Let's start fresh! SATA, PCI-Express, USB2.0! Time for a clean break! Get rid of all the legacy crap. We're supposed to upgrade every three years anyway, so let's really upgrade.
IBM got the contract out of part of the buyout. It was probably never looked at. They would have assumed that IBM had control of derivative works, and SCO is saying that they do.
I think that however this case turns out, it has much more to do with IP and ownership than any of us first realized, and, without the presiding judge following the exact letter of the law and not the spirit of it, IBM could very well destroy the "Transferrability of IP" in the common market. That would be fine by me.
It gets worse. At what point do the developers become 'infected'? IBM can't very well replace the code, because the developers know too much AIX. Linus can't replace the code, because he's seen the original, and any changes that he makes can be considered derivitave works. Hell, I can't even replace the RCU code, because the/. editors placed a link in the article which I blindly followed and have now become tainted as well. Basically, according to the licensing, we have to shoot all people who have used linux, unix, and possibly even solaris, burn all record of Unix ever happening, and train new developers without the teachers having any prior knowledge of programming, using no resources currently in print.
If that's not viral, I don't know what is. It's a slash-and-burn to say the least!
It could scale, but not as effectively as Unix. That's where the RCU code comes into play. Without the RCU code, the bottlenecks barely show on 4-way systems, but begin compounding after that. From the RCU Article linked in the story:
"Using RCU it is possible to provide some relief from acquiring dcache_lock. The main work of dentry cache is done in looking up existing dentries in the cache which is done by d_lookup routine. Using RCU (and lazy-lru algorithm), we could do lock-less lookup for dentries and bring down the contention for dcache_lock while running dbench considerably from 16.5% to 0.95% on an 8-way SMP box. Recent measurements by Anton showed > 20% improvement in dbench throughput for a 32-way PPC machine."
So even though a standard SMP kernel pre-patch could do SMP, the (allegedly)Sco owned code made the kernel 'Enterprise Ready'
SCO claims that the SMP jumped from 4 processors to 64, something that the linux community could not do on their own, mostly because none of us could afford a 64 proc machine (I know that I can't). So in order for it to run on it, somebody who has access to a 64 proc machine and a lot of experience with Unix source for said machine had to put up the work.
Sco says that IBM did it, and did it illegally.
One CD? That better come with a helluva lot of pictures on it, because there is absolutely no way they can fill the thing even if they had full PDF and MS DOC versions of each program, with i18n versions of each.
I thought that only AOL gave out free coasters.
Re:MOD THIS WHORE DOWN! NOW!
on
Sun's Last Stand
·
· Score: 0, Offtopic
It only gets modded up if there's references to Cmdr Taco and gay activities.
If it's the same article that I'm thinking of, then it's the exact opposite of what I want to do. I think that they were using open standards to replace the exchange functionality, whereas I want to use the exchange functionality to let non-exchange clients work with an exchange server, without the use of the OWA.
Microsoft is free to compete on the merits of their product, not on the secrecy of their protocols.
Ahhh, I wish you were the DOJ. Your rational reasoning goes against what all the large comapnies have fought so hard for.
In an ideal world, your statement would be accurate. Unfortunately we no longer live in a world, we live in a giant department store.
The DOJ settlement does allow for this. The problem does not lie in the fact that they are charging for the protocol inspection, but there are further terms which could be considered discriminatory. For instance, the independant audit at the licensee's expense with no assurance that the inspector is not reporting back extra information to Microsoft, or the severely crippling NDA which prompted one executive to say that "Basically, I'd have to shoot the engineers when they came back,". Now those are what the DOJ has to center in on, not some admission ticket price.
Usually, I'm an anti-MS zealot, but this time I agree with you. It's not a fee, it's a deterrent. I would love to get a copy of MS's protocols so I could write a proper exchange connector for unix. But I don't have a hundred grand to pony up, so it ain't gonna happen.
This isn't unfair competitive practices, this is competitive practices designed to protect their trade secrets. They are essentially showing the holy grail of linux computing, so why shouldn't they charge an admission fee?
Bounties are usually like 2 to 10 percent of what the guy owes, although that usally only applies to bail bonds. If it were me, I'd bump the ante up to a cool ten million, as long as I could actually get at least $20 million from him. But that means that he'd have to be brought back alive...
Just replace it with a real mirror. By the time they figure out that it's not just broken, you're home free.
As a side note, they will charge you for towels. Lightbulbs however, they tend to overlook. Same goes for shower heads and the batteries out of the remote.
(thread of what you've stolen out of hotel rooms starts here)
Slashdot Mirror
Hmm, seing 'We Were Soldiers' with all the violence removed? I'd have only wasted half an hour of my life instead of two.
I'm all for it!
If they tightly integrate their search engine with Windows while making it work well, things might change.
You mean things will stay the same.
We bitched about IE being a core part of the OS, now we get to bitch that MSN is. Antitrust lawsuits, here we come again!
Hell, I'm still using a celeron 450 with 256MB RAM, and it's still holding up to what I need it to do.
Let's start fresh! SATA, PCI-Express, USB2.0! Time for a clean break! Get rid of all the legacy crap. We're supposed to upgrade every three years anyway, so let's really upgrade.
Oh Shit, we all have iMac's now...
IBM got the contract out of part of the buyout. It was probably never looked at. They would have assumed that IBM had control of derivative works, and SCO is saying that they do.
I think that however this case turns out, it has much more to do with IP and ownership than any of us first realized, and, without the presiding judge following the exact letter of the law and not the spirit of it, IBM could very well destroy the "Transferrability of IP" in the common market. That would be fine by me.
See ya Mickey, you just got stepped on by IBM.
It gets worse. At what point do the developers become 'infected'? IBM can't very well replace the code, because the developers know too much AIX. Linus can't replace the code, because he's seen the original, and any changes that he makes can be considered derivitave works. Hell, I can't even replace the RCU code, because the /. editors placed a link in the article which I blindly followed and have now become tainted as well. Basically, according to the licensing, we have to shoot all people who have used linux, unix, and possibly even solaris, burn all record of Unix ever happening, and train new developers without the teachers having any prior knowledge of programming, using no resources currently in print.
If that's not viral, I don't know what is. It's a slash-and-burn to say the least!
It could scale, but not as effectively as Unix. That's where the RCU code comes into play. Without the RCU code, the bottlenecks barely show on 4-way systems, but begin compounding after that. From the RCU Article linked in the story: "Using RCU it is possible to provide some relief from acquiring dcache_lock. The main work of dentry cache is done in looking up existing dentries in the cache which is done by d_lookup routine. Using RCU (and lazy-lru algorithm), we could do lock-less lookup for dentries and bring down the contention for dcache_lock while running dbench considerably from 16.5% to 0.95% on an 8-way SMP box. Recent measurements by Anton showed > 20% improvement in dbench throughput for a 32-way PPC machine." So even though a standard SMP kernel pre-patch could do SMP, the (allegedly)Sco owned code made the kernel 'Enterprise Ready'
SCO claims that the SMP jumped from 4 processors to 64, something that the linux community could not do on their own, mostly because none of us could afford a 64 proc machine (I know that I can't). So in order for it to run on it, somebody who has access to a 64 proc machine and a lot of experience with Unix source for said machine had to put up the work. Sco says that IBM did it, and did it illegally.
When will they get their letter?
As long as they don't touch my car...
I use my garage for something better than that...
I park my car in it!
Nah, that was all the devious work of the egg council, as a disinformation ploy.
One for you, and one for me. Two for you...
</Homer>
One CD? That better come with a helluva lot of pictures on it, because there is absolutely no way they can fill the thing even if they had full PDF and MS DOC versions of each program, with i18n versions of each.
I thought that only AOL gave out free coasters.
It only gets modded up if there's references to Cmdr Taco and gay activities.
:)
You must be new here
If it's the same article that I'm thinking of, then it's the exact opposite of what I want to do. I think that they were using open standards to replace the exchange functionality, whereas I want to use the exchange functionality to let non-exchange clients work with an exchange server, without the use of the OWA.
Microsoft is free to compete on the merits of their product, not on the secrecy of their protocols.
Ahhh, I wish you were the DOJ. Your rational reasoning goes against what all the large comapnies have fought so hard for.
In an ideal world, your statement would be accurate. Unfortunately we no longer live in a world, we live in a giant department store.
The DOJ settlement does allow for this. The problem does not lie in the fact that they are charging for the protocol inspection, but there are further terms which could be considered discriminatory. For instance, the independant audit at the licensee's expense with no assurance that the inspector is not reporting back extra information to Microsoft, or the severely crippling NDA which prompted one executive to say that "Basically, I'd have to shoot the engineers when they came back,". Now those are what the DOJ has to center in on, not some admission ticket price.
Usually, I'm an anti-MS zealot, but this time I agree with you. It's not a fee, it's a deterrent. I would love to get a copy of MS's protocols so I could write a proper exchange connector for unix. But I don't have a hundred grand to pony up, so it ain't gonna happen.
This isn't unfair competitive practices, this is competitive practices designed to protect their trade secrets. They are essentially showing the holy grail of linux computing, so why shouldn't they charge an admission fee?
Bounties are usually like 2 to 10 percent of what the guy owes, although that usally only applies to bail bonds. If it were me, I'd bump the ante up to a cool ten million, as long as I could actually get at least $20 million from him. But that means that he'd have to be brought back alive...
Man, now that's justice!
If only we could do the same thing to spammers, instead of suing them for $1000.00 per UCE.
Just send an anonymous tip to SCO. We'll find out in court...
1. Please put your request in writing or in an email addressed to info@Linksys.com
I think that you forgot something, or the post office got way more efficient.
Just replace it with a real mirror. By the time they figure out that it's not just broken, you're home free.
As a side note, they will charge you for towels. Lightbulbs however, they tend to overlook. Same goes for shower heads and the batteries out of the remote.
(thread of what you've stolen out of hotel rooms starts here)
Ahhh, that made my night. Thanks.