Every example you just posted requires you to actually examine the output of each of the commands, and apply brittle and convolted text parsing structures like grep and awk. All of if these break when the author alters the output text format. PowerShell has none of those limitations. If an author of ps adds a new property to each object, he does not need to be concerned with previous users of his cmdlet, because nobody is actually parsing his output. His output is strongly typed objects. If a previous user didn't consume his new property, it doesn't matter, they'll continue to not consume it.
Instead of building scripts based on brittle text parsing, they are built on a self documenting model that provides. There is no text parsing. That's extra work. Why do it?
First, your ISP probably won't randomly give you a new prefix. There's no need. You'll probably have a static one assigned. Because there are enough prefixes for every one of their customers to just have on assigned.
Second, manual configuration of IPv6 addresses is almost completely unnecessary, since addresses can be statically assigned within a prefix based on the hardware address of the device. The router gives out the prefix, the device will always have a predictable address underneath it.
Second, you'd probably access these devices using some sort of multicast discovery protocol anyways. A UPnP profile for lighting control already exists, for instance.
Actually, Mono runs all of the non-UI and non-WCF 4.0 code I have. Just fine. And it runs most of the WCF client side stuff just fine.
That includes Linq, and almost everything everybody uses on a daily basis.
It should be pointed out that Mono doesn't magically make the UI differences between iOS, Android and whatever go away. You still have to use native widgets per platform.
So, you can make a core assembly, that you share between all three, and independent UI's on top of that. Sounds like a great situation to me.
I think this CEO is pretty far off the mark. There are good reasons for scrutinizing.Net developers during a hiring process. It is very easy to write software without knowing anything about the task at hand in MS's environment. That does contribute to a large amount of developers out there who say they can program, but couldn't even tell you what a linked list is. It's a serious problem. I'm hiring for a.Net shop and confront it over and over again.
The CEO however goes off and makes a comparison between.Net and McDonalds, saying you can't write good code in it. That's simply not true. You can do anything with it you can do with any other OO/VM platform, but you don't HAVE to. You literally can get away without writing good code. Doesn't mean you can't write good code with it.
Murder is illegal homicide. It's not murder until after it is illegal.
The question of whether it is a life is unimportant. It's a life from the moment of conception. But who cares? I have no problem with the legalized killing of a life (abortion.)
To be honest, I don't know what I'd want a home router to do with IPv6. A firewall would be nice, I guess. And maybe automatically setting up 6to4, and giving out 6to4 addresses?
But if the ISP is handing you a/64, then all it really needs to do is be a passthrough bridge. Each machine gets it's own real honest to goodness public address. Cable modems would certainly need to support it.
A dual stack configuration just means you are running both IPv4 and IPv6. You will have a set of addresses. Your hosts probably already have private IPv4 addresses. Those will remain, and continue working exactly as they are now.
As IPv6 is made available, your router will obtain a public IPv6 address, and hand it out addresses within that subnet to internal hosts. Your hosts will then have BOTH a private IPv4 address, and a public IPv6 address. When your host looks up a name, such as www.google.com, two sets of addresses will be returned: A and AAAA. Your host will make a decision on which local address to use to reach it. Microsoft has a published document describing this process for Windows. I'm imagining it's the same or similar on other OSes.
You'll just have two addresses. One will be a IPv4 that will be NATed, the other won't. Easy. Done.
I know for a fact that every ISP I've talked to is already perfecting an IPv6 infrastructure. Comcast is just talking about it publicly. Every business ISP I've spoken to and asked for addresses has actually given me a time table.
If you have Windows 7, Windows Vista, or any normal Linux distribution in the last few years, they automatically have IPv6 enabled, and obtain their addresses using stateless auto configuration. I'm not sure anybody has dhcp6-client running by default, yet, but stateless auto config is pretty nifty.
Unless you've DISABLED IPv6, or are running some custom setup, you'll probably start grabbing addresses whenever your ISP starts handing them out. Unless you have an edge router, which most people do. Those will need to be updated. Some already are.
If you have Win7 or Vista, you've probably already got public Teredo or 6to4 addresses. If your ISP puts in a close 6to4 router, then the overhead is pretty minimal.
All of my hosts that properly support dual stack, at my organization, have IPv6 addresses. Publicly routable. We're using 6to4 on our edge routers. And that's fine. It's easy to set up. The only "hard part" is reading the addresses... but that passes pretty quickly.
Soon as our ISP offers real IPv6 addresses, we'll just switch to them. IPv6 even has an address migration strategy. It can give out two subnets, and mark one as preferred, and the other as deprecated. Hosts will have both until some point where all hosts have the new ones, and DNS has been updated. It's super flexible, and very well built.
I can vouch that IPv6 works just fine, right now. A dual stack solution is just fine. There's no downside (yet). As public web sites start offering IPv6, and home routers get fixed and replaced with IPv6 supporting solutions, and services like Comcast move over, traffic will just move to IPv6. Sure, we'll need dual stack for ages... but who cares? That's fine.
That's silly. IPv6 is accessible by, for example, every Windows Vista or Windows 7 user, right now. They have Teredo and 6to4 enabled BY DEFAULT. So unless a user bothers to turn these off, they already have public addresses.
And I disagree. Public resources are what? HTTP, services like AIM. As soon as those have IPv6 addresses, traffic starts flowing over IPv6 instead of IPv4. Eventually IPv4 becomes minimal. And some day, in the super distant future, people can start turning it off without even realizing it.
I have IPv6 deployed across my org, with public IPv6 addresses. IPSEC all around. It's great. Yeah, everybody still has NATted addresses. But that's FINE. I can use the benefits of IPv6 right now. From home I can remote desktop to any of my internal machines, SSH to any internal machine, use internal resources like file shares, etc. It's great. Roaming machines get Teredo or 6to4 addresses as well, so they can completely participate with the local network.
Yes, it's going to be years before IPv4 is gone. Probably over a decade. But that's okay. Get started. There are great benefits to IPv6 right now. Consider it an additional feature.
Slashdot Mirror
Every example you just posted requires you to actually examine the output of each of the commands, and apply brittle and convolted text parsing structures like grep and awk. All of if these break when the author alters the output text format. PowerShell has none of those limitations. If an author of ps adds a new property to each object, he does not need to be concerned with previous users of his cmdlet, because nobody is actually parsing his output. His output is strongly typed objects. If a previous user didn't consume his new property, it doesn't matter, they'll continue to not consume it.
Instead of building scripts based on brittle text parsing, they are built on a self documenting model that provides. There is no text parsing. That's extra work. Why do it?
Off switch that works without remembering to hit the off switch will probably save more energy.
Last second = third.
First, your ISP probably won't randomly give you a new prefix. There's no need. You'll probably have a static one assigned. Because there are enough prefixes for every one of their customers to just have on assigned.
Second, manual configuration of IPv6 addresses is almost completely unnecessary, since addresses can be statically assigned within a prefix based on the hardware address of the device. The router gives out the prefix, the device will always have a predictable address underneath it.
Second, you'd probably access these devices using some sort of multicast discovery protocol anyways. A UPnP profile for lighting control already exists, for instance.
I think the sentence specifically does not say what you think it says.
Actually Mono runs 4.0 code just fine. Obviously you can't use WPF on it.
Actually, Mono runs all of the non-UI and non-WCF 4.0 code I have. Just fine. And it runs most of the WCF client side stuff just fine.
That includes Linq, and almost everything everybody uses on a daily basis.
It should be pointed out that Mono doesn't magically make the UI differences between iOS, Android and whatever go away. You still have to use native widgets per platform.
So, you can make a core assembly, that you share between all three, and independent UI's on top of that. Sounds like a great situation to me.
I think this CEO is pretty far off the mark. There are good reasons for scrutinizing .Net developers during a hiring process. It is very easy to write software without knowing anything about the task at hand in MS's environment. That does contribute to a large amount of developers out there who say they can program, but couldn't even tell you what a linked list is. It's a serious problem. I'm hiring for a .Net shop and confront it over and over again.
The CEO however goes off and makes a comparison between .Net and McDonalds, saying you can't write good code in it. That's simply not true. You can do anything with it you can do with any other OO/VM platform, but you don't HAVE to. You literally can get away without writing good code. Doesn't mean you can't write good code with it.
I've never heard him present unscientific nonsense, and I've read all his books.
We draw arbitrary lines. No surprise there. We do the same for driving and drinking and voting age.
Murder is illegal homicide. It's not murder until after it is illegal.
The question of whether it is a life is unimportant. It's a life from the moment of conception. But who cares? I have no problem with the legalized killing of a life (abortion.)
Oh. I run all Windows servers. Because they're, you know, better for some stuff.
To be honest, I don't know what I'd want a home router to do with IPv6. A firewall would be nice, I guess. And maybe automatically setting up 6to4, and giving out 6to4 addresses?
But if the ISP is handing you a /64, then all it really needs to do is be a passthrough bridge. Each machine gets it's own real honest to goodness public address. Cable modems would certainly need to support it.
(does not hold true for PPP-ish services)
I don't understand how running a dual stack is a problem.
A dual stack configuration just means you are running both IPv4 and IPv6. You will have a set of addresses. Your hosts probably already have private IPv4 addresses. Those will remain, and continue working exactly as they are now.
As IPv6 is made available, your router will obtain a public IPv6 address, and hand it out addresses within that subnet to internal hosts. Your hosts will then have BOTH a private IPv4 address, and a public IPv6 address. When your host looks up a name, such as www.google.com, two sets of addresses will be returned: A and AAAA. Your host will make a decision on which local address to use to reach it. Microsoft has a published document describing this process for Windows. I'm imagining it's the same or similar on other OSes.
You'll just have two addresses. One will be a IPv4 that will be NATed, the other won't. Easy. Done.
I know for a fact that every ISP I've talked to is already perfecting an IPv6 infrastructure. Comcast is just talking about it publicly. Every business ISP I've spoken to and asked for addresses has actually given me a time table.
6to4 works until then.
Correction, Windows 7 and Vista will get dhcp6 addresses. Unsure about most Linux distributions.
If you have Windows 7, Windows Vista, or any normal Linux distribution in the last few years, they automatically have IPv6 enabled, and obtain their addresses using stateless auto configuration. I'm not sure anybody has dhcp6-client running by default, yet, but stateless auto config is pretty nifty.
Unless you've DISABLED IPv6, or are running some custom setup, you'll probably start grabbing addresses whenever your ISP starts handing them out. Unless you have an edge router, which most people do. Those will need to be updated. Some already are.
If you have Win7 or Vista, you've probably already got public Teredo or 6to4 addresses. If your ISP puts in a close 6to4 router, then the overhead is pretty minimal.
All of my hosts that properly support dual stack, at my organization, have IPv6 addresses. Publicly routable. We're using 6to4 on our edge routers. And that's fine. It's easy to set up. The only "hard part" is reading the addresses... but that passes pretty quickly.
Soon as our ISP offers real IPv6 addresses, we'll just switch to them. IPv6 even has an address migration strategy. It can give out two subnets, and mark one as preferred, and the other as deprecated. Hosts will have both until some point where all hosts have the new ones, and DNS has been updated. It's super flexible, and very well built.
Evaluating the their end game isn't particularly useful for gauging whether or not the migration plan will be effective.
I can vouch that IPv6 works just fine, right now. A dual stack solution is just fine. There's no downside (yet). As public web sites start offering IPv6, and home routers get fixed and replaced with IPv6 supporting solutions, and services like Comcast move over, traffic will just move to IPv6. Sure, we'll need dual stack for ages... but who cares? That's fine.
The support is built into Vista and Windows 7, and enabled by default, right now.
That's silly. IPv6 is accessible by, for example, every Windows Vista or Windows 7 user, right now. They have Teredo and 6to4 enabled BY DEFAULT. So unless a user bothers to turn these off, they already have public addresses.
And I disagree. Public resources are what? HTTP, services like AIM. As soon as those have IPv6 addresses, traffic starts flowing over IPv6 instead of IPv4. Eventually IPv4 becomes minimal. And some day, in the super distant future, people can start turning it off without even realizing it.
But all of that is okay.
I have IPv6 deployed across my org, with public IPv6 addresses. IPSEC all around. It's great. Yeah, everybody still has NATted addresses. But that's FINE. I can use the benefits of IPv6 right now. From home I can remote desktop to any of my internal machines, SSH to any internal machine, use internal resources like file shares, etc. It's great. Roaming machines get Teredo or 6to4 addresses as well, so they can completely participate with the local network.
Yes, it's going to be years before IPv4 is gone. Probably over a decade. But that's okay. Get started. There are great benefits to IPv6 right now. Consider it an additional feature.