Japan has managed to create the first Petaflop supercomputer, called MDGrape-3, with just 4808 chips...
FLOP = floating operation [per second].
PETA = 10 ^ 15, or "a quadrillion".
(10 ^ 15) / 4808 = about 207,986,688,852, which would indicate that each chip is running at several hundred TERA-hertz [and, even then, the machine would have to possess an operating system so efficient that it could consistently perform one floating point operation per clock increment, which seems extraordinarily unlikely].
Or is this an "analog" computer and are these "analog" FLOPS?
Anonymous Coward:They want to see fiascos like the Xbox project and all the other marketplace failures terminated.
Anonymous Coward:The Xbox is almost a certainty to get the axe
The X-Box is their most innovative "rip-off" since at least Active Directory. They've sold 5 million units, and they think that, with Sony's delay in transitioning to the Cell processor, they can sell 15 million more in the next year. The gross revenues for the XBox alone are pretty much equal to [or even exceed] the gross revenues of Novell and Red Hat COMBINED.
Have you ever watched a boy [or even grown men, with professional degrees, like MDs, JDs, PhDs, etc] playing a video game? The stuff is utterly addictive. M$FT giving up the video game market would be like RJ Reynolds or Philip Morris giving up the cigarette market - it's a guaranteed revenue stream from now until the end of time.
In fact, in twenty or thirty years, I can see the various state Attorneys General suing M$FT, just like they sued the tobacco companies, in a massive class action law suit, on behalf of all those l00s3rs who wasted their lives away on that nonsense.
The XBox division might be bleeding red ink right now, but good grief, long term it has the potential to be phenomenally lucrative.
Actually, it's worse than slave labor. A slave master has to feed and clothe his slaves, and keep a roof over their heads.
In the real world, an enterprise-worthy CMS might cost easily several million in upfront R&D [anywhere from 10 to 50 man-years worth of labor, just to get to the "alpha" stage, with a total compensation package of easily $200,000 per man per year], and that's before you start regression testing and then moving to something you might call a "beta" version.
$5,000 is roughly what one contract programmer would cost for a WEEK. A week isn't even enough time to start making your stupid little "object note cards" [yeah, I'm dating myself]. Hell, a week isn't even enough time to start drawing stupid pictures on the whiteboard with the dry-erase markers.
Wake up people - you're being had. These corporations want FOSS out of you SO THAT THEY WON'T HAVE TO PAY YOU WHAT YOU DESERVE!!!
But seriously - anyone who falls for this con job deserves to be had.
Sounds like what The Hurd was supposed to be...
on
Driving Plan 9
·
· Score: 1
Exporting a name space is part of the deal, this presents many gifts that were not deliberately shoe-horned in such as remote step debugging across architectures, sending sound to a remote soundcard, importing a remote machine's network stack instead of using a gateway (including non-plan9 machines via ssh), importing remote filesystems (including non plan 9 machines). All this is facilitated by the 9p protocol [bell-labs.com].
This sounds remarkably similar to what Richard Stallman's The Hurd was supposed to be. [Speaking of which - Duke Nuke'em Forever ain't got nothing on The Hurd.]
In terms of languages that understand transparent network distribution - has anyone ported Erlang [or something similarly modern] to this platform?
Also, are there stripped-down versions of Plan 9 that can function in the RTOS space?
This sort of thing [highly parallelizable, highly fault-tolerant computing] was done more than a decade ago, at Hewlett-Packard, in the old Teramac group.
It sounds like Corel Draw can export to either AI or EPS, so I'd say that you would want to call up the people who burned the original CDR file, and ask them, "Could you please open the file, and choose either 'File | Save As... | Adobe Illustrator (AI)' or 'File | Save As... | Encapsulated PostScript (EPS)'? And then either mail me the file on a CD-ROM, or upload the file to a site like RapidShare, and then email me a link to the uploaded file so that I can download it? Thanks!"
If they have broadband, and if they upload to RapidShare, then the whole experience shouldn't take much more than about ten minutes. If they burn the file to a CD-ROM and snail-mail it to you, then it could take the better part of a week [or more].
Novell, with somewhere between 100 million & 200 million installed seats of Novell Directory Services ["eDirectory"], has got to be the world's leading vendor of RSA-based identification, authorization, and encryption products.
Personally, I think it would really r0x0r if the new OS shipped with an object-relational file system that had metadata, and a SQL-esque query syntax, and automated fall-over network distribution and...
It looks like [Messman's] been blamed for Novell's poor performance in the Linux space versus Red Hat. But can Linux ever be a real cash cow?
Wonder if Hovsepian will be on the phone with Hubert Mantel?
Or is that whole KDE/Gnome thang just a bridge too far?
A little off-topic, but here are a few ideas I've always had for Novel:
1) Write [or purchase] an in-house COMPILER! You've been releasing operating systems for more than twenty years now, yet you've never released your own compiler?!? Steve Ballmer may be a jackass, but he's absolutely correct when he says that it's all about "Developers, developers, developers!" If you can't write your own, then purchase Codewarrior from Motorola/Freescale. [By the way, my advice would be the same for AMD: If you're really serious about confronting Intel, then you've gotta produce something that can challenge the Intel C/C++ compiler.]
2) I don't want to say "abandon the desktop", because you've got a history of interest & subsequent disinterest in the desktop going back more than a decade, to WordPerfect and Corel, but here's an idea that involves expansion beyond the desktop: Take a long, hard look at the embedded market. NDS for QNX, NDS for VxWorks, NDS for Sybian, NDS for iTunes, etc, etc, etc. There are gonna be literally BILLIONS of those little devices sold in the next few years, and somebody's gonna hafta provide enterprise-level/mission-critical security for all of 'em. Might as well be the best enterprise-level/mission-critical outfit in the world.
3) Oh, and while you're at it: PLEASE port NDS to FreeBSD/OpenBSD.
4) Get the hell outta Beantown and return to your Utah roots. Blue States are where companies go to die. Red States are where they go to thrive.
You're wrong. You can cypher it with the public key and it can't be recovered without the private key, which is safe at his computer.
The system you've described is not [internally] closed - it requires interaction with an external agent [which, in theory, at least, means that the FBI/NSA can track the transaction and find the bad guy].
Ideally, though, you would want this sort of thing to be closed [with the encryption & decryption entirely internal to the system itself].
Now I'm not an expert in encryption [really hardly even a novice], but it's always seemed to me that there must be some meta-theorem in encryption theory which states that this is impossible - that an encrypted system which is capable of de-encrypting itself is necessarily capable of being de-encrypted by a third party [with sufficient time and money on its hands].
But is that true? Has such a meta-theorem been stated and proved?
Copyright law was a deal between the public and content producers that gave content producers the right to limit distribution for a limited time, in exchange for the requirement that their works fall into the public domain (i.e., can be copied against their wishes) after that period.
Fair enough: Classically, intellectual property was protected for about 17 years.
# 1 Another Day In Paradise - Phil Collins
# 2 Miss You Much - Janet Jackson
# 3 Straight Up - Paula Abdul
# 4 Right Here Waiting - Richard Marx
# 5 Lost In Your Eyes - Debbie Gibson
# 6 Like A Prayer - Madonna
# 7 We Didn't Start The Fire - Billy Joel
# 8 Two Hearts - Phil Collins
# 9 When I see You Smile - Bad Enough
# 10 Blame It On The Rain - Milli Vanilli
# 11 Forever Your Girl - Paula Abdul
# 12 Girl I'm Gonna Miss You - Milli Vanilli
# 13 Toy Soldiers - Martika
# 14 Cold Hearted - Paula Abdul
# 15 Don't Wanna Lose You - Gloria Estefan
# 16 Wind Beneath My Wings - Bette Midler
# 17 My Prerogative - Bobby Brown
# 18 She Drives Me Crazy - Fine Young Cannibals
# 19 The Look - Roxette
# 20 If You Don't Know Me By Now - Simply Red
How many of those "files" are being actively "shared" 17 years later?
For that matter, how many of those "files" are still being listened to 17 years later?
My question is; where they contributing anything new to the maket recently
It may not be all that "recent", but if you're a C++ programmer, you might want to download a copy of this documentation before the bankruptcy trustees pull the plug on the server:
You start and end your comment by lambasting Java:
Now, before I take a moment to rag on your ridiculous RAM comment, let me assure you that I hate Java from that ground up. I find it to be little more than a virus...
P.S. - Just to remind you, I hate Java. I would never use it again if I can avoid it and my goal in life to is make other people stop using it too...
Yet the body of your comment contains nothing but the highest praise for Java:
So here's where Java shines, because of the garbage collection system and because of the relocatable memory architecture, memory is managed in such a way which decreases the cycles spent in allocation and deallocation of buffers...
This is a Java shining point. Java includes so many libraries, that nearly every application can be written very quickly and with little effort...
Objects don't have to be read from disc more often than necessary in Java, so in theory a decent JVM can in fact use a precache (such as Insignia's AOT compiled versions) and load objects very quickly for exectution...
So why do you hate Java?
Are you a Lisp/Smalltalk/Haskell/Erlang guy?
Or does it have something to do with your background in embedded devices?
I have formed my own personal postulate/theory/law... and it's corollary:
It is impossible to completely test a sufficiently complex system in every possible way to be certain that it's bug-free.
Along those lines - many years ago, Professor Turing set out to find a test for [among other things] the possible presence of an infinite loop within a computer program.
It would be hard enough to do that without having to break in a slew of new characters: Lt. Col. Cameron Mitchell, General Hank Landry, and Valla.
Don't forget Dr. Carolyn Lam, fresh from the crew of Andromeda [well, I guess she WAS Andromeda itself].
By the way, this past Monday night, they replayed Heroes
Part I and Part I; boy, after a couple of years, you kinda forget what a cutie pie Dr. Janet Fraiser was.
nVidia & IBM/Sony/Cell/Playstation can perform only 32-bit single-precision floating point calculations in hardware. [IBM/Sony can, at least in theory, perform 64-bit double-precision floating point calculations, but the implementation involves some weird software emulation thingamabob which invokes a massive performance penalty.]
ATi is even worse - last I checked, they could perform only 24-bit "three-quarters"-precision floating point calculations in hardware.
And just in case you aren't aware, 32-bit single-precision floats are essentially worthless for anyone doing even the simplest mathematical calculations; for instance, with 32-bit single-precision floats, integer granularity is lost at 2 ^ 24 = 16M, i.e.
Now while 64-bit double-precision floats [or "doubles"] are probably accurate enough for most financial calculations, where, generally speaking, accuracy is only needed to the nearest 1/100th [i.e. to the nearest cent], 64-bit doubles are still more or less worthless to the mathematician, physicist, and engineer.
For instance, consider the work of Professor Kahan at UC-Berkeley:
At the time, Kahan was arguing in favor of using the full power of the Intel/AMD 80-bit extended precision doubles [i.e. embedding 64-bit doubles in an 80-bit space, performing calculations with the greater accuracy afforded therein, and then rounding the result back down to 64-bits and returning that as your answer], but, truth be told, the Sine Qua Non of hardware-based calculations is true 128-bit "quad-precision" floating point calculations as performed in hardware.
Sun has a "quad-precision" floating point number for Solaris/SPARC, but, sadly, it's a software hack, and, like IBM/Sony/Cell/Playstation, far too slow to be used in practice.
I believe that IBM makes a chip for the Z-Series mainframe, which can perform 128-bits in hardware, but I imagine that it's prohibitively expensive [if you could even convince IBM to sell it to you in the first place].
The best configuration here would probably look like a fancy-schmantzy Digitial Signal Processor [DSP] chipset, from someone like Texas Instruments, capable of 128-bit hardware calculations, mounted onto a card that would plug into something very fast, like a 16x PCIe bus, which in turn would be connected to a HyperTransport bus [but boy, wouldn't it be really cool if the DSP lay directly on the HyperTransport bus itself?].
By the way, if anyone knows of a company that's making such a card, with stable drivers [or, God forbid, a motherboard with a socket for a 128-bit DSP on the HyperTransport bus], then please tell me about it, 'cause I'd be very interested in purchasing such a thing.
We used iFolder 2 in a Novell cluster and it was a really nice product. Our laptop users loved it. None of the hassles of Windows offline folders (which seems to try to use a sledgehammer to nail in a tack) iFolder simply sits in the background and watches your iFolder for changes/new files and seamlessly syncs them with the iFolder on the server.
If a file becomes dirty [i.e. if it's altered], then, traditionally, Windows copied THE ENTIRE FILE to the server.
So if you were goofing around with a 1MB PDF, say, and you saved some changes, then Windows would push the entire 1MB across the wire.
iFolder, on the other hand, pushed ONLY THE DIRTY SECTORS - if, for instance, your alterations were confined to a single 512 byte sector, then only that sector was pushed across the wire.
So, in effect, Windows was like encrypted FTP, whereas iFolder was more like an encrypted file system over TCP.
Traditionally the great drawback to iFolder was that its passwords and passkeys were not stored in NDS, so you had to remember a set of passwords and passkeys that were distinct from your NDS password, which put a real damper on things. Don't know whether they've cleaned that up in v3, but having to memorize a second set of credentials kinda obviated the whole purpose of turning to Novell in the first place.
I spent a month looking to replace an employee, and got a whopping 2 resume's from a month of online ads and paper ads. And this was for an entry level position. All I needed was somebody fresh out that knew the software I use (the most popular in my industry, by the way) - I would train them in the in-house particulars. I got two resumes - one from a guy with no experience in the software at all, and one from a guy wanting $50k. For a $20-$25k position.
You've just proved, irrefutably, that it's not a $20-$25k position.
Look, it's not that hard to earn $40K hanging sheetrock
50wks/yr X 40hrs/wk X $20/hr = $40,000/yr
Entry-level janitors earn $25K/yr. To ask someone to work on crucial business logic for a salary like that is, quite frankly, to humiliate them.
The US intelligence agencies have had for years the capability to analyze some of the Internet traffic going through the US.
From everything I've heard, even if you look at nothing more than photographs from classical spy satellites, the CIA/NSA is just swamped with mountains upon mountains of data, and can't make heads or tails of what they've got.
To analyze "all" of the traffic on the internet, and make sense of any tiny fraction of it, would require a machine that would need to be designed and built be hyper-intelligent pan-dimensional mice.
Besides: Who watches the watchers? Who watches the watcher watchers? Who watches the watcher watcher watchers?
At some point, the thing starts to look like Bertrand Russell's Set of All Sets...
Slashdot Mirror
Japan has managed to create the first Petaflop supercomputer, called MDGrape-3, with just 4808 chips...
FLOP = floating operation [per second].
PETA = 10 ^ 15, or "a quadrillion".
(10 ^ 15) / 4808 = about 207,986,688,852, which would indicate that each chip is running at several hundred TERA-hertz [and, even then, the machine would have to possess an operating system so efficient that it could consistently perform one floating point operation per clock increment, which seems extraordinarily unlikely].
Or is this an "analog" computer and are these "analog" FLOPS?
And no, I did not RTFA.
Anonymous Coward: They want to see fiascos like the Xbox project and all the other marketplace failures terminated.
Anonymous Coward: The Xbox is almost a certainty to get the axe
The X-Box is their most innovative "rip-off" since at least Active Directory. They've sold 5 million units, and they think that, with Sony's delay in transitioning to the Cell processor, they can sell 15 million more in the next year. The gross revenues for the XBox alone are pretty much equal to [or even exceed] the gross revenues of Novell and Red Hat COMBINED.
Have you ever watched a boy [or even grown men, with professional degrees, like MDs, JDs, PhDs, etc] playing a video game? The stuff is utterly addictive. M$FT giving up the video game market would be like RJ Reynolds or Philip Morris giving up the cigarette market - it's a guaranteed revenue stream from now until the end of time.
In fact, in twenty or thirty years, I can see the various state Attorneys General suing M$FT, just like they sued the tobacco companies, in a massive class action law suit, on behalf of all those l00s3rs who wasted their lives away on that nonsense.
The XBox division might be bleeding red ink right now, but good grief, long term it has the potential to be phenomenally lucrative.
Actually, it's worse than slave labor. A slave master has to feed and clothe his slaves, and keep a roof over their heads.
In the real world, an enterprise-worthy CMS might cost easily several million in upfront R&D [anywhere from 10 to 50 man-years worth of labor, just to get to the "alpha" stage, with a total compensation package of easily $200,000 per man per year], and that's before you start regression testing and then moving to something you might call a "beta" version.
$5,000 is roughly what one contract programmer would cost for a WEEK. A week isn't even enough time to start making your stupid little "object note cards" [yeah, I'm dating myself]. Hell, a week isn't even enough time to start drawing stupid pictures on the whiteboard with the dry-erase markers.
Wake up people - you're being had. These corporations want FOSS out of you SO THAT THEY WON'T HAVE TO PAY YOU WHAT YOU DESERVE!!!
But seriously - anyone who falls for this con job deserves to be had.
Exporting a name space is part of the deal, this presents many gifts that were not deliberately shoe-horned in such as remote step debugging across architectures, sending sound to a remote soundcard, importing a remote machine's network stack instead of using a gateway (including non-plan9 machines via ssh), importing remote filesystems (including non plan 9 machines). All this is facilitated by the 9p protocol [bell-labs.com].
This sounds remarkably similar to what Richard Stallman's The Hurd was supposed to be. [Speaking of which - Duke Nuke'em Forever ain't got nothing on The Hurd.]
In terms of languages that understand transparent network distribution - has anyone ported Erlang [or something similarly modern] to this platform?
Also, are there stripped-down versions of Plan 9 that can function in the RTOS space?
Thanks!
This sort of thing [highly parallelizable, highly fault-tolerant computing] was done more than a decade ago, at Hewlett-Packard, in the old Teramac group.
Background here, here, here, etc.
Sorry about that; apparently I hit "Submit" instead of "Preview".
racism isn't just about believing one race to be superior, it's also about stereotyping; i.e. all arabs are terrorists
Like the submitter & the editor of this article, you are using the word "Arab" incorrectly.
There are Arab Jews, Arab Christians, Arab Zorastrians, Arab Animists, Arab Agnostics, Arab Atheists, etc, etc, etc.
The ones we are worried about, however, are the Arab MUSLIMS.
I've been given a Corel Draw Document (.cdr) as cover art for (charity) music CD... They accept... Illustrator...
I don't know the first thing about any of this, but I do know that Illustrator files end in ".ai", so I googled thusly:
Converting
The first hit I got was to this discussion from about three months ago:
Convert CDR to AI or EPS
It sounds like Corel Draw can export to either AI or EPS, so I'd say that you would want to call up the people who burned the original CDR file, and ask them, "Could you please open the file, and choose either 'File | Save As... | Adobe Illustrator (AI)' or 'File | Save As... | Encapsulated PostScript (EPS)'? And then either mail me the file on a CD-ROM, or upload the file to a site like RapidShare, and then email me a link to the uploaded file so that I can download it? Thanks!"
If they have broadband, and if they upload to RapidShare, then the whole experience shouldn't take much more than about ten minutes. If they burn the file to a CD-ROM and snail-mail it to you, then it could take the better part of a week [or more].
Same guys that bought VMWARE
Novell, with somewhere between 100 million & 200 million installed seats of Novell Directory Services ["eDirectory"], has got to be the world's leading vendor of RSA-based identification, authorization, and encryption products.
So if Novell ever goes belly-up, then it seems like EMC might be very interested in the bankruptcy sale.
Oh, and did I mention that Novell has this software product called "Novell Storage Services"?
Personally, I think it would really r0x0r if the new OS shipped with an object-relational file system that had metadata, and a SQL-esque query syntax, and automated fall-over network distribution and...
Toshiba is losing $200 per unit, of its new HD DVD player, in order to gain some marketshare.
I thought it was supposed to be "illegal" to sell a product below cost.
Or maybe that "law" only applies to Microsoft.
It looks like [Messman's] been blamed for Novell's poor performance in the Linux space versus Red Hat. But can Linux ever be a real cash cow?
Wonder if Hovsepian will be on the phone with Hubert Mantel?
Or is that whole KDE/Gnome thang just a bridge too far?
A little off-topic, but here are a few ideas I've always had for Novel:
I submitted an entire thread about the abomniation that is the Slashdot makeover, but of course it was rejected.
For the record, I predict a massive loss in readership [hence advertising revenue].
You're wrong. You can cypher it with the public key and it can't be recovered without the private key, which is safe at his computer.
The system you've described is not [internally] closed - it requires interaction with an external agent [which, in theory, at least, means that the FBI/NSA can track the transaction and find the bad guy].
Ideally, though, you would want this sort of thing to be closed [with the encryption & decryption entirely internal to the system itself].
Now I'm not an expert in encryption [really hardly even a novice], but it's always seemed to me that there must be some meta-theorem in encryption theory which states that this is impossible - that an encrypted system which is capable of de-encrypting itself is necessarily capable of being de-encrypted by a third party [with sufficient time and money on its hands].
But is that true? Has such a meta-theorem been stated and proved?
Copyright law was a deal between the public and content producers that gave content producers the right to limit distribution for a limited time, in exchange for the requirement that their works fall into the public domain (i.e., can be copied against their wishes) after that period.
Fair enough: Classically, intellectual property was protected for about 17 years.
Now here are twenty songs from the Billboard Top 100 for 1989:
How many of those "files" are being actively "shared" 17 years later?For that matter, how many of those "files" are still being listened to 17 years later?
Well, other than in elevators, maybe...
x-guru: In an interview this morning, Daffy Duck agreed with Jiaotong University. "Indeed, his behavior is despicable", said Mr. Duck.
Anonymous Coward: Sylvester's catch phrase was "sufferin' succotash!"
Or, as Mel Blanc would have put it:
any tools for downloading all of the STL site?
The main download page is here:
This is what I downloaded, plus the experimental C++ I/O Library:My question is; where they contributing anything new to the maket recently
It may not be all that "recent", but if you're a C++ programmer, you might want to download a copy of this documentation before the bankruptcy trustees pull the plug on the server:
I'm sacrificing Mod Points to ask this question.
You start and end your comment by lambasting Java:
Yet the body of your comment contains nothing but the highest praise for Java: So why do you hate Java?Are you a Lisp/Smalltalk/Haskell/Erlang guy?
Or does it have something to do with your background in embedded devices?
I have formed my own personal postulate/theory/law... and it's corollary: It is impossible to completely test a sufficiently complex system in every possible way to be certain that it's bug-free.
Along those lines - many years ago, Professor Turing set out to find a test for [among other things] the possible presence of an infinite loop within a computer program.
Sadly, though, he didn't get very far with that line of inquiry...
It would be hard enough to do that without having to break in a slew of new characters: Lt. Col. Cameron Mitchell, General Hank Landry, and Valla.
Don't forget Dr. Carolyn Lam, fresh from the crew of Andromeda [well, I guess she WAS Andromeda itself].
By the way, this past Monday night, they replayed Heroes Part I and Part I; boy, after a couple of years, you kinda forget what a cutie pie Dr. Janet Fraiser was.
nVidia & IBM/Sony/Cell/Playstation can perform only 32-bit single-precision floating point calculations in hardware. [IBM/Sony can, at least in theory, perform 64-bit double-precision floating point calculations, but the implementation involves some weird software emulation thingamabob which invokes a massive performance penalty.]
ATi is even worse - last I checked, they could perform only 24-bit "three-quarters"-precision floating point calculations in hardware.
And just in case you aren't aware, 32-bit single-precision floats are essentially worthless for anyone doing even the simplest mathematical calculations; for instance, with 32-bit single-precision floats, integer granularity is lost at 2 ^ 24 = 16M, i.e.
Now while 64-bit double-precision floats [or "doubles"] are probably accurate enough for most financial calculations, where, generally speaking, accuracy is only needed to the nearest 1/100th [i.e. to the nearest cent], 64-bit doubles are still more or less worthless to the mathematician, physicist, and engineer.For instance, consider the work of Professor Kahan at UC-Berkeley:
In particular, read a few of these papers from the late nineties: At the time, Kahan was arguing in favor of using the full power of the Intel/AMD 80-bit extended precision doubles [i.e. embedding 64-bit doubles in an 80-bit space, performing calculations with the greater accuracy afforded therein, and then rounding the result back down to 64-bits and returning that as your answer], but, truth be told, the Sine Qua Non of hardware-based calculations is true 128-bit "quad-precision" floating point calculations as performed in hardware.Sun has a "quad-precision" floating point number for Solaris/SPARC, but, sadly, it's a software hack, and, like IBM/Sony/Cell/Playstation, far too slow to be used in practice.
I believe that IBM makes a chip for the Z-Series mainframe, which can perform 128-bits in hardware, but I imagine that it's prohibitively expensive [if you could even convince IBM to sell it to you in the first place].
The best configuration here would probably look like a fancy-schmantzy Digitial Signal Processor [DSP] chipset, from someone like Texas Instruments, capable of 128-bit hardware calculations, mounted onto a card that would plug into something very fast, like a 16x PCIe bus, which in turn would be connected to a HyperTransport bus [but boy, wouldn't it be really cool if the DSP lay directly on the HyperTransport bus itself?].
By the way, if anyone knows of a company that's making such a card, with stable drivers [or, God forbid, a motherboard with a socket for a 128-bit DSP on the HyperTransport bus], then please tell me about it, 'cause I'd be very interested in purchasing such a thing.
We used iFolder 2 in a Novell cluster and it was a really nice product. Our laptop users loved it. None of the hassles of Windows offline folders (which seems to try to use a sledgehammer to nail in a tack) iFolder simply sits in the background and watches your iFolder for changes/new files and seamlessly syncs them with the iFolder on the server.
If a file becomes dirty [i.e. if it's altered], then, traditionally, Windows copied THE ENTIRE FILE to the server.
So if you were goofing around with a 1MB PDF, say, and you saved some changes, then Windows would push the entire 1MB across the wire.
iFolder, on the other hand, pushed ONLY THE DIRTY SECTORS - if, for instance, your alterations were confined to a single 512 byte sector, then only that sector was pushed across the wire.
So, in effect, Windows was like encrypted FTP, whereas iFolder was more like an encrypted file system over TCP.
Traditionally the great drawback to iFolder was that its passwords and passkeys were not stored in NDS, so you had to remember a set of passwords and passkeys that were distinct from your NDS password, which put a real damper on things. Don't know whether they've cleaned that up in v3, but having to memorize a second set of credentials kinda obviated the whole purpose of turning to Novell in the first place.
I spent a month looking to replace an employee, and got a whopping 2 resume's from a month of online ads and paper ads. And this was for an entry level position. All I needed was somebody fresh out that knew the software I use (the most popular in my industry, by the way) - I would train them in the in-house particulars. I got two resumes - one from a guy with no experience in the software at all, and one from a guy wanting $50k. For a $20-$25k position.
You've just proved, irrefutably, that it's not a $20-$25k position.
Look, it's not that hard to earn $40K hanging sheetrock
Entry-level janitors earn $25K/yr. To ask someone to work on crucial business logic for a salary like that is, quite frankly, to humiliate them.The US intelligence agencies have had for years the capability to analyze some of the Internet traffic going through the US.
From everything I've heard, even if you look at nothing more than photographs from classical spy satellites, the CIA/NSA is just swamped with mountains upon mountains of data, and can't make heads or tails of what they've got.
To analyze "all" of the traffic on the internet, and make sense of any tiny fraction of it, would require a machine that would need to be designed and built be hyper-intelligent pan-dimensional mice.
Besides: Who watches the watchers? Who watches the watcher watchers? Who watches the watcher watcher watchers?
At some point, the thing starts to look like Bertrand Russell's Set of All Sets...