It is one thing to poke fun at yourselves. But when your company itself produces machines that need to be considered trustworthy, having such a signature does not help to promote any sort of trust. Remember that trust is not always based on fact, but on the perception others hold of you.
I doubt the company cares much. The publicity and market attention they will receive from this deal will be more than enough to compensate for any licensing wrongdoings.
It says the total cost difference between a tablet and comparable notebook is about $200. Of that amount only $30-$60 is due to hardware, the rest is the extra software licensing cost. That is a $140-$170 premium for Windows XP Tablet Edition. To me, for a machine that costs a few thousand, even a $200 difference does not seem that much. Or maybe people just haven't gotten used to the technology enough to make it a worthwhile purchase yet?
It is sad, we have arrived in a day and age where it seems as though every new technology that comes around the block needs to make it big in the first couple years , or it is considered a failure. Real improvements in productivity don't happen that way. They can take many years before the returns are actually realized. The people who use the technology don't learn it overnight. In fact, it is only now that many companies are finally starting to see a decent return on their investments in technology in the late 1990's.
Re:Makes perfect sense to ask slashdot...
on
Digital 35mm SLRs?
·
· Score: 0, Troll
.. Yeesh, next thing you know we are going to have an Ask Slashdot on Relationship Advice.
If the company really is valued at $15-20 billion, I doubt Microsoft would be willing to part with that much of their cash hoard. Also, it would be considered a hostile takeover, which does not work very well for software companies unless they have reached maturity.
"By becoming public, google loses the ability to continue with constant steady growth and innovative R&D. These things will invariably lead to short sighted planning by the management to "make the numbers" for the next quarter, 6 months, or year. "Growth" will be expected year after year - the innovative ideas that have made google so successful will give way."
Actually, quite the opposite. By staying private they lose the ability to keep the talent on board by issuing those high-valued employees stock in the company. Sergey Brin himself has mentioned this several times in the past. While being private gives the company more freedom with its financial affairs, it needs an IPO to keep growing and move forward.
I think the point to be made though is that not ALL open source software is inherently more secure than all closed source software. It all varies from project to project.
That is of course assuming they are able to obtain bankruptcy protection:
"X10 filed what the bankruptcy court termed a "deficient" filing, meaning that it lacked a statement of its financial affairs. The court set a 15-day deadline for the completion of the filing, or X10 risks a dismissal."
Availability of money to borrow cheap is only a good thing for an economy if that money is used profitably.
Not true. As long as the executives aren't taking the money off to the Carribean with them and spending it on tourism there (which can then be considered an import). If the money is spent domestically, it will go to the various local suppliers and that helps the economy by itself. Also, even if the company is losing money, if they are producing something then that is helping GDP and the economy too.
Probably so that in 25 years, they don't have to revisit it again and implement an IPv8. Also, the design of IPv6 is very different than IPv4. The 128 bits are actually two distinct 64bit identifiers combined together. The first 64 bits indicates the subnet. Of that first 64 bits, 48 are there to be used in partitioning the network in different ways (it's an oversimplification I know and I am dumbing down some details). The last 64 bits are your 'interface identifier', this is the equivalent of your 48bit MAC address. Only now the MAC address is going to be part of your address.
Yes, I read somewhere the average routing table sizes for an IPv6 router will shrink to about 8000 entries, as compared to the 50,000 or so that they have to maintain today. Not sure if it will really speed up the nodal processing delays by factor of 5x or 6x though.
As far as I know, they are not going to be dividing up the addresses into classes like before. There will be specific size blocks like a/32 or a/48 that users will get. But only the tier-1 one ISPs will actually be assigned blocks from the regional authorities. Everyone else will just request certain size blocks from their direct provider. All of the details are discussed in the RFCs.
I think there is a good chance that IP spoofing will be reduced in IPv6. Because of the hierarchial design of IPv6, I think that most ISPs will probably configure their near-edge/gateway routers to drop any packets that have source addresses not within their respective network.
It is true, the IPv4 address space is not as depleted or near exhaustion as many have made it out to be. However, that fact is hiding something else. The reason the IPv4 pool not been completely exhausted is the regional authorities are keeping a very tight fist on how many addresses are being given out. This is the same reason why most people cannot get more than one IP address from their ISP without paying extra. And the ISP's themselves are fighting for extra address space.
Once the world starts to go IPv6, there won't be the same problems with getting extra IPs. Lower tier ISPs won't have to go all the way up the food chain to request more address blocks. I believe the policy with IPv6 is you just request more from your provider. These benefits will hopefully trickle down to end-users like us, and we will be able to hook up our multiple computers, laptop, VoIP phone, and toaster without paying substantially extra.
My understanding is that the 6bone was not a production network.
Also, I imagine the reason the DoD is building their own network right now, is so that they have more control over it. They don't need home users bringing havoc over the network while they are conducting their tests. It may also be a temporary network, and they could have further plans down the road to introduce further major changes to it.
There are a lot of good reasons for everyone to upgrade. There is a good article over at CommsWorld about this. Basically the main reason for upgrading is innovation. Once everyone can attach a public IP address to all of their devices, there will be a lot of cool stuff that will come out.
(Note: the article was originally linked to from CircleID)
Either that, or most software will finally learn to adapt to either 2 ways: 1) Automatically learn your IP, or the most likely way 2) Use DNS hostnames.
Instead of relying on only IP addresses, most software will simply use a gethostbyname( ) method to automatically learn the IP address.
It is hard to respect a writer who equates Linux zealots, with people that seek nothing less than the complete failure of our modern society and who are willing to fight to death for it.
I think the writer missed out on a big group of people: He seemed to go straight from describing a group of platform-neutral people (Pros) to people who promote Linux without any factual backup (Priests). That's skipping out on a whole lot of people who really prefer the Linux platform for many good reasons. I would consider myself fairly platform-neutral, but I at least respect those that favor Linux over the other platforms.
I think the writer has invoked a modern day version of Godwin's Law: replacing the use of Nazis with Terrorists.
You misunderstand what client/server is. A server is simply a process that sits on a machine listening for connections, and a client is the one responsible for initiating connections. p2p applications are client/server oriented, its just that the happen to fulfill both roles of client and server sometimes.
When your computer makes a TCP connection to another host, your machine is a client and the other is a server. Its as simple as that. As long as TCP is around, the client/server model will always remain.
You are partly right. Web applications are based on thin-client/server architecture. What the author should have said was that the typical fat-client/server architecture is fading away. I think this is true for most business application and productivity-type applications. However, as long as TCP is around there will always be new innovative applications that use the typical client/server method.
Slashdot Mirror
It is one thing to poke fun at yourselves. But when your company itself produces machines that need to be considered trustworthy, having such a signature does not help to promote any sort of trust. Remember that trust is not always based on fact, but on the perception others hold of you.
Because it is an employee of the company, and it appears to reflect a position that someone working for such a company should hopefully not have.
The goal is to figure out exactly what the users need and offer no more.
And that is an unattainable goal because not every user has the same set of needs.
I doubt the company cares much. The publicity and market attention they will receive from this deal will be more than enough to compensate for any licensing wrongdoings.
It says the total cost difference between a tablet and comparable notebook is about $200. Of that amount only $30-$60 is due to hardware, the rest is the extra software licensing cost. That is a $140-$170 premium for Windows XP Tablet Edition. To me, for a machine that costs a few thousand, even a $200 difference does not seem that much. Or maybe people just haven't gotten used to the technology enough to make it a worthwhile purchase yet?
It is sad, we have arrived in a day and age where it seems as though every new technology that comes around the block needs to make it big in the first couple years , or it is considered a failure. Real improvements in productivity don't happen that way. They can take many years before the returns are actually realized. The people who use the technology don't learn it overnight. In fact, it is only now that many companies are finally starting to see a decent return on their investments in technology in the late 1990's.
.. Yeesh, next thing you know we are going to have an Ask Slashdot on Relationship Advice.
If the company really is valued at $15-20 billion, I doubt Microsoft would be willing to part with that much of their cash hoard. Also, it would be considered a hostile takeover, which does not work very well for software companies unless they have reached maturity.
"By becoming public, google loses the ability to continue with constant steady growth and innovative R&D. These things will invariably lead to short sighted planning by the management to "make the numbers" for the next quarter, 6 months, or year. "Growth" will be expected year after year - the innovative ideas that have made google so successful will give way."
Actually, quite the opposite. By staying private they lose the ability to keep the talent on board by issuing those high-valued employees stock in the company. Sergey Brin himself has mentioned this several times in the past. While being private gives the company more freedom with its financial affairs, it needs an IPO to keep growing and move forward.
I think the point to be made though is that not ALL open source software is inherently more secure than all closed source software. It all varies from project to project.
Availability of money to borrow cheap is only a good thing for an economy if that money is used profitably.
Not true. As long as the executives aren't taking the money off to the Carribean with them and spending it on tourism there (which can then be considered an import). If the money is spent domestically, it will go to the various local suppliers and that helps the economy by itself. Also, even if the company is losing money, if they are producing something then that is helping GDP and the economy too.
Probably so that in 25 years, they don't have to revisit it again and implement an IPv8. Also, the design of IPv6 is very different than IPv4. The 128 bits are actually two distinct 64bit identifiers combined together. The first 64 bits indicates the subnet. Of that first 64 bits, 48 are there to be used in partitioning the network in different ways (it's an oversimplification I know and I am dumbing down some details). The last 64 bits are your 'interface identifier', this is the equivalent of your 48bit MAC address. Only now the MAC address is going to be part of your address.
Okay, I understand and agree with most of your post. But how is getting rid of the IP header checksums a good thing?
Yes, I read somewhere the average routing table sizes for an IPv6 router will shrink to about 8000 entries, as compared to the 50,000 or so that they have to maintain today. Not sure if it will really speed up the nodal processing delays by factor of 5x or 6x though.
As far as I know, they are not going to be dividing up the addresses into classes like before. There will be specific size blocks like a /32 or a /48 that users will get. But only the tier-1 one ISPs will actually be assigned blocks from the regional authorities. Everyone else will just request certain size blocks from their direct provider. All of the details are discussed in the RFCs.
So use DNS instead? Other than DNS servers, I don't see why applications still require you to put in an IP address exclusively over a hostname.
I think there is a good chance that IP spoofing will be reduced in IPv6. Because of the hierarchial design of IPv6, I think that most ISPs will probably configure their near-edge/gateway routers to drop any packets that have source addresses not within their respective network.
It is true, the IPv4 address space is not as depleted or near exhaustion as many have made it out to be. However, that fact is hiding something else. The reason the IPv4 pool not been completely exhausted is the regional authorities are keeping a very tight fist on how many addresses are being given out. This is the same reason why most people cannot get more than one IP address from their ISP without paying extra. And the ISP's themselves are fighting for extra address space.
Once the world starts to go IPv6, there won't be the same problems with getting extra IPs. Lower tier ISPs won't have to go all the way up the food chain to request more address blocks. I believe the policy with IPv6 is you just request more from your provider. These benefits will hopefully trickle down to end-users like us, and we will be able to hook up our multiple computers, laptop, VoIP phone, and toaster without paying substantially extra.
My understanding is that the 6bone was not a production network.
Also, I imagine the reason the DoD is building their own network right now, is so that they have more control over it. They don't need home users bringing havoc over the network while they are conducting their tests. It may also be a temporary network, and they could have further plans down the road to introduce further major changes to it.
There are a lot of good reasons for everyone to upgrade. There is a good article over at CommsWorld about this. Basically the main reason for upgrading is innovation. Once everyone can attach a public IP address to all of their devices, there will be a lot of cool stuff that will come out.
(Note: the article was originally linked to from CircleID)
Either that, or most software will finally learn to adapt to either 2 ways: 1) Automatically learn your IP, or the most likely way 2) Use DNS hostnames. Instead of relying on only IP addresses, most software will simply use a gethostbyname( ) method to automatically learn the IP address.
It is hard to respect a writer who equates Linux zealots, with people that seek nothing less than the complete failure of our modern society and who are willing to fight to death for it.
I think the writer missed out on a big group of people: He seemed to go straight from describing a group of platform-neutral people (Pros) to people who promote Linux without any factual backup (Priests). That's skipping out on a whole lot of people who really prefer the Linux platform for many good reasons. I would consider myself fairly platform-neutral, but I at least respect those that favor Linux over the other platforms.
I think the writer has invoked a modern day version of Godwin's Law: replacing the use of Nazis with Terrorists.
But RAID doesn't work if half the drives break in transport.
You misunderstand what client/server is. A server is simply a process that sits on a machine listening for connections, and a client is the one responsible for initiating connections. p2p applications are client/server oriented, its just that the happen to fulfill both roles of client and server sometimes.
When your computer makes a TCP connection to another host, your machine is a client and the other is a server. Its as simple as that. As long as TCP is around, the client/server model will always remain.
You are partly right. Web applications are based on thin-client/server architecture. What the author should have said was that the typical fat-client/server architecture is fading away. I think this is true for most business application and productivity-type applications. However, as long as TCP is around there will always be new innovative applications that use the typical client/server method.