Obviously this system would require a revision to the OpenID system, whereby the authentication provider maintains the database, and the site is allowed to query it based on an identifier, assuming the user has allowed the site to do so.
If OpenID allowed each site to assign users a reputation, and had an opt in to advertise that reputation, and sites could opt in (after receiving requests, etc) to use reputation from other sites, two sites would be allowed to peer with each other (shared rep), and there would need to be some sort of exponential scaling to adding and removing site opt-ins, and random delays on how long it would take effect (1-24 hours.)
That would prevent these abuses: * A user would not be forced to have site's that abused their reputation permanently marring their account. * A site would not be forced to allow other site's reputation affect the "other user's reputation" pool, and the opt-in prevents spam reputation providers, there would also be a facility for scaling reputation for each site. * Being only able to see the user's total "internal" and "external" rep prevents a site from naively determining the reputation a user has on another site for unknown reasons * Random delays on the time it takes to have a site opt in to use another site's user reputations would mean a site can't opt in, query all of their users, opt out, and use those scores for nefarious purposes. They are, essentially, "stuck with it."
Presumably OpenID would however allow a service such as querying the mean, standard deviation, minimum and maximum or other metrics for what another site's reputation is, that would help prevent gaming, and of course there would have to be some sort of system whereby a site can have a minimum / maximum effect on a user's reputation, in addition to the scaling factor.
You say that with such conviction, someone might be tricked into thinking it's true.
But I'm absolutely certain Microsoft not working harder with third parties (Nvidia, Creative, to name a couple) resulted in millions, maybe even a few billion dollars in lost sales for Vista. After all, Vista was death by a thousand cuts for the enterprise, and that cost them dearly. Some, not all, of the problems were not even Microsoft's fault. Lazy or unmotivated third party vendors released poor drivers despite having over a year of release candidates and betas, and continued to release these drivers in monthly or quarterly installments that resulted in marginal improvements. I think Nvidia alone accounted for nearly 1/3 of all blue screens. Microsoft released numbers documented here:
So no, I think Microsoft painfully learned that they could not rely on the OEMs to do the hardware compatibility testing "for them." And even if those crashes were encountered on primarily enthusiast PCs with high end graphics cards and sound cards, those statistics and the anecdotes about them reached businesses at the speed of rumor, and before you knew it, Vista was sunk as a business sell. It did alright, after all it still had Microsoft's force, long term support contracts meant businesses could deploy as little or as many Vista desktops as they wanted, but it didn't do great.
Hyper-V R2 has none of the limitations you mentioned, and Hyper-V's initial RTM released shortly after Server 2008 doesn't even have some of those limitations either.
High Availability: Hyper-V R2 supports live migration (no downtime, memory copy) and quick migration (hibernate & restore). As well as clustering servers to act as a pool for VMs. Same as VMWare. Though I don't think this scales as well as VMWare's products.
Drive letters: You can have more than 24VMs if you don't assign a drive letter and instead use the volume ID, or if you use a third party shared volume setup. And Hyper-V R2 adds clustered volume support so you can have unlimited VMs on the same volume on your storage, and unlimited volumes attached to your cluster.
And you're exactly right about IBM FUD, which is why I feel confident using Hyper-V because it's stupidly easy to set up and repair, and Microsoft has shoved it off on enough big businesses that by version 2 that I feel comfortable running orders of magnitude fewer VMs.
But I don't run a shop that needs truly massive scalability, and can't budget for VMWare's products. Your case differs, so your choices differed.
I can't blame you for picking the industry leader, yada yada yada.
Err, Nokia doesn't advertise the N# line of phones? Blackberry doesn't advertise their phones, especially for business? Samsung never advertised their phones?
Shoot, I think you're attributing way too much to Apple here. The app store had explosive growth because it was a smartphone that reached mass appeal with the mainstream, Gen WhateverItIs crowd. It's not like Windows Mobile phones and Blackberry phones and Symbian phones never had applications before. True, their marketing was paid for in bottom shelf vodka, and it showed.
Lastly, the FCC is getting involved because of handset tying, but also all the related practices. Why are data plans so cheap compared to text messages, for example? I can download a gigabyte or two in a month of heavy usage, the equivalent of over 7,000,000 text messages. There are numerous issues being examined by the FCC here.
Considering Hyper-V R2 is free, I don't think it's appropriate to even refer to sales.
I mean, it's not only free, but it does out of the box more than any free VMWare hypervisor does, and it's a heck of a lot easier than it currently is to set up a high availability VM cluster on Linux.
Fortunately the commerce clause really does cover something like the large inter-state communication networks, and allows the FCC to prevent one ISP from implementing draconian provisions that affect all of us.
It proves that an fMRI, like most machines, needs to be carefully operated and the mechanisms understood, as there are risks of false positives for results.
The paper is about intentionally observing a dead creature, and coming across a few false positives and why that happened.
The reason it's being investigating is because Apple's iPhone is exclusive to AT&T, and they're investigating the matters of handset exclusivity and long contracts and a number of other industry practices that may not be in the best interest of the consumer, and can prevent competition.
Now normally that would be OK. We're not talking about monopolies, but the reason it's not here is because AT&T and all the other mobile vendors are using our spectrum. The Federal government has licensed the spectrum to them to benefit us, and when they are doing things with their business to prevent us from using that spectrum in the way, or with the device we want on the network we want, then it is a problem.
When it is demonstrably easy to switch a jailbroken iPhone from AT&T to T-Mobile, then the FCC has proof that the exclusivity is solely about maintaining market dominance. When the handset manufacturer, now tied to the dominant market player, is arbitrarily rejecting apps, particularly apps from companies that they are in some form of competition with, then the FCC has connected all the dots from Google, to Apple, to AT&T, and is wondering wtf is going on with the spectrum they licensed for the good of the people.
So you get 4 of these things and you RAID5 the 4 of them. So it's actually RAID50, but c'est la vie.
Think of it like the SSD being itself a RAID array, and you can just RAID it like normal with other SSDs. Duh. It's exactly what IT admins have been doing. For bulk data or large writes use RAID5/RAID6, for database IO use RAID10. Don't concern yourself with what the SSD is doing. It may be more or less reliable than a hard drive. Probably more. But ignore that, treat it like a regular hard drive, just really fast.
Due to the 8/10 encoding on SATA, SAS, and a few other serial technologies, it's really easy to convert between megabits/gigabits of total bandwidth and megabits/gigabits of encoded bandwidth. For SATA/SAS 3Gib/s, it's 300MiB/s. For 6Gib/s, it's 600MiB/s.
As a user of Windows 7, I found it exceedingly helpful. I was pleased when Clippy popped up and said, "It looks like you're trying to infect your computer, do you want some help?" At which point Clippy showed me how to use Aero Shake(tm) to get rid of all the distracting popups that would divert me from trying to find the source of all malware. After I encountered a fork in the road, so to speak, Clippy demonstrated Aero Snap(tm) so I could compare the sites I was surfing side by side. At long last, I found truly good malware on a *stan website. Top level domain was for some country like Miyagistan. Thankfully, I bought Windows(tm) 7 Ultimate Edition(tm) and downloaded the appropriate language pack so the viruses I downloaded would be more at home.
Running it was as easy as clicking on it and clicking "Continue." Ever since then I've been living in a peaceful coexist
No, if you construct something from copyright, then that is a derivative work as well and infringing.
If I buy from you a book, and make a copy of said book, that is OK. But if I then sell the copy, then I have infringed on your copyright license to me.
Ok, rather than delve into SMP, NUMA, and cache sizes, just from one point of view there is a huge difference between various environments. In a phone and a desktop, you want latency reduced as much as possible on any foreground operation. In a similar way, a cluster of lightweight webservers would want to minimize average latency to respond to requests.
On the other hand, a database server, virtual machine server handling a variety of workloads, a beefy dual processor workstation, etc, are all going to have different needs.
At least in the Microsoft camp, I'm fairly sure different schedulers are used for the desktop, server, and cluster areas.
IMO, if Linux wants to have its year, it needs to accept that when I come home and take off my IT admin/programmer/developer cap and put on my home user/web browser/video game playing cap, I have different needs. I want different things from my OS.
One thing Rock Band has done, and presumably this came from somewhere else or has propagated to Guitar Hero and other rhythm games, is that you can set the video latency and audio latencies separately and finely tune the system so that it looks and sounds like you want it to be.
Rock Band 2's guitar controller actually has a tiny light sensitive component and a cheap microphone, so that you can auto-set your game. It's really very handy, and took only fifteen seconds or so. The result was that when a note crosses the "active line" of the game is when I should both strum it / hit it / sing it and hear the result.
Are you certain there is no way to do the same thing with DDR?
And it does get even worse, you have to remember there are smartphones and embedded chips running Linux, and there are very large many-many core machines running it as well. Intel recently demonstrated a 128 core computer using 16 of their new Nehalem-EX chips. With hyperthreading, Windows Server or Linux would show 256 CPU graphs.
Clearly my phone should not run the same scheduler my desktop does, and my desktop has different needs than a middle of the road server, and that server's needs may differ enormously from the needs of that 128 core monster.
Again, this is just adding emphasis to the argument that pluggable schedulers are a Good Thing(TM).
Corporations tend to do slimy things for the sake of profit, but a lot of these allegations are tenuous at best. Hey, I'd love to see proof OOXML was bribed into being a standard, to see its ISO certification disappear. It is a nasty, long, dense standard that doesn't help anyone but them. It'd be great if instead they contributed vital enhancements to ODF.
On the other hand, they're on the receiving end of a truly absurd amount of FUD at Slashdot.
They -are- a patent troll. Just like any other company that sues for obvious, trivial software patents and non-inventions. They're all seeking to make money by extortion or lawsuit for something that really isn't patent-worthy, hence: patent troll.
At times, even Microsoft has been a patent troll, so have many other companies. Stop treating everyone differently with these trivial patents, rather than playing favorites just because a company you don't like is involved.
So? Almost all software patents have implementations. That doesn't make them legit.
I don't get why Slashdot screams bloody murder when software patents are brought up in conversation about open source products, but when Microsoft or one of the "bad guys" is on the receiving end of a lawsuit it's A-OK with you guys.
You are the cancer that is killing/., is what I'm trying to say. They aren't a patent troll? Come on, grow up. You can be a patent troll with a product, just like all those other software patent trolls that have a product, or have licensed it at some point, or whatever. If having patents and a product means you're not a patent troll, then I guess that mean's Google's patent on their home page is legit? Or IBM's patent for a business process to produce patents, or whatever it was? Because they use both of those, obviously, so that means if they sue someone for it, they're clearly not trolls?
That's what your logic works out to. How about we adjust your definition of patent troll so you can stop apologizing for i4i. New definition of patent troll: a company or individual who uses patents that are not legitimate non-obvious, novel inventions in order to acquire money by suing whomever looks like a good target, usually in an East Texas court room.
Err, sorry, I mean, do you have any EVIDENCE for any of that?
Why is it OK to be against software patents on one day, except when it's someone you despise on the wrong end of a lawsuit? What makes this patent holier than all the rest, and why do you think OpenOffice and other programs would remain safe? i4i's promise not to sue Sun later on is not binding, and there are only half a million projects out there that use XML and support custom schemas. Fuck, Word is just the tip of the iceberg here. Microsoft's successor to the GDI framework, WPF, uses XML and supports custom schemas defined by the user (or in this case, developer.) I can add a element for example, with custom binding, properties, etc. Is that infringing?
Help me make sense of this. Where does this insanity end?
Slashdot Mirror
11.5 million. Wait, err... ;)
Obviously this system would require a revision to the OpenID system, whereby the authentication provider maintains the database, and the site is allowed to query it based on an identifier, assuming the user has allowed the site to do so.
If OpenID allowed each site to assign users a reputation, and had an opt in to advertise that reputation, and sites could opt in (after receiving requests, etc) to use reputation from other sites, two sites would be allowed to peer with each other (shared rep), and there would need to be some sort of exponential scaling to adding and removing site opt-ins, and random delays on how long it would take effect (1-24 hours.)
That would prevent these abuses:
* A user would not be forced to have site's that abused their reputation permanently marring their account.
* A site would not be forced to allow other site's reputation affect the "other user's reputation" pool, and the opt-in prevents spam reputation providers, there would also be a facility for scaling reputation for each site.
* Being only able to see the user's total "internal" and "external" rep prevents a site from naively determining the reputation a user has on another site for unknown reasons
* Random delays on the time it takes to have a site opt in to use another site's user reputations would mean a site can't opt in, query all of their users, opt out, and use those scores for nefarious purposes. They are, essentially, "stuck with it."
Presumably OpenID would however allow a service such as querying the mean, standard deviation, minimum and maximum or other metrics for what another site's reputation is, that would help prevent gaming, and of course there would have to be some sort of system whereby a site can have a minimum / maximum effect on a user's reputation, in addition to the scaling factor.
You say that with such conviction, someone might be tricked into thinking it's true.
But I'm absolutely certain Microsoft not working harder with third parties (Nvidia, Creative, to name a couple) resulted in millions, maybe even a few billion dollars in lost sales for Vista. After all, Vista was death by a thousand cuts for the enterprise, and that cost them dearly. Some, not all, of the problems were not even Microsoft's fault. Lazy or unmotivated third party vendors released poor drivers despite having over a year of release candidates and betas, and continued to release these drivers in monthly or quarterly installments that resulted in marginal improvements. I think Nvidia alone accounted for nearly 1/3 of all blue screens. Microsoft released numbers documented here:
http://www.crn.com/hardware/206905475;jsessionid=MNAOZ34HU4KWHQE1GHPCKH4ATMY32JVN
So no, I think Microsoft painfully learned that they could not rely on the OEMs to do the hardware compatibility testing "for them." And even if those crashes were encountered on primarily enthusiast PCs with high end graphics cards and sound cards, those statistics and the anecdotes about them reached businesses at the speed of rumor, and before you knew it, Vista was sunk as a business sell. It did alright, after all it still had Microsoft's force, long term support contracts meant businesses could deploy as little or as many Vista desktops as they wanted, but it didn't do great.
Hyper-V R2 has none of the limitations you mentioned, and Hyper-V's initial RTM released shortly after Server 2008 doesn't even have some of those limitations either.
High Availability: Hyper-V R2 supports live migration (no downtime, memory copy) and quick migration (hibernate & restore). As well as clustering servers to act as a pool for VMs. Same as VMWare. Though I don't think this scales as well as VMWare's products.
Drive letters: You can have more than 24VMs if you don't assign a drive letter and instead use the volume ID, or if you use a third party shared volume setup. And Hyper-V R2 adds clustered volume support so you can have unlimited VMs on the same volume on your storage, and unlimited volumes attached to your cluster.
And you're exactly right about IBM FUD, which is why I feel confident using Hyper-V because it's stupidly easy to set up and repair, and Microsoft has shoved it off on enough big businesses that by version 2 that I feel comfortable running orders of magnitude fewer VMs.
But I don't run a shop that needs truly massive scalability, and can't budget for VMWare's products. Your case differs, so your choices differed.
I can't blame you for picking the industry leader, yada yada yada.
Err, Nokia doesn't advertise the N# line of phones? Blackberry doesn't advertise their phones, especially for business? Samsung never advertised their phones?
Shoot, I think you're attributing way too much to Apple here. The app store had explosive growth because it was a smartphone that reached mass appeal with the mainstream, Gen WhateverItIs crowd. It's not like Windows Mobile phones and Blackberry phones and Symbian phones never had applications before. True, their marketing was paid for in bottom shelf vodka, and it showed.
Lastly, the FCC is getting involved because of handset tying, but also all the related practices. Why are data plans so cheap compared to text messages, for example? I can download a gigabyte or two in a month of heavy usage, the equivalent of over 7,000,000 text messages. There are numerous issues being examined by the FCC here.
Wow, you make my time sound so cheap when you put it like that.
Considering Hyper-V R2 is free, I don't think it's appropriate to even refer to sales.
I mean, it's not only free, but it does out of the box more than any free VMWare hypervisor does, and it's a heck of a lot easier than it currently is to set up a high availability VM cluster on Linux.
Fortunately the commerce clause really does cover something like the large inter-state communication networks, and allows the FCC to prevent one ISP from implementing draconian provisions that affect all of us.
It proves that an fMRI, like most machines, needs to be carefully operated and the mechanisms understood, as there are risks of false positives for results.
The paper is about intentionally observing a dead creature, and coming across a few false positives and why that happened.
The reason it's being investigating is because Apple's iPhone is exclusive to AT&T, and they're investigating the matters of handset exclusivity and long contracts and a number of other industry practices that may not be in the best interest of the consumer, and can prevent competition.
Now normally that would be OK. We're not talking about monopolies, but the reason it's not here is because AT&T and all the other mobile vendors are using our spectrum. The Federal government has licensed the spectrum to them to benefit us, and when they are doing things with their business to prevent us from using that spectrum in the way, or with the device we want on the network we want, then it is a problem.
When it is demonstrably easy to switch a jailbroken iPhone from AT&T to T-Mobile, then the FCC has proof that the exclusivity is solely about maintaining market dominance. When the handset manufacturer, now tied to the dominant market player, is arbitrarily rejecting apps, particularly apps from companies that they are in some form of competition with, then the FCC has connected all the dots from Google, to Apple, to AT&T, and is wondering wtf is going on with the spectrum they licensed for the good of the people.
So you get 4 of these things and you RAID5 the 4 of them. So it's actually RAID50, but c'est la vie.
Think of it like the SSD being itself a RAID array, and you can just RAID it like normal with other SSDs. Duh. It's exactly what IT admins have been doing. For bulk data or large writes use RAID5/RAID6, for database IO use RAID10. Don't concern yourself with what the SSD is doing. It may be more or less reliable than a hard drive. Probably more. But ignore that, treat it like a regular hard drive, just really fast.
Due to the 8/10 encoding on SATA, SAS, and a few other serial technologies, it's really easy to convert between megabits/gigabits of total bandwidth and megabits/gigabits of encoded bandwidth. For SATA/SAS 3Gib/s, it's 300MiB/s. For 6Gib/s, it's 600MiB/s.
As a user of Windows 7, I found it exceedingly helpful. I was pleased when Clippy popped up and said, "It looks like you're trying to infect your computer, do you want some help?" At which point Clippy showed me how to use Aero Shake(tm) to get rid of all the distracting popups that would divert me from trying to find the source of all malware. After I encountered a fork in the road, so to speak, Clippy demonstrated Aero Snap(tm) so I could compare the sites I was surfing side by side. At long last, I found truly good malware on a *stan website. Top level domain was for some country like Miyagistan. Thankfully, I bought Windows(tm) 7 Ultimate Edition(tm) and downloaded the appropriate language pack so the viruses I downloaded would be more at home.
Running it was as easy as clicking on it and clicking "Continue." Ever since then I've been living in a peaceful coexist
Copying the full book as long as I own and keep all copies is legitimate.
No, if you construct something from copyright, then that is a derivative work as well and infringing.
If I buy from you a book, and make a copy of said book, that is OK. But if I then sell the copy, then I have infringed on your copyright license to me.
Ok, rather than delve into SMP, NUMA, and cache sizes, just from one point of view there is a huge difference between various environments. In a phone and a desktop, you want latency reduced as much as possible on any foreground operation. In a similar way, a cluster of lightweight webservers would want to minimize average latency to respond to requests.
On the other hand, a database server, virtual machine server handling a variety of workloads, a beefy dual processor workstation, etc, are all going to have different needs.
At least in the Microsoft camp, I'm fairly sure different schedulers are used for the desktop, server, and cluster areas.
IMO, if Linux wants to have its year, it needs to accept that when I come home and take off my IT admin/programmer/developer cap and put on my home user/web browser/video game playing cap, I have different needs. I want different things from my OS.
One thing Rock Band has done, and presumably this came from somewhere else or has propagated to Guitar Hero and other rhythm games, is that you can set the video latency and audio latencies separately and finely tune the system so that it looks and sounds like you want it to be.
Rock Band 2's guitar controller actually has a tiny light sensitive component and a cheap microphone, so that you can auto-set your game. It's really very handy, and took only fifteen seconds or so. The result was that when a note crosses the "active line" of the game is when I should both strum it / hit it / sing it and hear the result.
Are you certain there is no way to do the same thing with DDR?
Do you offer a No Kids On Lawn guarantee while they're away? It's very important.
And it does get even worse, you have to remember there are smartphones and embedded chips running Linux, and there are very large many-many core machines running it as well. Intel recently demonstrated a 128 core computer using 16 of their new Nehalem-EX chips. With hyperthreading, Windows Server or Linux would show 256 CPU graphs.
Clearly my phone should not run the same scheduler my desktop does, and my desktop has different needs than a middle of the road server, and that server's needs may differ enormously from the needs of that 128 core monster.
Again, this is just adding emphasis to the argument that pluggable schedulers are a Good Thing(TM).
Corporations tend to do slimy things for the sake of profit, but a lot of these allegations are tenuous at best. Hey, I'd love to see proof OOXML was bribed into being a standard, to see its ISO certification disappear. It is a nasty, long, dense standard that doesn't help anyone but them. It'd be great if instead they contributed vital enhancements to ODF.
On the other hand, they're on the receiving end of a truly absurd amount of FUD at Slashdot.
They -are- a patent troll. Just like any other company that sues for obvious, trivial software patents and non-inventions. They're all seeking to make money by extortion or lawsuit for something that really isn't patent-worthy, hence: patent troll.
At times, even Microsoft has been a patent troll, so have many other companies. Stop treating everyone differently with these trivial patents, rather than playing favorites just because a company you don't like is involved.
So? Almost all software patents have implementations. That doesn't make them legit.
I don't get why Slashdot screams bloody murder when software patents are brought up in conversation about open source products, but when Microsoft or one of the "bad guys" is on the receiving end of a lawsuit it's A-OK with you guys.
You are the cancer that is killing /., is what I'm trying to say. They aren't a patent troll? Come on, grow up. You can be a patent troll with a product, just like all those other software patent trolls that have a product, or have licensed it at some point, or whatever. If having patents and a product means you're not a patent troll, then I guess that mean's Google's patent on their home page is legit? Or IBM's patent for a business process to produce patents, or whatever it was? Because they use both of those, obviously, so that means if they sue someone for it, they're clearly not trolls?
That's what your logic works out to. How about we adjust your definition of patent troll so you can stop apologizing for i4i. New definition of patent troll: a company or individual who uses patents that are not legitimate non-obvious, novel inventions in order to acquire money by suing whomever looks like a good target, usually in an East Texas court room.
Does that sound like i4i to you?
Evidence?
Err, sorry, I mean, do you have any EVIDENCE for any of that?
Why is it OK to be against software patents on one day, except when it's someone you despise on the wrong end of a lawsuit? What makes this patent holier than all the rest, and why do you think OpenOffice and other programs would remain safe? i4i's promise not to sue Sun later on is not binding, and there are only half a million projects out there that use XML and support custom schemas. Fuck, Word is just the tip of the iceberg here. Microsoft's successor to the GDI framework, WPF, uses XML and supports custom schemas defined by the user (or in this case, developer.) I can add a element for example, with custom binding, properties, etc. Is that infringing?
Help me make sense of this. Where does this insanity end?
If you right click a file in Windows and go to Properties you see:
Size: 2.47 KB (2,539 bytes)
Size on disk: 4.00 KB (4,096 bytes)
I thought Mac OS X was supposed to be easy?