Real science fiction fans deride the use of Sci-Fi as a moniker, I guess it is becoming obvious why. Soap operas in space are not science fiction, 900 number reject asshole "psychics" that scam old people are not science fiction.
Science fiction is about expanding the way we thing about the present, by showing us probable or possible futures, based on the science and technology we currently have. It only barely encompasses the "supernatural", only insofar as the "supernatural" is just science we do not yet understand, which, by definition isn't really supernatural.
perhaps this will lead people to do the same in future instances for money-calls. This shows it can be done.
Can be done under the right circumstances. I think that if someone came along and ran a campaign just like Blender's with the exact same software demand as Blender, and just as many followers as Blender, it wouldn't be as sucessful.
I think a lot of people gave money to this because it was novel, and wanted to prove a point. I know that played a part in my dontations, I suck at art, and my pitiful attempts at 3D art years ago on my 486 are all I have ever touched as far as 3D art goes. I think I made a sphere hovering over a checkerboard floor. It rocked. heh
Anyway, sure, given a proper cooling-off period for people to get some more charity money together, and a suficiently popular software product, it could be done again. In the grand scheme of things though, 100K isn't much money at all. It was a real bargain, and it basically says NaN was giving up hope on making Blender a closed source commercial sucess.
It's still much better than the code rotting away in some company's vaults for no reason.
I think that is the good precedent, that software that is deemed worthless to a company might be sold to open source as it's last stop before EOL. That does an end-run around the insane copyright laws that lock software up forever (100 years is forever in computer terms), even after it has ended its useful life for the copyright holder.
So to sum up, I agree with you, with caveats. If every company makes a run to sell their older software in the next couple months, it won't work, but a small trickle of software with a large enough following, spaced out sufficiently, could really start building the "GNU domain" (surrogate for public domain, since copyright law is so fucked up)
That buffer may not exist. The "Pending" were people that said they were going to pay, but were supposedly waiting on their Pay Pal account to go through. Ton is cancelling all pending transactions that are more than 4 weeks old. I wouldn't be surprised if that $8000 turned out to be more like $1000.
And of course it runs over fiber optic cable, which isn't exactly penny kit. So the industry decided to try running it over Ethernet.
I agree that iSCSI is kind of a stupid idea, but all the fibre channel I have ever worked with is copper, two pair, that terminates in a D-sub connector with four pins. I didn't think anyone bothered running it over fiber anymore.
A lot of mini-ITX motherboards have composite or S-Video out, however, they also have Macrovision, so you need to spend about the same price as the motherboard on a stabilizer box to remove this intentional crippling.
Even worse, nothing may exist at all! After all, your complete sensory input comes from only a 5 or 6 foot large object that you seem to have control over. It's entirely possible there is some "Evil Deceiver" that is tricking you into thinking it's all real!
I wonder why no one has thought of this before? Someone should write a treatise or something.
Or, the other case I see as probably being common, they are adding another new Linux server, but keeping their legacy Windows or Novell system around because it works for them, and just running both of them, waiting until they have time to finish the migration and totally ditch the old server.
That is the current situation at my workplace (Novell), and I know we aren't alone. These phased migrations are hard to measure statistically, since there isn't a trackable event (like a purchase) when a company decides to finally ditch the old system.
"For those that have recently purchased new Linux servers, 31 percent were adding capacity, 31 percent were replacing Windows systems, 24 percent were replacing Unix and 14 percent were replacing other operating systems."
Bad at math? It's OK, I heard on CNN that 50% of all people are below average in math.
In any case:
From those numbers, 69% of new installations were migrations, the rest were "adding capacity".
31% of total new installations were replacing legacy Windows systems.
31%/69% is about 45%. 45% of all migrations to Linux were from legacy Windows systems.
Makes external boxes that you hook to a normal SCSI controller card, comes in the form of boxes that can take ATA drives, or SCSI drives. Also makes fibre channel boxes of a similar nature. Appears as one large SCSI drive to the host OS, compatible with basically any OS.
Low end crappy standalone ATA-to-SCSI boxes, similar to the above ones, also makes very crappy contoller cards, only useful if you are using them for software RAID, don't use thier hardware RAID for anything. Promise cards are also picky about the BIOS on the motherboard they are installed on. Their standalone box prices are way overpriced for what you get. Their controller cards are cheap, but acceptable, for software RAID. More than one 6 channel controller per computer is not supported, more than three two channel cards is not supported. Linux kernel module is mature though.
3ware makes hardware ATA RAID controllers that are very fast, and relatively expensive. "Unlimited" number of controllers per computer, I've ran up to four 8-port cards in a single computer. Cabling is a mess when you get a lot of drives in a single system, if you need that many, seriously consider one of the above standalone boxes. Linux kernel module is open source and vendor maintained. Management software for Linux is free but closed source.
----------------- Of course, Linux has software RAID built right in, and with a distro like Red Hat, you can set up software RAID when you install the OS in a simple GUI. You can use whatever disks you have installed, but for good results, you need to only have one disk per ATA channel, be it on your motherboard or a Promise card.
We currently have 56 Maxtor drives in various RAIDs where I work. Some of them are coming up on a year old, with no failures yet (100, 120, 160GB). Of course, the fault tolerance that is built in to our systems can withstand quite a few failures before we would be in any trouble (up to half), and even then there are tapes behind most of it.
Our G4 Macs that have IBM drives though, those are dropping left and right, and most of those are less than a year old.
You're telling me... I tried to order a ATA-SCSI box full of (14) 160GB drives, and the assholes priced it so the drives were about $500 a piece, compared to the price of the box empty. He basically told me "Good Luck finding them on your own".
I guess we will just get the empty box and scrounge around until we can find 160s. I bet SCSI is hurting, considering how Maxtor can't keep the ATAs on the shelf they are selling so fast. With the ATA-SCSI switched bus RAID boxes being so cheap, so much faster than a shared bus SCSI arrangement, and just as reliable, it's no wonder they are selling out.
For the record, since the original poster seems confused (parent of parent),
Maxtor currently sells: 100, 120, 160GB 5400 RPM drives
WD is *supposed* to be selling: 180, 200GB 7200 RPM drives, but I havn't seen any for sale yet.
Maxtor 160GB street retail price around $240, WD 200GB MSRP at $400
The Maxtor 160s have been out for over 6 months now, WD just released their high capacity last month.
If intellectual property laws are repealed, then Linux as we know it is doomed.
Who said anything about repealing anything completely?
And Linux's only advantages-- better stability and security--are only as safe as its code base.
If IP laws were repealed, GNU/Linux could freely take whatever technology that was formerly covered by patent or copyright, the advantages of open source would be even more apparent, since companies that rely on IP only for income (otherwise known as leeches, or Rambus), will go out of business, leaving only the people who are really producing non-IP value to make the money.
With LEO sats, you could get that down to something more reasonable. My current satellite service gets minimum 600ms latency real world, in geostationary orbit, 35,000km or so up. LEO sats can be much much lower, 300-800km. At LEO, the latency is no longer an issue, in fact, you might get better latency than land lines under some circumstances!
Ebay offers escrow also, which would infringe on that most likely. Doesn't necessarily mean it will stand up in the end. Escrow isn't exactly a non-obvious innovation without prior art.
Slashdot Mirror
John Edward -- All fiction and no science.
Real science fiction fans deride the use of Sci-Fi as a moniker, I guess it is becoming obvious why. Soap operas in space are not science fiction, 900 number reject asshole "psychics" that scam old people are not science fiction.
Science fiction is about expanding the way we thing about the present, by showing us probable or possible futures, based on the science and technology we currently have. It only barely encompasses the "supernatural", only insofar as the "supernatural" is just science we do not yet understand, which, by definition isn't really supernatural.
Do you really want the FCC regulating the Internet? Be very careful what you wish for man.
perhaps this will lead people to do the same in future instances for money-calls. This shows it can be done.
Can be done under the right circumstances. I think that if someone came along and ran a campaign just like Blender's with the exact same software demand as Blender, and just as many followers as Blender, it wouldn't be as sucessful.
I think a lot of people gave money to this because it was novel, and wanted to prove a point. I know that played a part in my dontations, I suck at art, and my pitiful attempts at 3D art years ago on my 486 are all I have ever touched as far as 3D art goes. I think I made a sphere hovering over a checkerboard floor. It rocked. heh
Anyway, sure, given a proper cooling-off period for people to get some more charity money together, and a suficiently popular software product, it could be done again. In the grand scheme of things though, 100K isn't much money at all. It was a real bargain, and it basically says NaN was giving up hope on making Blender a closed source commercial sucess.
It's still much better than the code rotting away in some company's vaults for no reason.
I think that is the good precedent, that software that is deemed worthless to a company might be sold to open source as it's last stop before EOL. That does an end-run around the insane copyright laws that lock software up forever (100 years is forever in computer terms), even after it has ended its useful life for the copyright holder.
So to sum up, I agree with you, with caveats. If every company makes a run to sell their older software in the next couple months, it won't work, but a small trickle of software with a large enough following, spaced out sufficiently, could really start building the "GNU domain" (surrogate for public domain, since copyright law is so fucked up)
That buffer may not exist. The "Pending" were people that said they were going to pay, but were supposedly waiting on their Pay Pal account to go through. Ton is cancelling all pending transactions that are more than 4 weeks old. I wouldn't be surprised if that $8000 turned out to be more like $1000.
like 30 wayans brothers shows 24/7
Doesn't that describe the current WB?
since 6 pages of legalese in the GPL doesn't add up to free beer OR free speech.
The bill of rights is about a page long, the constitution is several pages long.
Sometimes you need to define some terms of freedom to ensure its survival.
He went back into Taco's mind, where he came from.
Yeah, don't pick on those poor SS officers, they were just following orders!
I think you meant NAS. SAN is not really anything like what you describe.
And of course it runs over fiber optic cable, which isn't exactly penny kit. So the industry decided to try running it over Ethernet.
I agree that iSCSI is kind of a stupid idea, but all the fibre channel I have ever worked with is copper, two pair, that terminates in a D-sub connector with four pins. I didn't think anyone bothered running it over fiber anymore.
A lot of mini-ITX motherboards have composite or S-Video out, however, they also have Macrovision, so you need to spend about the same price as the motherboard on a stabilizer box to remove this intentional crippling.
Promise sells such small boxes. Overpriced in my opinion though.
[sic] is viewed as elitist by many. It's usually used when a reporter wants to make someone look stupid.
Even worse, nothing may exist at all! After all, your complete sensory input comes from only a 5 or 6 foot large object that you seem to have control over. It's entirely possible there is some "Evil Deceiver" that is tricking you into thinking it's all real!
I wonder why no one has thought of this before? Someone should write a treatise or something.
2000 miles across. 8 of the largest drives are located in Minnesota for offsite backup of the most important stuff. I work in Virginia.
In other news, 50% of all Slashdot readers are "above average" nitpickers. :)
(I meant arithmetic mean)
Heh, I think in that case, the private firm will be doing more damage to itself!
Or, the other case I see as probably being common, they are adding another new Linux server, but keeping their legacy Windows or Novell system around because it works for them, and just running both of them, waiting until they have time to finish the migration and totally ditch the old server.
That is the current situation at my workplace (Novell), and I know we aren't alone. These phased migrations are hard to measure statistically, since there isn't a trackable event (like a purchase) when a company decides to finally ditch the old system.
"For those that have recently purchased new Linux servers, 31 percent were adding capacity, 31 percent were replacing Windows systems, 24 percent were replacing Unix and 14 percent were replacing other operating systems."
Bad at math? It's OK, I heard on CNN that 50% of all people are below average in math.
In any case:
From those numbers, 69% of new installations were migrations, the rest were "adding capacity".
31% of total new installations were replacing legacy Windows systems.
31%/69% is about 45%. 45% of all migrations to Linux were from legacy Windows systems.
Does anyone know of a product like this?
Yes, but you won't like the price generally, for the real plug and play units.
ACNC
Makes external boxes that you hook to a normal SCSI controller card, comes in the form of boxes that can take ATA drives, or SCSI drives. Also makes fibre channel boxes of a similar nature. Appears as one large SCSI drive to the host OS, compatible with basically any OS.
HardData
Same deal as ACNC basically, but is more of a VAR of AXUS products of this type. Penguin on homepage a plus.
Promise
Low end crappy standalone ATA-to-SCSI boxes, similar to the above ones, also makes very crappy contoller cards, only useful if you are using them for software RAID, don't use thier hardware RAID for anything. Promise cards are also picky about the BIOS on the motherboard they are installed on. Their standalone box prices are way overpriced for what you get. Their controller cards are cheap, but acceptable, for software RAID. More than one 6 channel controller per computer is not supported, more than three two channel cards is not supported. Linux kernel module is mature though.
3ware
3ware makes hardware ATA RAID controllers that are very fast, and relatively expensive. "Unlimited" number of controllers per computer, I've ran up to four 8-port cards in a single computer. Cabling is a mess when you get a lot of drives in a single system, if you need that many, seriously consider one of the above standalone boxes. Linux kernel module is open source and vendor maintained. Management software for Linux is free but closed source.
-----------------
Of course, Linux has software RAID built right in, and with a distro like Red Hat, you can set up software RAID when you install the OS in a simple GUI. You can use whatever disks you have installed, but for good results, you need to only have one disk per ATA channel, be it on your motherboard or a Promise card.
We currently have 56 Maxtor drives in various RAIDs where I work. Some of them are coming up on a year old, with no failures yet (100, 120, 160GB). Of course, the fault tolerance that is built in to our systems can withstand quite a few failures before we would be in any trouble (up to half), and even then there are tapes behind most of it.
Our G4 Macs that have IBM drives though, those are dropping left and right, and most of those are less than a year old.
You're telling me... I tried to order a ATA-SCSI box full of (14) 160GB drives, and the assholes priced it so the drives were about $500 a piece, compared to the price of the box empty. He basically told me "Good Luck finding them on your own".
I guess we will just get the empty box and scrounge around until we can find 160s. I bet SCSI is hurting, considering how Maxtor can't keep the ATAs on the shelf they are selling so fast. With the ATA-SCSI switched bus RAID boxes being so cheap, so much faster than a shared bus SCSI arrangement, and just as reliable, it's no wonder they are selling out.
For the record, since the original poster seems confused (parent of parent),
Maxtor currently sells:
100, 120, 160GB 5400 RPM drives
WD is *supposed* to be selling:
180, 200GB 7200 RPM drives, but I havn't seen any for sale yet.
Maxtor 160GB street retail price around $240, WD 200GB MSRP at $400
The Maxtor 160s have been out for over 6 months now, WD just released their high capacity last month.
If intellectual property laws are repealed, then Linux as we know it is doomed.
Who said anything about repealing anything completely?
And Linux's only advantages-- better stability and security--are only as safe as its code base.
If IP laws were repealed, GNU/Linux could freely take whatever technology that was formerly covered by patent or copyright, the advantages of open source would be even more apparent, since companies that rely on IP only for income (otherwise known as leeches, or Rambus), will go out of business, leaving only the people who are really producing non-IP value to make the money.
With LEO sats, you could get that down to something more reasonable. My current satellite service gets minimum 600ms latency real world, in geostationary orbit, 35,000km or so up. LEO sats can be much much lower, 300-800km. At LEO, the latency is no longer an issue, in fact, you might get better latency than land lines under some circumstances!
Ebay offers escrow also, which would infringe on that most likely. Doesn't necessarily mean it will stand up in the end. Escrow isn't exactly a non-obvious innovation without prior art.