This, in a market space that's saturated with me-too also-rans.
It's not that Amazon's offering is horrible, it's that it's not notable in a field littered with the corpses of other not-notable failed products.
BTW: Very few "normal" people are actually aware of hypermiling. Techies/nerds more so.
You can pretty significantly improve your fuel economy by using a "hybrid" approach, which I do to raise the average fuel economy of my car by about 20%, which is significant. This, by the way, includes enough "pedal to the medal" incidents that I do get to enjoy the 200 HP engine in my beautiful convertible!
Simple things, like trailing cars going through timed lights, letting off the gas a mile or so before your turn off so you bleed speed from 65 to 50 or so before exiting and watching a half mile ahead for red lights can improve fuel economy significantly without pissing people off. If you were in the car while I drove, it's likely you wouldn't notice unless I said something.
Ha ha... paying attention goes oh so much deeper than countdown timers...
What most people don't know is that you can improve your fuel economy rather dramatically using a variety of techniques commonly referred as "Hyper-miling". I didn't think much of it myself until I got a car that has a fuel economy computer built into the dash, and then it started to click.
See, brakes are death to fuel economy. Sounds obvious, but what isn't obvious is what that translates to in real world use.
Example: negotiating a red light. Most people don't pay attention to red lights until they are half a block or so away. If it's red, they start to apply the brake, and then as the light stubbornly refuses to turn green, they apply more and more brake until they stop behind the next car. Which is exactly the *wrong* way to get best fuel economy. Instead, you should be looking ahead as far as possible, and apply the brake as early as possible to reduce speed as early as possible to increase the amount of time it takes to cover the block distance while losing as little forward momentum as possible. Instead of waiting until the last minute and losing all forward momentum, you brake early and keep perhaps 30 MPH. This means that you don't have to accelerate to 30 MPH and you save that much fuel.
It was rather surprising to me how much difference I could accomplish using these techniques! On the freeway, if I drive around 50 MPH unless going up a hill, then more like 40-45, the normal 25-28ish MPG becomes closer to 34 MPG. Around the town, normally, my car (a 4-seat Chrysler convertible) gets around 18-20 MPG, but using these techniques about braking and reduced acceleration, I can get over 30 MPG on town surface streets! (flat land) Unfortunately, I do have to get used to being flipped off in order to achieve this.
In any event, you *can* get a rather sharp increase in fuel economy by paying attention to the forces of momentum, timing and friction.
Nothing like watching shows on illegitimate, foreign-hosted websites laden with pr0n ads and drive-by zero-day flash exploits... Sorry, no. I, for one, have a detectable amount of self respect.
It looks like it should be a fairly simple matter.
... which demonstrates that you don't understand the problem. See, people die and will die in every production car ever made. Accidents happen, and with enough money spent, virtually every single death could be prevented, with enough additional shielding, crumple zones, and whatnot. But the result would be a car that nobody could afford, not even by a long shot.
So, in a very literal sense, every single car you've ever driven is a balance between the amount you're willing to pay and the amount of risk you're willing to assume. At what point does a death become a death due to a design flaw? Well, guess what: there is no obvious, intuitive "line" that would be easy to sue over.
An ignition switch occasionally dies? Would you *want* a car company that didn't improve its designs as it learns more about them? But on the balancing side, you literally cannot afford to pay for a car getting a recall every time an improvement is made.
There are a ton of shades of grey, and it's not callous, it's just sensible business to balance costs with safety. Doing it any other way would be the foolish way to do things.
Slashdot Mirror
But in practice, there is no meaningful difference between "QoS" and "Net Neutrality" as there is no meaningful definition of what constitutes a "type" of traffic.
For example, NetFlix uses a traffic based on SilverLight's streaming video capability. Comcast only has to change this ever so slightly so it's a different "type" and then they can claim QoS. There is no meaningful limit to the rules that can be used to determine QoS, thus there is no meaningful distinction between QoS and "Net Neutrality".
I wonder how many of those jobs for "big data" involve data sets that will sit on a USB drive?
PHB: "But it's a TERABYTE, at least!"
I just remember how critical everybody here on /. was about the iPod. How many years ago was that?
In case you hadn't noticed, Credit Card companies secure your credit card using techniques very similar to A/V vendors' products. They do heuristic scanning of transactions, looking for consumer spending patterns and throwing red flags when they change significantly. You can wax poetic all you want about "smart cards" but the system is big enough that we'll probably *never* be without similar methods for protecting your bank account
In my laptop, I have an SSD. Upgrading the HDD cost about as much as a new laptop and cost significantly less. I've been able to buy 2+ years of time on my old laptop with an upgrade at significantly less cost.
So the numbers make sense, here!
We host a heavily database-driven app. Use of an SSD reduces latency by at *least* 95% in our testing. It's a no-brainer. Even if we replaced the SSDs every single year, we'd still come out way ahead. SSDs are where it's at for perfromance!
I don't understand why a spine is necessary. If the market is crying out for XP, why not just simply migrate XP to an annual license of $20 and let people keep their beloved 512 MB P IVs going for as long as their bits keep shuffling?
Microsoft had an awesome opportunity with WinXP and they just threw it away...
I would argue, it's not that he's *right*, it's that he's right in a way that still doesn't help.
Wife: "Honey, do you know where my keys are?"
Husband: "Pretty sure they're right where you left them!".
Technically correct, completely unhelpful.
And that's how QMail is. (or was, I stopped using it years ago) Qmail is a nightmare I'd rather forget. Sure, it is/was very secure, modularly written with strong privilege separation, built-in clustering support, etc. But because of the horribly restrictive license it was under, you had to download it and apply a half dozen patches in order to get it to do what you wanted. Worse, the patches were often somewhat in conflict, so you end up doing patches in a specific order to get what you want, and/or manually editing files. And then, when you are done, you have something that not only has its own init system, it actually conflicts with the standard init system so you have to pretty much disable every other service on the said server.
But you know what seems like the most asshole part of all? Updates are infrequent at best, and the license doesn't allow you to distribute updated versions. It just isn't going to get any better. Everywhere else, you run a single command (EG: yum install postfix) but to get qmail going takes weeks.
I'm a big fan of DJB's code quality, and the new Crypto being released as LGPL means I would actually get behind using his code. I'm just glad that heartbleed means that the critical security infrastructure will finally get somewhere near the attention it deserves...
Two words: Synchronized Panes. For nothing else, this is enough.
Need to do a semi-repetitive task that's not quite annoying enough to script on a bunch of servers? TMUX to the rescue! You can open 10 windows to 10 different servers, and synchronize what you type so it shows up on all screens, or click on an individual screen to run just that command there.
For boring admin chores, it's a god-send.
but the pure electric car isn't going to be ready until a) massive updates to the power grid b) swappable batteries c) battery tech that lets cars go 500-1000 miles on a charge.
Why the boolean logic?
In case you hadn't noticed, pure electric cars are stomping the ever loving crap out of the luxury/performance car market. So long as the cars are selling at a growing pace, they are here to stay and are ready for the people who continue to buy them.
And as long as this happens, manufacturers will make continuing improvements to the cars they make.
A) The power grid is constantly being worked on. As people buy more cars, the grid will be upgraded to match demand.
B) Swappable batteries might be one of those improvements. But they don't seem to be required, at least not yet.
C) 1000 miles on a charge? Show me any common car that gets anything like that range.
Lots of people expect the world to change all of a sudden. But it doesn't really. Instead, continuing incremental changes gradually make the world into a different place. Those incremental changes have rather drastically changed how people interact in just the 30 or so years that I can personally remember.
By reading this comment, you waive all rights to the ownership of your car...
OP starts with: "I am the sole sysadmin for nearly 50 servers (win/linux) across several contracts. ..."
This implies that he's paid hourly. Contracts implies that he's a consultant. If there's anything that a consultant craves, it's billable hours...
Your points are valid, in a sense. But do you really think that people are going to stop trusting Open Source technologies? What isn't part of the conversation is just how terribly horrible OpenSSL actually is. It's a readability nightmare. The patch makes my eyes bleed, makes me weep gently to myself as I rock myself in an attempt to succor the horrific nightmare that code of this quality is what drives most Internet "security".
I so sorely wish more consideration was given towards NACL as a replacement for OpenSSL. It's clean, elegant, readable. Bugs will be shallower because readers might have *some idea* what is going on. And with an LGPL license, it should be quite embeddable.
IMHO, OpenSSL should be toss summarily as soon as possible. Beneath its horrific API and code lurk untold numbers of nascent, undiscovered holes no doubt already being exploited by our good friends at the NSA.
Writing security code is *hard*, folks. Making it hard to read only makes it impossible to debug...
^^^ This. I bought an officially branded Comcast MODEM on ebay for $21 after shipping. It paid for itself in just 3 months, and hasn't given me a lick of trouble in YEARS.
Well, that is certainly possible if that's important to you. .
You also could have clicked the link I provided,l and perhaps typed the word "ethernet"... There are lots of models covering many different use cases. My unit works great over wifi, which is great since I have no Ethernet in my bedroom...
Google TV isn't a failure, it's just not the only success.
I have a Google TV stick and I love it! It is just a tablet that uses my TV as its screen and a wireless keyboard as its input. It is about the size of a thumb drive. It cost $40 on Amazon (.search for mk808b to get the exact model I'm watching Hulu+ on as I write this)
See my post history for details: this is quite successful. I have no idea what Google would want to improve...
Play store: no problem. How else do you think I installed Netflix, CBS.com, Hulu, uTorrent, and all the other apps?
Seriously, just imagine a tablet running on your TV using a mouse/remote instead of a touch screen. That's what I use every day. (And what is currently playing Sherlock Holmes a la Hulu; my wife loves that show)
I have a "Google TV" and I love it! Also called a "TV Stick" they are best sellers on Amazon with many models to choose from starting at around $25. I bought an MK808B for my bedroom TV and it's hard not to love.
1) It cost $40.
2) It uses my already existing TV
3) It streams Hulu, Netflix, CBS, NBC, and any other TV network that bothers with an Android app over wifi.
4) It uses about 2.5 watts of power.
5) It's not much bigger than a thumb stick.
6) It works seamlessly with an "air mouse" wireless remote.
7) It plays MP4 videos fluidly and runs uTorrent without issue.
8) It has room for two USB devices and an SD card.
9) Effortless support for 1080p resolution.
What more do you want from set top box that actually hides behind the TV?
Seriously: a major airplane "disappears" despite evidence that it wasn't really crashed. Everybody's wondering who dunnit and how, and whether or not it will become another impromptu bomb.
There's a *lot* you can carry on a 777. $50 mil is a lot, but the amount of damage such a plane could do with a little direction makes $50 mil look like peanuts. And it's pretty clear that anybody with the skills to make it disappear as completely as it did is capable of more than just a little direction.
You know, it could be that Tesla enlists the help of elite hackers who have compromised other high-flying products to harden their systems before somebody gets killed?
The problem here is that the product has no specific point to it - it exists *solely to produce vendor lock in*. Since it's little more than a re-badged Android TV stick there's really nothing special at all about it. This, in a market space that's saturated with me-too also-rans.
It's not that Amazon's offering is horrible, it's that it's not notable in a field littered with the corpses of other not-notable failed products.
BTW: Very few "normal" people are actually aware of hypermiling. Techies/nerds more so.
You can pretty significantly improve your fuel economy by using a "hybrid" approach, which I do to raise the average fuel economy of my car by about 20%, which is significant. This, by the way, includes enough "pedal to the medal" incidents that I do get to enjoy the 200 HP engine in my beautiful convertible!
Simple things, like trailing cars going through timed lights, letting off the gas a mile or so before your turn off so you bleed speed from 65 to 50 or so before exiting and watching a half mile ahead for red lights can improve fuel economy significantly without pissing people off. If you were in the car while I drove, it's likely you wouldn't notice unless I said something.
Bezos isn't crazy? How much do you want to bet on that?
The Drones idea has more holes in it than an R/O water filter....
Ha ha... paying attention goes oh so much deeper than countdown timers...
What most people don't know is that you can improve your fuel economy rather dramatically using a variety of techniques commonly referred as "Hyper-miling". I didn't think much of it myself until I got a car that has a fuel economy computer built into the dash, and then it started to click.
See, brakes are death to fuel economy. Sounds obvious, but what isn't obvious is what that translates to in real world use.
Example: negotiating a red light. Most people don't pay attention to red lights until they are half a block or so away. If it's red, they start to apply the brake, and then as the light stubbornly refuses to turn green, they apply more and more brake until they stop behind the next car. Which is exactly the *wrong* way to get best fuel economy. Instead, you should be looking ahead as far as possible, and apply the brake as early as possible to reduce speed as early as possible to increase the amount of time it takes to cover the block distance while losing as little forward momentum as possible. Instead of waiting until the last minute and losing all forward momentum, you brake early and keep perhaps 30 MPH. This means that you don't have to accelerate to 30 MPH and you save that much fuel.
It was rather surprising to me how much difference I could accomplish using these techniques! On the freeway, if I drive around 50 MPH unless going up a hill, then more like 40-45, the normal 25-28ish MPG becomes closer to 34 MPG. Around the town, normally, my car (a 4-seat Chrysler convertible) gets around 18-20 MPG, but using these techniques about braking and reduced acceleration, I can get over 30 MPG on town surface streets! (flat land) Unfortunately, I do have to get used to being flipped off in order to achieve this.
In any event, you *can* get a rather sharp increase in fuel economy by paying attention to the forces of momentum, timing and friction.
Nothing like watching shows on illegitimate, foreign-hosted websites laden with pr0n ads and drive-by zero-day flash exploits... Sorry, no. I, for one, have a detectable amount of self respect.
It looks like it should be a fairly simple matter.
... which demonstrates that you don't understand the problem. See, people die and will die in every production car ever made. Accidents happen, and with enough money spent, virtually every single death could be prevented, with enough additional shielding, crumple zones, and whatnot. But the result would be a car that nobody could afford, not even by a long shot.
So, in a very literal sense, every single car you've ever driven is a balance between the amount you're willing to pay and the amount of risk you're willing to assume. At what point does a death become a death due to a design flaw? Well, guess what: there is no obvious, intuitive "line" that would be easy to sue over.
An ignition switch occasionally dies? Would you *want* a car company that didn't improve its designs as it learns more about them? But on the balancing side, you literally cannot afford to pay for a car getting a recall every time an improvement is made.
There are a ton of shades of grey, and it's not callous, it's just sensible business to balance costs with safety. Doing it any other way would be the foolish way to do things.