From the article:
The RPOW server
is running on a high-security processor card, the IBM 4758 Secure
Cryptographic Coprocessor, validated to FIPS-140 level 4.... Please keep in mind that if there are problems I may need to reload the server code, which will invalidate any RPOW tokens which people have previously created.
So, in other words, it passes out little tokens that are worth something... but just until the server is taken out.
Ok, so its running FIPS-certified code on FIPS-certified hardware. Still, how sure can you be that it will keep running 24/365 for years on end? If that private key is needed for proof of authenticity, and that key never leaves the board, that makes it, among other things, one heckuva terrorist target.
"I watched the tapes of the Nuremburg experiments that showcased how people put in positions of authority could be ordered to torture and kill other people and that the majority of those tested in the study failed the "humanity" test."
What's the name of that law about when the argument gets to comparing the opponent with Nazi's?
Godwin's Law is a natural law of Usenet named after Mike Godwin
(godwin@eff.org) concerning Usenet "discussions". It reads, according to
the Jargon File:
As a Usenet discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.
I'm quite familier (sic) with title 17 of the us code, state and local laws might differ and there is a possability (sic) of a law I havn't (sic) seen yet. If I'm wrong (leagaly (sic) not moraly (sic)) then please cite the laws for me.
This kind of post really cracks me up. If I were able to get past the 20+ horrendous spelling and grammatical errors, I might believe this guy knows what he's talking about. I mean, he claims he's quite familier (sic) with the pertinent codes. However, when it looks like an 8-year-old wrote it, I'm hard pressed to believe there's any coherent thought behind it. The only things missing are the lined paper and a burnt-umber crayon.
Note to others:
A little skill in spelling will get you far in life. Would you hire someone who's resume looked like the parent post?
Note to prospective employers:
Make your candidates write something during their interview. This guy will undoubtedly have his resume professionally written, and you certainly wouldn't want him issuing your press releases!
Shockwave doesn't seem to have a problem getting the flash player installed on damn near every computer all on their own.
The only reason I have Flash installed on my computer is because the latest Firefox builds have a very handy Flash Click To View extension (Thanks Ted!) allowing me to choose when flash will run! Before that, if a web site had a flash intro page without a "Skip Intro" link, or a menu system that required Flash, my fingers just browsed on over to someone else's site.
"This is a very difficult issue for us," Marsh said. "It is a huge disappointment to us that we would be thrown in the same bucket, so to say, with the SCO Group. We didn't make any admission that their IP was used in Linux. Our public position is certainly not to support that, and our intent was simply to take us out of the loop, not to make us public enemy number one."
- - Robert Marsh, CEO, EV1Servers
Captain Cheese, I suspect that many people simply don't trust a CEO that would make such a foolish, but expensive, impulse decision. If you believe SCO has a snowball's chance in hell, then you buy "insurance." However, if you're in tune enough with the tech community, then you know better than to throw your money away.
And where does CEO Marsh's money come from? The people who host their servers with his company! Every dollar spent on SCO means a dollar that they cannot spend on servers, maintenance, preventative maintenance, etc.
I'm all for the public outcry and boycotts. The more noise us Linux supporters make, the less people will buy from SCO, the quicker they fold and this thing goes away.. It just takes some people a little while to learn that you're either with us, or your against us on this issue.
What if the power company clocks the datastream by using the 50 Hz (IIRC) that AC power cycles at?
As someone else mentioned, its 60Hz in the USA (including Cincinnati), not 50Hz.
Additionally, IF you were able to modulate the signals enough to overcome variations in power draw, without disturbing devices that expect a relatively steady voltage, you'd get a total theoretical bandwidth of (gasp!) 60Hz. Considering that modems surpassed that, oh, at their CONCEPTION, it hardly qualifies as broadband. I don't see people signing up for 50bps (or even 50,000bps) shared-pipe service in droves..
Oh, c'mon, guys. Can't you see that Mr. Aucsmith is just trolling the world?? Move along.. Nothing to see here. The best way to deal with trolls is to ignore thm. Responding only encourages their actions!
BTW, I have a slashcode improvement request: I'd like the ability to moderate front page articles as "-1 Troll"
This patent application is thorough, obviously written by someone who knows how to push patents through. The last three pages of the PDF have a listing of 25 claims that they say are the embodiment of the "invention." I'm not a patent examiner, and am not in the place to refute each of them. However, its clear that they aren't claiming a patent on basic pager functionality. As mentioned elsewhere, they give examples of prior art.
There do seem to be some improvements listed. Foremost appears to be the ability to view a scaled version of the desktops in full screen instead of just the little icons in the pager. For instance, with 4 virtual desktops, they describe a scaled view where each desktop is essentially 1/4 of the screen. If you have two browser windows open in two different desktops, such a view would enable you to visually determine which is which. I don't remember seeing such a feature in other VWMs. They also describe animating the transition between this view and the full desktops via shrinking/expanding the active desktop.
While this does seem to be an improvement over existing pagers, many will argue its triviality. I personally suspect it's still enough to get the patent issued and drag someone into court...
Intel has brought us a lot of real inventions. I'm thinking of HyperThreading, and even Itanium...
Not to nitpick, but Hyperthreading is not an Intel "innovation." It is actually based off of DEC Alpha technology that they bought when Compaq sold their CPU business off. If Alpha had survived another iteration, we'd have seen Symmetric Multi-Threading (SMT) labeled with much less marketing spin than we currently have.
Even IA-64, Intel's incarnation of Very Long Instruction Word (VLIW) technology, was not new when Intel decided to base Itanium on it.
As a side note, I always wished AMD had named the Opteron/Athlon64 series Atinum (as in pl-Atinum) or Alladium (as in p-Alladium). Either metal is more precious than t-Itanium, which is only considered a minor metal in the markets. Besides being a great play on words, they start with 'A', fitting in with the whole AuthenticAMD theme... although they probably would have gotten sued like Lindows was.. *sigh*
I'm still waiting to see AMD cross-license and make use of SMT technology in the Hammer line. *drool*
... how are we anywhere close to maxing out the 10^10 numbers available?
The US numbering plan format is often described as NPA-NXX-XXXX. NPA is the Numbering Plan Area (aka area code). The N in NXX is because the local exchange number never starts with a 0 or a 1; that first digit determines whether your placing a local vs. a long-distance or operator-assisted call. For that reason, each NPA really only has a theoretical maximum of 8 million numbers to allocate. The same restriction applies to area codes. That alone has reduced that 10 billion number down to around 6.4 billion (800 * 8 million).
There are other restrictions as well, such as the N11 numbers (411, 911, etc.) that will never be an NPA or an NXX. They're reserving some NPAs for additional toll-free codes beyond 800, 888, 877 and 866. The 555 NXX block is reserved for non-valid numbers. That's why they pop up all the time in movies. Also, consider that a small village with 3000 phone lines is still allocated an entire NXX exchange of around 10,000, and you can see where whole ranges of numbers are lost.
The first time they maxed out the US dialing plan (about a decade ago?), they eliminated the requirement for the middle digit in all NPAs to be 0 or 1. At that time, the theoretical maximum would have been close to 8 * 2 * 10 * 8 million, which is close to 1.25 billion That's still enough for around 4 numbers per person, based on your statistics. However, when you consider that an average single working person had a home phone, a cell phone or pager, and an office phone, that's not too preposterous.
In any flexible numbering plan, you can't just take x^y to come up with the maximum possible combinations. Take IANA, as you mentioned.. there are 32 bits, but there aren't 2^32 valid IPv4 addresses. There's multicast, reserved blocks, loopback adapter, netmasks, and so on. There will always be some ranges sacrificed for flexibility and features.
Yes, and that's most likely why current telecom providers want this to be considered a regulated service...
Interestingly, Jeff has several friends in the telecom and datacom industries, among them AT&T, Qwest, Worldcom, Global Crossing and Cisco. Outside of that, the venerable EFF is also in favor of his petition. (Donate now!)
I didn't include it in the story in order to avoid trampling his site too much, but he still has the original petition [PDF] available online.
Maintaining the metaphor, I suppose its more like FWD stepping in front of the gun to see if the FCC pulls the trigger, rather than the FCC swinging the gun so that it points at FWD.
The best defense is a good offense -- Vince Lombardi
I was recently the lucky recipient of a refund check resulting from a class action lawsuit against Citibank. I haven't cashed it yet.. but I may frame it as an example of how stupid some lawsuits can be...
It was for a whopping $0.06! Postage (even bulk rate) cost several times what I ended up with. The big winners were some shmucky law firm, and the USPS.
I can't find the link, but I remember hearing once that code optimized for the Pentium actually had a larger speed improvement on the Athlon than it did on the Pentium. I don't recall which models were under test or what their clock speeds were, but it was a few years ago now. It wasn't publicized much, of course, because it bloodied Intel's nose..
The point is that AMD CPUs perform optimized and unoptimized tasks more efficiently (wrt clock speed) than Intel CPUs. While Intel has focused on exceedingly high clockspeed (witness the 31 stage pipeline..) and compiler improvements (IA-64/VLIW anyone?), AMD has gone for parallelism and efficiency in utilization.
AMD pioneered vector math with 3DNow (Correct me if I'm mistaken, but I'm pretty sure 3DNow predates MMX...) but, instead of Intel using AMD's technology, they developed MMX in response. I believe code optimized for either extension saw about the same improvement. However, as market leader, Intel led the market where they wanted. And since more developers coded for MMX than 3DNow, AMD added MMX. In doing so, AMD's chips started benefitting just as much or more from Intel's extensions.
Historically, its been Intel introducing the extensions, and AMD quietly implementing them in response. The x86-64 extensions are shaping up to be the first example of an AMD-pioneered extension that Intel will implement!
As for Intel vs. GCC compilers, GCC is a very flexible cross compiler that targets many different architectures. Intel's compilers are very architecture specific (hint: they target only Intel's architectures). Apples and oranges. Like Java and the CLR; their design goals are worlds apart.
Stock Evaluation 101:
Blindly evaluating a business's statistics won't give you a complete picture. If you look at SCO's current business, they have no viable product. Their earnings are based entirely off of their much bally-hooed "IP portfolio." Since that portfolio is on such shaky ground, if their allegations prove false or their IP is unenforcable, we have every reason to believe the "company is going down the tubes." This community has a particularly insightful view of the strengths of their claims (as opposed to Joe CEO, or Jim Finance). So, Mr. AC is likely not talking out of his ass.. he's making an informed (although colorful) assessment of his belief of SCOs future earnings potential.
Re:The replacement is already here
on
United Linux Dead
·
· Score: 4, Interesting
Bruce, don't get me wrong, I like you and the work you do. I've got you on my friends list. I'll probably fiddle with UserLinux when it comes out. But this is close to inexcusible.
United Linux was, to my understanding, a corporate response to RHEL ES/AS, the "server" products. My understanding of UserLinux is that it is a grassroots response to RHEL WS, the "desktop" product. Of course, any Linux can be used to run server apps, but the point is UserLinux's target is the desktop. United Linux had plans of certifying the "big iron" apps like Oracle, SAP, etc., that large corporations feel they need support for. How long before I'll be able to get Oracle to support their latest datacenter DB product on UserLinux like I could right now on RHEL AS? I'm afraid it'll take more than a grassroots effort to compete with Redhat's server lineup...
The person who is the billing contact will be sued. No lawsuit has played out in court to know what the result of a lawsuit brought against a person who gave access to someone who shared copyrighted material would be.
I wouldn't be surprised to see search warrants make an appearance in these trials.. After all, its the courts that issue them, and that's where the warrants issued to the ISPs to discover ownership come from.
Furthermore, if the judge doesn't feel an IP address and a couple log files are proof enough, I'm sure the RIAA is going to try to get the local police to storm down the door and confiscate some hard drives. That would really rile up the conspiracy theorists!
In completely unrelated news, I hear that sales of wireless fileservers are up...
My apologies.. I didn't see the question. In fact, I still didn't see it in my cursory search.
If you don't mind rolling your own NAS, I recommend NetZerver. I've had one for a few year and like it. It does NFS, CIFS, Appleshare, Netware, HTTP, FTP, and SmartMirror for backup. Its available in IDE (6 channels) or SCSI (15 devs) and is the size of a long CDROM. You can dynamically grow and shrink your RAID5 group. The latest version even supports MS AD, for what that's worth. Count me as a happy customer..
Slashdot Mirror
From the article: ... Please keep in mind that if there are problems I may need to reload the server code, which will invalidate any RPOW tokens which people have previously created.
... but just until the server is taken out.
The RPOW server is running on a high-security processor card, the IBM 4758 Secure Cryptographic Coprocessor, validated to FIPS-140 level 4.
So, in other words, it passes out little tokens that are worth something
Ok, so its running FIPS-certified code on FIPS-certified hardware. Still, how sure can you be that it will keep running 24/365 for years on end? If that private key is needed for proof of authenticity, and that key never leaves the board, that makes it, among other things, one heckuva terrorist target.
- What's the name of that law about when the argument gets to comparing the opponent with Nazi's?
Godwin's Law1. What is Godwin's Law?
Godwin's Law is a natural law of Usenet named after Mike Godwin (godwin@eff.org) concerning Usenet "discussions". It reads, according to the Jargon File:
- As a Usenet discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.
LOL! Good catch.Does it offer FLAC support?
That's why they recommend pepperocini's for first-time inmates...
"It burns! It burns!"
Or.. for the cut-n-paste challenged:
http://www.redhat.com/about/presscenter/
I'm quite familier (sic) with title 17 of the us code, state and local laws might differ and there is a possability (sic) of a law I havn't (sic) seen yet. If I'm wrong (leagaly (sic) not moraly (sic)) then please cite the laws for me.
This kind of post really cracks me up. If I were able to get past the 20+ horrendous spelling and grammatical errors, I might believe this guy knows what he's talking about. I mean, he claims he's quite familier (sic) with the pertinent codes. However, when it looks like an 8-year-old wrote it, I'm hard pressed to believe there's any coherent thought behind it. The only things missing are the lined paper and a burnt-umber crayon.
Dear sumdumass,
Are you familiar with Merriam-Webster?
Regards,
Note to others:
A little skill in spelling will get you far in life. Would you hire someone who's resume looked like the parent post?
Note to prospective employers:
Make your candidates write something during their interview. This guy will undoubtedly have his resume professionally written, and you certainly wouldn't want him issuing your press releases!
</decaying-spelling-rant>
And don't forget...
Do not taunt Happy Fun Ball.
Shockwave doesn't seem to have a problem getting the flash player installed on damn near every computer all on their own.
The only reason I have Flash installed on my computer is because the latest Firefox builds have a very handy Flash Click To View extension (Thanks Ted!) allowing me to choose when flash will run! Before that, if a web site had a flash intro page without a "Skip Intro" link, or a menu system that required Flash, my fingers just browsed on over to someone else's site.
"This is a very difficult issue for us," Marsh said. "It is a huge disappointment to us that we would be thrown in the same bucket, so to say, with the SCO Group. We didn't make any admission that their IP was used in Linux. Our public position is certainly not to support that, and our intent was simply to take us out of the loop, not to make us public enemy number one."
- - Robert Marsh, CEO, EV1Servers
Captain Cheese, I suspect that many people simply don't trust a CEO that would make such a foolish, but expensive, impulse decision. If you believe SCO has a snowball's chance in hell, then you buy "insurance." However, if you're in tune enough with the tech community, then you know better than to throw your money away.
And where does CEO Marsh's money come from? The people who host their servers with his company! Every dollar spent on SCO means a dollar that they cannot spend on servers, maintenance, preventative maintenance, etc.
I'm all for the public outcry and boycotts. The more noise us Linux supporters make, the less people will buy from SCO, the quicker they fold and this thing goes away.. It just takes some people a little while to learn that you're either with us, or your against us on this issue.
What if the power company clocks the datastream by using the 50 Hz (IIRC) that AC power cycles at?
As someone else mentioned, its 60Hz in the USA (including Cincinnati), not 50Hz.
Additionally, IF you were able to modulate the signals enough to overcome variations in power draw, without disturbing devices that expect a relatively steady voltage, you'd get a total theoretical bandwidth of (gasp!) 60Hz. Considering that modems surpassed that, oh, at their CONCEPTION, it hardly qualifies as broadband. I don't see people signing up for 50bps (or even 50,000bps) shared-pipe service in droves..
it was said by the Microsoft Security Chief.
Oh, c'mon, guys. Can't you see that Mr. Aucsmith is just trolling the world?? Move along.. Nothing to see here. The best way to deal with trolls is to ignore thm. Responding only encourages their actions!
BTW, I have a slashcode improvement request: I'd like the ability to moderate front page articles as "-1 Troll"
This patent application is thorough, obviously written by someone who knows how to push patents through. The last three pages of the PDF have a listing of 25 claims that they say are the embodiment of the "invention." I'm not a patent examiner, and am not in the place to refute each of them. However, its clear that they aren't claiming a patent on basic pager functionality. As mentioned elsewhere, they give examples of prior art.
There do seem to be some improvements listed. Foremost appears to be the ability to view a scaled version of the desktops in full screen instead of just the little icons in the pager. For instance, with 4 virtual desktops, they describe a scaled view where each desktop is essentially 1/4 of the screen. If you have two browser windows open in two different desktops, such a view would enable you to visually determine which is which. I don't remember seeing such a feature in other VWMs. They also describe animating the transition between this view and the full desktops via shrinking/expanding the active desktop.
While this does seem to be an improvement over existing pagers, many will argue its triviality. I personally suspect it's still enough to get the patent issued and drag someone into court...
Intel has brought us a lot of real inventions. I'm thinking of HyperThreading, and even Itanium ...
Not to nitpick, but Hyperthreading is not an Intel "innovation." It is actually based off of DEC Alpha technology that they bought when Compaq sold their CPU business off. If Alpha had survived another iteration, we'd have seen Symmetric Multi-Threading (SMT) labeled with much less marketing spin than we currently have.
Even IA-64, Intel's incarnation of Very Long Instruction Word (VLIW) technology, was not new when Intel decided to base Itanium on it.
As a side note, I always wished AMD had named the Opteron/Athlon64 series Atinum (as in pl-Atinum) or Alladium (as in p-Alladium). Either metal is more precious than t-Itanium, which is only considered a minor metal in the markets. Besides being a great play on words, they start with 'A', fitting in with the whole AuthenticAMD theme... although they probably would have gotten sued like Lindows was.. *sigh*
I'm still waiting to see AMD cross-license and make use of SMT technology in the Hammer line. *drool*
... how are we anywhere close to maxing out the 10^10 numbers available?
The US numbering plan format is often described as NPA-NXX-XXXX. NPA is the Numbering Plan Area (aka area code). The N in NXX is because the local exchange number never starts with a 0 or a 1; that first digit determines whether your placing a local vs. a long-distance or operator-assisted call. For that reason, each NPA really only has a theoretical maximum of 8 million numbers to allocate. The same restriction applies to area codes. That alone has reduced that 10 billion number down to around 6.4 billion (800 * 8 million).
There are other restrictions as well, such as the N11 numbers (411, 911, etc.) that will never be an NPA or an NXX. They're reserving some NPAs for additional toll-free codes beyond 800, 888, 877 and 866. The 555 NXX block is reserved for non-valid numbers. That's why they pop up all the time in movies. Also, consider that a small village with 3000 phone lines is still allocated an entire NXX exchange of around 10,000, and you can see where whole ranges of numbers are lost.
The first time they maxed out the US dialing plan (about a decade ago?), they eliminated the requirement for the middle digit in all NPAs to be 0 or 1. At that time, the theoretical maximum would have been close to 8 * 2 * 10 * 8 million, which is close to 1.25 billion That's still enough for around 4 numbers per person, based on your statistics. However, when you consider that an average single working person had a home phone, a cell phone or pager, and an office phone, that's not too preposterous.
In any flexible numbering plan, you can't just take x^y to come up with the maximum possible combinations. Take IANA, as you mentioned.. there are 32 bits, but there aren't 2^32 valid IPv4 addresses. There's multicast, reserved blocks, loopback adapter, netmasks, and so on. There will always be some ranges sacrificed for flexibility and features.
Unfortunately, MS can delay this and consider it legitimate. They have a workaround available: IE6.
Catch up.. The $50M investment was a venture capital offering funded largely by RBC Centura. That's old news.
Yes, and that's most likely why current telecom providers want this to be considered a regulated service...
Interestingly, Jeff has several friends in the telecom and datacom industries, among them AT&T, Qwest, Worldcom, Global Crossing and Cisco. Outside of that, the venerable EFF is also in favor of his petition. (Donate now!)
I didn't include it in the story in order to avoid trampling his site too much, but he still has the original petition [PDF] available online.
I agree.. but it got your attention, didn't it?
Maintaining the metaphor, I suppose its more like FWD stepping in front of the gun to see if the FCC pulls the trigger, rather than the FCC swinging the gun so that it points at FWD.
The best defense is a good offense -- Vince Lombardi
I was recently the lucky recipient of a refund check resulting from a class action lawsuit against Citibank. I haven't cashed it yet.. but I may frame it as an example of how stupid some lawsuits can be...
It was for a whopping $0.06! Postage (even bulk rate) cost several times what I ended up with. The big winners were some shmucky law firm, and the USPS.
Ridiculous!
My bad.. I thought I remembered the first chips with 3DNow not including MMX. Thanks for doing the research for me.. 8-)
Where is the AMD's answer to Intel's compiler?
I can't find the link, but I remember hearing once that code optimized for the Pentium actually had a larger speed improvement on the Athlon than it did on the Pentium. I don't recall which models were under test or what their clock speeds were, but it was a few years ago now. It wasn't publicized much, of course, because it bloodied Intel's nose..
The point is that AMD CPUs perform optimized and unoptimized tasks more efficiently (wrt clock speed) than Intel CPUs. While Intel has focused on exceedingly high clockspeed (witness the 31 stage pipeline..) and compiler improvements (IA-64/VLIW anyone?), AMD has gone for parallelism and efficiency in utilization.
AMD pioneered vector math with 3DNow (Correct me if I'm mistaken, but I'm pretty sure 3DNow predates MMX...) but, instead of Intel using AMD's technology, they developed MMX in response. I believe code optimized for either extension saw about the same improvement. However, as market leader, Intel led the market where they wanted. And since more developers coded for MMX than 3DNow, AMD added MMX. In doing so, AMD's chips started benefitting just as much or more from Intel's extensions.
Historically, its been Intel introducing the extensions, and AMD quietly implementing them in response. The x86-64 extensions are shaping up to be the first example of an AMD-pioneered extension that Intel will implement!
As for Intel vs. GCC compilers, GCC is a very flexible cross compiler that targets many different architectures. Intel's compilers are very architecture specific (hint: they target only Intel's architectures). Apples and oranges. Like Java and the CLR; their design goals are worlds apart.
Please, by all means, invest in them then.
Stock Evaluation 101:
Blindly evaluating a business's statistics won't give you a complete picture. If you look at SCO's current business, they have no viable product. Their earnings are based entirely off of their much bally-hooed "IP portfolio." Since that portfolio is on such shaky ground, if their allegations prove false or their IP is unenforcable, we have every reason to believe the "company is going down the tubes." This community has a particularly insightful view of the strengths of their claims (as opposed to Joe CEO, or Jim Finance). So, Mr. AC is likely not talking out of his ass.. he's making an informed (although colorful) assessment of his belief of SCOs future earnings potential.
The UserLinux project is United Linux done right.
*cough* *c SHAMELESS PLUG!!! ough* *ahem* Err... excuse me.
Bruce, don't get me wrong, I like you and the work you do. I've got you on my friends list. I'll probably fiddle with UserLinux when it comes out. But this is close to inexcusible.
United Linux was, to my understanding, a corporate response to RHEL ES/AS, the "server" products. My understanding of UserLinux is that it is a grassroots response to RHEL WS, the "desktop" product. Of course, any Linux can be used to run server apps, but the point is UserLinux's target is the desktop. United Linux had plans of certifying the "big iron" apps like Oracle, SAP, etc., that large corporations feel they need support for. How long before I'll be able to get Oracle to support their latest datacenter DB product on UserLinux like I could right now on RHEL AS? I'm afraid it'll take more than a grassroots effort to compete with Redhat's server lineup...
The person who is the billing contact will be sued. No lawsuit has played out in court to know what the result of a lawsuit brought against a person who gave access to someone who shared copyrighted material would be.
I wouldn't be surprised to see search warrants make an appearance in these trials.. After all, its the courts that issue them, and that's where the warrants issued to the ISPs to discover ownership come from.
Furthermore, if the judge doesn't feel an IP address and a couple log files are proof enough, I'm sure the RIAA is going to try to get the local police to storm down the door and confiscate some hard drives. That would really rile up the conspiracy theorists!
In completely unrelated news, I hear that sales of wireless fileservers are up...
My apologies.. I didn't see the question. In fact, I still didn't see it in my cursory search.
If you don't mind rolling your own NAS, I recommend NetZerver. I've had one for a few year and like it. It does NFS, CIFS, Appleshare, Netware, HTTP, FTP, and SmartMirror for backup. Its available in IDE (6 channels) or SCSI (15 devs) and is the size of a long CDROM. You can dynamically grow and shrink your RAID5 group. The latest version even supports MS AD, for what that's worth. Count me as a happy customer..