It varies by state, however, there are usually 3 different laws. Failure to move right (which speed limit typically does not matter), impeding the flow of traffic (often referred to as wolf packing), and failure to use a turn off (x number of cars want to pass on 2 lane road).
You do realize that YOU are breaking the law by not moving over when you are driving in the left lane? Yes, even if you are speeding. YOU should learn to read the damn traffic laws, forfeit your license, or both.
You could make that argument I suppose. I wasn't thinking of Apple because it's not really in the same market as Microsoft. Apple sells hardware and they license their OS to run on it. Microsoft's market is stand-alone OS software sales, and Apple isn't part of that, nor is any phone or tablet. So I guess it depends on how you look at it.
I suppose Microsoft could (in some fantasy world), drop it's stand-alone OS completely, and resell linux on their surface hardware, but that's a silly idea.
Oh wait, no, there isn't. And I suspect no, you won't spend the 200 million man hours writing the OS and then release it for free either. Until then, $100 seems like a pretty good deal.
So OSX, Linux and BSD doesn't count as competitors? And the last two either aren't free, or haven't taken a massive amount of development?
You sort of missed the point. Jedidiah proposes that the value of Windows if it had no vendor-lockin would be $0, and $100 is overpriced. Compared to what? It's sort of like saying all houses are worth $0, because well there are those homes for homeless programs where people chip in and build a house and give it away for FREE. So therefore, all houses are worthless. Just because linux and BSD exist, and many thousands of people "donated" their time into making them, doesn't make Windows worthless either. It is sort of a silly argument he is making, taken from a self-centric point of view.
1. NVidia's drivers for x86 is miles ahead of their ARM drivers. 2. Slowing progress? The x86 platform is pretty energy efficient now, it is almost silly to develop for ARM now considering by developing for x86 you can cover all your bases from desktop to mobile.
This may suprise you, but that is only true if you want to support old hardware. Anything cutting edge is more likely supported by windows than Linux, and since I have more cutting edge hardware than old relics, it isn't true for me.
Currently, when I run games, they typically use less than 20% of my CPU, and I bet if I disabled all my unused services that I don't need to run the current game, I could get that down to, oh... 19.95% CPU usage, so I could have one more core sitting idle for.05% longer.
$100 is overpriced for this years version of a well entrenched monopoly product. If not for vendor-lock, the value of that product would be $0.
I think you are on to something there. Write me an OS that can run all my games and applications on it in a user friendly manner that I don't have to spend days googling the right command to type in to get it what I need it to do, and support the major video cards out there running in SLI, the major chipsets, RAID drivers, etc etc, and sell it for $0, and I'm sure it will be a hit.
If it was so overly priced you'd have a ton of competitors out there. Oh wait, no, there isn't. And I suspect no, you won't spend the 200 million man hours writing the OS and then release it for free either. Until then, $100 seems like a pretty good deal.
Simple math man, you take the last digit (9) and compare it to the last digit (2), and since 9>2*2, it's less than half. What do they teach you in school these days?
Google is actually a publicly traded international company with headquarters in the US. Is it that the headquarters that is in the US that bothers you? Or that the CEO is from the US? What makes either of those things more suspect than if Google was run by a German national with headquarters in the US, or a CEO born in the US with headquarters in Germany? Or is it that Google does business in the US? So if google had a German CEO, HQ in Germany, and did some business in the US, does that make it more suspect?
Please ignore the earthquakes. 14 people dead. It's much better than say the 5 deaths (direct) in the last decade from nuclear (0 from Fukushima, 1 from Mayapuri, and 4 from Mihama). Gas and oil is so much safer, that I'm sure this is the last instance of this to happen for the next 50 years. I mean the green eco-freaks couldn't possibly be wrong, could they?
Yes, but it does. Google SPDY for performance tests between it and HTTP 1.1. Because you can't effectively pipeline requests over HTTP 1.1, that feature isn't used nor supported in any current browser, nor will it likely ever be. Without it, a dropped packet will stall the arrival of subsequent data to the application, which then stalls the next request, effectively stopping the connection until it receives the missing packet.
HTTP 2.0 does not suffer from that (as much) because all the known needed resources have already been requested as soon as they become known that they will be needed. In addition, even the very first request can be sent in parallel if the server supports it.
Here is an example of the later. You hit www.mysite.com. It page consists of 1 HTML file, 3 CSS files, 10 JavaScript files, and 25 images. You request the HTML file, and the very first packet is dropped. the client hasn't received the HTML yet, so it doesn't know about the other 38 files it needs. So not only is the HTML file stalled, but all 38 others aren't being requested.
Now under HTTP 2.0, when you request the HTML file, it can anticipate you will need the other 38 files, and begin sending them to the client immediately. Sure, the application won't see them until the missing packet arrives, but as soon as the one packet is retransmitted, the application will receive all the data that was continuing to be sent after it all at once, and the connection never stalled.
Scenario 2: the HTML file is sent ok, the browser parses it, and begins requesting the additional resources (first of all, this is flawed because you don't need the entire HTML to begin parsing, but I am keeping it simple). So additional resources are spread across the 6 connections to the server, and one of them has a packet dropped. The server finishes sending all the data for the resource (or it is "in flight") long before the missing packet is detected. Since the client application does not receive the data, it can not request additional resources to be sent over that connection. That connection is effectively stalled now until the missing packet is received, and the client is now only using 5 connections efficiently.
Again, in HTTP 2.0, this doesn't have the same effect because it can request all 38 additional resources as soon as possible. The data is being sent to the client in parallel over the single connection.
Does that help clear it up? If you want actual performance tests, google SPDY, and you will see that this method actually does improve performance significantly when packets are lost.
That seems silly. You don't need an internet connection to your house to pay taxes. You can walk down to the nearest internet cafe, starbucks, McDonalds, or pay a tax service to file it for you. Inconvenient sure, but isn't that the point?
Slashdot Mirror
It varies by state, however, there are usually 3 different laws. Failure to move right (which speed limit typically does not matter), impeding the flow of traffic (often referred to as wolf packing), and failure to use a turn off (x number of cars want to pass on 2 lane road).
You do realize that YOU are breaking the law by not moving over when you are driving in the left lane? Yes, even if you are speeding. YOU should learn to read the damn traffic laws, forfeit your license, or both.
You could make that argument I suppose. I wasn't thinking of Apple because it's not really in the same market as Microsoft. Apple sells hardware and they license their OS to run on it. Microsoft's market is stand-alone OS software sales, and Apple isn't part of that, nor is any phone or tablet. So I guess it depends on how you look at it.
I suppose Microsoft could (in some fantasy world), drop it's stand-alone OS completely, and resell linux on their surface hardware, but that's a silly idea.
Oh wait, no, there isn't. And I suspect no, you won't spend the 200 million man hours writing the OS and then release it for free either. Until then, $100 seems like a pretty good deal.
So OSX, Linux and BSD doesn't count as competitors? And the last two either aren't free, or haven't taken a massive amount of development?
You sort of missed the point. Jedidiah proposes that the value of Windows if it had no vendor-lockin would be $0, and $100 is overpriced. Compared to what? It's sort of like saying all houses are worth $0, because well there are those homes for homeless programs where people chip in and build a house and give it away for FREE. So therefore, all houses are worthless. Just because linux and BSD exist, and many thousands of people "donated" their time into making them, doesn't make Windows worthless either. It is sort of a silly argument he is making, taken from a self-centric point of view.
1. NVidia's drivers for x86 is miles ahead of their ARM drivers.
2. Slowing progress? The x86 platform is pretty energy efficient now, it is almost silly to develop for ARM now considering by developing for x86 you can cover all your bases from desktop to mobile.
Are you trying to say that my numbers are off somehow? Hell, I'll be super generous today and let's just say 1.6%. Does that change things for you?
This may suprise you, but that is only true if you want to support old hardware. Anything cutting edge is more likely supported by windows than Linux, and since I have more cutting edge hardware than old relics, it isn't true for me.
Do the right thing MS.. Move to *nix back end like the other 0.80% that use *nix
Fixed that for you.
Currently, when I run games, they typically use less than 20% of my CPU, and I bet if I disabled all my unused services that I don't need to run the current game, I could get that down to, oh... 19.95% CPU usage, so I could have one more core sitting idle for .05% longer.
$100 is overpriced for this years version of a well entrenched monopoly product. If not for vendor-lock, the value of that product would be $0.
I think you are on to something there. Write me an OS that can run all my games and applications on it in a user friendly manner that I don't have to spend days googling the right command to type in to get it what I need it to do, and support the major video cards out there running in SLI, the major chipsets, RAID drivers, etc etc, and sell it for $0, and I'm sure it will be a hit.
If it was so overly priced you'd have a ton of competitors out there. Oh wait, no, there isn't. And I suspect no, you won't spend the 200 million man hours writing the OS and then release it for free either. Until then, $100 seems like a pretty good deal.
Stop with your silly fake opinions.
Or moved to the mobile platform.
Isn't that what I said?
We take a felons right to vote (without a doubt a more important right) so why can't we take away their copy right.
I think you are confused. Felons can vote, they just can't run for presidency.
And those people who get so upset should consider a career change, or perhaps working in the creative-technical department of children's toys.
It's because anything multiplied by 1 is anything.
Simple math man, you take the last digit (9) and compare it to the last digit (2), and since 9>2*2, it's less than half. What do they teach you in school these days?
Well then I would have to count the gulf oil spill in 2006 that killed 11 and contaminated 68,000 sq miles as well.
But is it Web Scale?
Google is actually a publicly traded international company with headquarters in the US. Is it that the headquarters that is in the US that bothers you? Or that the CEO is from the US? What makes either of those things more suspect than if Google was run by a German national with headquarters in the US, or a CEO born in the US with headquarters in Germany? Or is it that Google does business in the US? So if google had a German CEO, HQ in Germany, and did some business in the US, does that make it more suspect?
How exactly is Google a private US company?
That's why I promote fracking in my back yard! Let someone else get the big quake!
Please ignore the earthquakes. 14 people dead. It's much better than say the 5 deaths (direct) in the last decade from nuclear (0 from Fukushima, 1 from Mayapuri, and 4 from Mihama). Gas and oil is so much safer, that I'm sure this is the last instance of this to happen for the next 50 years. I mean the green eco-freaks couldn't possibly be wrong, could they?
Yes, but it does. Google SPDY for performance tests between it and HTTP 1.1. Because you can't effectively pipeline requests over HTTP 1.1, that feature isn't used nor supported in any current browser, nor will it likely ever be. Without it, a dropped packet will stall the arrival of subsequent data to the application, which then stalls the next request, effectively stopping the connection until it receives the missing packet.
HTTP 2.0 does not suffer from that (as much) because all the known needed resources have already been requested as soon as they become known that they will be needed. In addition, even the very first request can be sent in parallel if the server supports it.
Here is an example of the later. You hit www.mysite.com. It page consists of 1 HTML file, 3 CSS files, 10 JavaScript files, and 25 images. You request the HTML file, and the very first packet is dropped. the client hasn't received the HTML yet, so it doesn't know about the other 38 files it needs. So not only is the HTML file stalled, but all 38 others aren't being requested.
Now under HTTP 2.0, when you request the HTML file, it can anticipate you will need the other 38 files, and begin sending them to the client immediately. Sure, the application won't see them until the missing packet arrives, but as soon as the one packet is retransmitted, the application will receive all the data that was continuing to be sent after it all at once, and the connection never stalled.
Scenario 2: the HTML file is sent ok, the browser parses it, and begins requesting the additional resources (first of all, this is flawed because you don't need the entire HTML to begin parsing, but I am keeping it simple). So additional resources are spread across the 6 connections to the server, and one of them has a packet dropped. The server finishes sending all the data for the resource (or it is "in flight") long before the missing packet is detected. Since the client application does not receive the data, it can not request additional resources to be sent over that connection. That connection is effectively stalled now until the missing packet is received, and the client is now only using 5 connections efficiently.
Again, in HTTP 2.0, this doesn't have the same effect because it can request all 38 additional resources as soon as possible. The data is being sent to the client in parallel over the single connection.
Does that help clear it up? If you want actual performance tests, google SPDY, and you will see that this method actually does improve performance significantly when packets are lost.
That seems silly. You don't need an internet connection to your house to pay taxes. You can walk down to the nearest internet cafe, starbucks, McDonalds, or pay a tax service to file it for you. Inconvenient sure, but isn't that the point?