it's not just L glass (which would go without saying - you don't get a 1-class body for sub-par glass), you would need a L prime, L zooms might not be corrected enough for the sensor resolution.
You do realize that 800 lines/mm means 10/8=1.25 microns, right? so not quite submicron. Size of the silver cristals is not all. You could put nanometer-sized sensors and achievable resolution would be quite poor - yeah, you count photons, but if there aren't enough photons to activate every sensor then your extra resolution is useless.
Also, 800 lines/mm are the (slow) laboratory 'max. resolution' conditions. You'll not get that using a 35mm camera, as the rest of the optical system will bring your max. resolution lower. Notably, the lens' resolving power becomes important sooner.
So the correct comparison is "effective resolution" in use, not max. achievable one under very special conditions.
Yeah, but you'd better use a damn good prime lens if you want to scan the film at 4kDPI. Otherwise it's a waste - you're limited by the lens, not by the storing medium.
Sun is not supposed to be datacenter only. Workstations can be used in scientific environments where for instance image data aquisition is necessary. Or using any custom-made sensor that would require a kernel driver to interface with the workstation. Hard to have that on Solaris, a lot easier on Linux.
If you read the reply you'll see the argument that Linux suns on some hardware that does not provide diagnostics/recovery capabilities. When it's possible, Linux tries to implement it (that also depends on the spec availability in a GPL-compatible form). It's probably true that Sun does a better job at it, as they control what hardware goes into their boxes and thus all the details about error recovery information available. Also, consumer PCs don't really need a high-availability server's degree of soft error recovery.
Yes, it's not a lot of the argument, but still an important part.
Tiff is HUGE (at least in uncompressed form, which seems to be what cameras use these days). On the other hand (quote from here)
DNG is based on the TIFF-EP format with extra metadata and scope for lossless JPEG compression, Adobe said. The format supports both mosaiced (CFA) and demosaiced interpolation.
Doh! Did you read the article? It's about the EC suggesting ISO. EC is not a business. It means governments might be interested in this - and gov. departments don't have the same integration requirements as businesses (also, they don't have the same funny DRM ideas). It might move some of them off MS Office, and that would be quite an achievement.
Also, if official documents are required to be submitted in OO.o format and MS Office does not support that, don't you think businesses will have to look for a supporting application? From TFA, EC asked for interoperability filters, which would exactly offering a conversion tool for businesses if MS gives this format a pass.
So yeah, I don't think this concerns businesses a lot - only a little at this stage. But it can have a snowball effect if it really catches the right start.
You seem to be making the common mistake here. It's not about Sun/OpenOffice per se, but about their document format. Which means, if it gets standardized upon by EU, then someone (presumably MS) will come up with a MS Office plugin to allow it to work with it - after all, how hard is it to make an XSLT sheet set?
The tool will be whatever integrates best with the organisation's workflow, but the format that tool will be using is the matter here.
Your logic also has a fatal flaw - it's not GPL software that's being sold. It's the "package".
If a particular distro gives you the choice of buying a box and downloading the isos, what is it you're buying? not the software, as that is also available for free. THE CONVENIENCE. You get the whole pack directly, plus manuals, plus some extra perks if that's the case (better access, commercial extensions, etc.). Also, you're buying some (limited) support. NOT THE GPL SOFTWARE.
Of course, if they don't provide the soft for free, then you're technically buying it (the binaries), but it's again a matter of convenience - getting a precompiled package instead of going LFS-like. There's still a way around this, though. They can provide sources , say a.src.rpm package that you can download, rebuild and install. That would be for free, thus not commercial. Or you can get the source directly from the author.
Think of it this way - if you're buying a book that is also available in the public domain, what is it that you're buying?
So I guess the point is, GPL software itself should not be regulated by this as commercial software is. If the author gives it as GPL then it's free. If the packager charges money for the binary, then the packager has to comply with commercial regulations.
I know you were just rethorical about it, but RTFA and at least you might be able to sound interesting.
Here's a bunch of clues, anyway:
No sane Dept. of Defense would rely on foreign software - so the contractors are French.
Mandrake is not the sole contractor.
The contract is for 3 years and aims at a CC-EAL5 certification; not exactly a typical Mandrake setup.
Linux and Security can mix - although not everyone uses that mix. Witness SELinux and it's offsprings.
Anyway, if they manage a EAL5 certification for this, they'll be able to laugh their asses out at Defense Dept.s that use (or even more, mandate the use) of Windows on their computers. Seeing that w2k only made it (dubiously) to EAL4.
EAL4 Methodically Designed, Tested and Reviewed. Analysis is supported by the low-level design of the modules of the TOE, and a subset of the implementation. Testing is supported by an independent search for obvious vulnerabilities. Development controls are supported by a life-cycle model, identification of tools, and automated configuration management.
EAL5 Semiformally Designed and Tested. Analysis includes all of the implementation. Assurance is supplemented by a formal model and a semiformal presentation of the functional specification and high level design, and a semiformal demonstration of correspondence. The search for vulnerabilities must ensure relative resistance to penetration attack. Covert channel analysis and modular design are also required.
The thing EMT64 does not have, for instance, is hardware IOMMU support. That kind of sucks, as it means DMA I/O has to go in the lower 4G of memory[*] - remember the low-mem buffers for legacy (16-bit) PCI devices? well, Intel just came up with an equivalent trick for processors.
[*] to quote Redhat:
Intel® EM64T does not support an IOMMU in hardware while AMD64 processors do. This means that physical addresses above 4GB (32 bits) cannot reliably be the source or destination of DMA operations. Therefore, the Red Hat Enterprise Linux 3 Update 2 kernel "bounces" all DMA operations to or from physical addresses above 4GB to buffers that the kernel pre-allocated below 4GB at boot time. This is likely to result in lower performance for IO-intensive workloads for Intel® EM64T as compared to AMD64 processors.
Ah, but you see, herein lies the difference. Software running on Linux is not as tightly integrated as the Windows counterpart. Conversely, the possibility of "breaking many apps" is quite remote (an underlying lib must be doing something wrong in the interface, hence by design - and none of the many devs using it to build the many apps must notice it, let alone the lib maintainer) Loose coupling is good sometimes.
One of the problems with MS is the messed-up API - documented, undocumented, internals exposed and so on. Most if not all of this involving MS apps. And, rest assured, the main "app breaking" MS is concerned about is their own. An IE patch should not break MS Office, right?
Who said there are only 2 compilers? He only tested 2, that's all. Here's a bit of a list - and notice that some of these are targeted specifically to scientific computing:
If Microsoft wants to go to a totally closed shop mentality as far as every single piece of hardware is concerned, they will probably quickly find Apple overtaking them.
Sorry to bust your bubble, but Apple does not have enough volume to threaten MS on the desktop. There are simply too many hardware players in the PC industry when compared to the Mac one. Software, too.
For the user that goes to BestBuy/CompUSA/whatever to buy a computer will likely end up in the HUGE PC area rather than in the small Apple one. And, as they won't care what OS runs on that computer as long as they can use it for what they want, guess what OS will they be taking home? Hint: this is why most hardware must 'work with Windows' to sell decently (witness iPod).
The funny thing is, until the time an 128-bit FS will really be needed any patents Sun has on ZFS will have expired. So whatever that day's Open Source OS of choice will be, it will at least support ZFS (and probably that time's 128-bit incarnation of several of today's FS's).
Somehow, an alternate history where 80286 was 64-bit instead of 16-bit (while everything else staying the same) comes to mind when reading the Sun's marketing on this.
I agree, too many spoilers already. Sorry about getting carried away. I'll try to keep this more 'neutral':-)
However, I don't buy Niven's picture of the possible ARM protector at all - a plotline about some outside race (Puppeteers?) neutralizing the virus everywhere it was left around in Protector would be more believable. First, the para-schizo image is hilarious for a protector - as he's not human anymore (quite different endocrine system, neural system, biochemistry). Heck, by Cro-Magnon standards I might be a sick weakling, but that's kind of irrelevant. Second, my point about Kzinti still being around and kicking stands - think about the current situation, where they are suposed to be outmaneuvering humans again (tech theft, war). No protector would allow it - and if it did happen, it means whoever is in charge in the human space is quite incompetent, thus there's no need for such drastic precautions. And I don't believe the doc was modified in any significant way, either - no time (and the puppeteer would have known it anyway, especially after 2 weeks of being around by himself in his own ship, so too risky). Finally, if a human protector did get to study the puppeteer ship before losing it (or will, as it's going back to human space) then the Ringworld is just as unsafe as ever. And btw, that ship is hardly unconspicuous.
I guess the main problem with having protectors in the first place is they're supposed to be a lot more intelligent than the author. Makes it kind of hard to predict their actions, especially in complex situations.
Physics is based on causality also - at several levels, too. Who'd have thought!
Here's a clue: a model of the Universe (or parts of it) is just that - a model. Meaning it describes the howaccurately enough, but does not explain the why. And guess what - causality is part of the reason.
But hey, don't let reason stand in the way of a good troll. This is/. after all.
That will depend on your hardware. If Sun will control the license tight enough (Java-style, as they seem to imply) then ports to platforms not agreed by Sun will be forever-beta at best. Look at the bickering about Solaris on IA64; and in spite of their talks, I don't really see why they would regard Solaris on Power as more than a lab experiment (it's a competing hw platform after all, and Sun is selling hardware)
Also, there will be the issue of 'controlled innovation', Sun's way or the highway. This has good parts and bad parts, as does anything, but it will not fit everybody's teacup - just as Linux does not right now.
I'm not sure about how using a.forward file (or a procmail forwarding rule) is forging. I like to forward a copy of my mails to a web account when I'm on vacation just to make sure I can read them whether or not I have a (trusted machine with a) ssh client available (read: internet cafes). I guess it's time to change that procmail script then.
Slashdot Mirror
it's not just L glass (which would go without saying - you don't get a 1-class body for sub-par glass), you would need a L prime, L zooms might not be corrected enough for the sensor resolution.
You do realize that 800 lines/mm means 10/8=1.25 microns, right? so not quite submicron. Size of the silver cristals is not all. You could put nanometer-sized sensors and achievable resolution would be quite poor - yeah, you count photons, but if there aren't enough photons to activate every sensor then your extra resolution is useless.
Also, 800 lines/mm are the (slow) laboratory 'max. resolution' conditions. You'll not get that using a 35mm camera, as the rest of the optical system will bring your max. resolution lower. Notably, the lens' resolving power becomes important sooner.
So the correct comparison is "effective resolution" in use, not max. achievable one under very special conditions.
Yeah, but you'd better use a damn good prime lens if you want to scan the film at 4kDPI. Otherwise it's a waste - you're limited by the lens, not by the storing medium.
Sun is not supposed to be datacenter only. Workstations can be used in scientific environments where for instance image data aquisition is necessary. Or using any custom-made sensor that would require a kernel driver to interface with the workstation. Hard to have that on Solaris, a lot easier on Linux.
If you read the reply you'll see the argument that Linux suns on some hardware that does not provide diagnostics/recovery capabilities. When it's possible, Linux tries to implement it (that also depends on the spec availability in a GPL-compatible form). It's probably true that Sun does a better job at it, as they control what hardware goes into their boxes and thus all the details about error recovery information available. Also, consumer PCs don't really need a high-availability server's degree of soft error recovery.
Yes, it's not a lot of the argument, but still an important part.
Doh! Did you read the article? It's about the EC suggesting ISO. EC is not a business. It means governments might be interested in this - and gov. departments don't have the same integration requirements as businesses (also, they don't have the same funny DRM ideas). It might move some of them off MS Office, and that would be quite an achievement.
Also, if official documents are required to be submitted in OO.o format and MS Office does not support that, don't you think businesses will have to look for a supporting application? From TFA, EC asked for interoperability filters, which would exactly offering a conversion tool for businesses if MS gives this format a pass.
So yeah, I don't think this concerns businesses a lot - only a little at this stage. But it can have a snowball effect if it really catches the right start.
You seem to be making the common mistake here. It's not about Sun/OpenOffice per se, but about their document format. Which means, if it gets standardized upon by EU, then someone (presumably MS) will come up with a MS Office plugin to allow it to work with it - after all, how hard is it to make an XSLT sheet set?
The tool will be whatever integrates best with the organisation's workflow, but the format that tool will be using is the matter here.
Your logic also has a fatal flaw - it's not GPL software that's being sold. It's the "package".
.src.rpm package that you can download, rebuild and install. That would be for free, thus not commercial. Or you can get the source directly from the author.
If a particular distro gives you the choice of buying a box and downloading the isos, what is it you're buying? not the software, as that is also available for free. THE CONVENIENCE. You get the whole pack directly, plus manuals, plus some extra perks if that's the case (better access, commercial extensions, etc.). Also, you're buying some (limited) support. NOT THE GPL SOFTWARE.
Of course, if they don't provide the soft for free, then you're technically buying it (the binaries), but it's again a matter of convenience - getting a precompiled package instead of going LFS-like. There's still a way around this, though. They can provide sources , say a
Think of it this way - if you're buying a book that is also available in the public domain, what is it that you're buying?
So I guess the point is, GPL software itself should not be regulated by this as commercial software is. If the author gives it as GPL then it's free. If the packager charges money for the binary, then the packager has to comply with commercial regulations.
The FA says we're running into the Virgo Cluster - ETA "a few billion years". So there's still time to pack.
Here's a bunch of clues, anyway:
Anyway, if they manage a EAL5 certification for this, they'll be able to laugh their asses out at Defense Dept.s that use (or even more, mandate the use) of Windows on their computers. Seeing that w2k only made it (dubiously) to EAL4.
Do you really see a major RIAA member like Sony encouraging the use of mp3/no-DRM in digital music?
How did the terminator become governor?
Coming from the future, he knew the right key combination for your friendly Diebold vote counter?
[*] to quote Redhat:
For speed, Fortran is still best. Most enginering codes are in Fortran.
... maybe only if you meant development speed (not arguing program speed one way or another).
That does not compute, logically - erm
Anyway, from the very paper you pointed to, C9x does complex math better than Fortran. Interesting - I wish there were some detail to it though.
Ah, but you see, herein lies the difference. Software running on Linux is not as tightly integrated as the Windows counterpart. Conversely, the possibility of "breaking many apps" is quite remote (an underlying lib must be doing something wrong in the interface, hence by design - and none of the many devs using it to build the many apps must notice it, let alone the lib maintainer) Loose coupling is good sometimes.
One of the problems with MS is the messed-up API - documented, undocumented, internals exposed and so on. Most if not all of this involving MS apps. And, rest assured, the main "app breaking" MS is concerned about is their own. An IE patch should not break MS Office, right?
Who said there are only 2 compilers? He only tested 2, that's all. Here's a bit of a list - and notice that some of these are targeted specifically to scientific computing:
1. GCC
2. Intel compiler (Intel only)
3. Comeau
4. PathScale (Opteron only)
5. Portland Group (PGI)
6. Borland
If Microsoft wants to go to a totally closed shop mentality as far as every single piece of hardware is concerned, they will probably quickly find Apple overtaking them.
Sorry to bust your bubble, but Apple does not have enough volume to threaten MS on the desktop. There are simply too many hardware players in the PC industry when compared to the Mac one. Software, too.
For the user that goes to BestBuy/CompUSA/whatever to buy a computer will likely end up in the HUGE PC area rather than in the small Apple one. And, as they won't care what OS runs on that computer as long as they can use it for what they want, guess what OS will they be taking home? Hint: this is why most hardware must 'work with Windows' to sell decently (witness iPod).
The funny thing is, until the time an 128-bit FS will really be needed any patents Sun has on ZFS will have expired. So whatever that day's Open Source OS of choice will be, it will at least support ZFS (and probably that time's 128-bit incarnation of several of today's FS's).
Somehow, an alternate history where 80286 was 64-bit instead of 16-bit (while everything else staying the same) comes to mind when reading the Sun's marketing on this.
I agree, too many spoilers already. Sorry about getting carried away. I'll try to keep this more 'neutral' :-)
However, I don't buy Niven's picture of the possible ARM protector at all - a plotline about some outside race (Puppeteers?) neutralizing the virus everywhere it was left around in Protector would be more believable. First, the para-schizo image is hilarious for a protector - as he's not human anymore (quite different endocrine system, neural system, biochemistry). Heck, by Cro-Magnon standards I might be a sick weakling, but that's kind of irrelevant. Second, my point about Kzinti still being around and kicking stands - think about the current situation, where they are suposed to be outmaneuvering humans again (tech theft, war). No protector would allow it - and if it did happen, it means whoever is in charge in the human space is quite incompetent, thus there's no need for such drastic precautions. And I don't believe the doc was modified in any significant way, either - no time (and the puppeteer would have known it anyway, especially after 2 weeks of being around by himself in his own ship, so too risky). Finally, if a human protector did get to study the puppeteer ship before losing it (or will, as it's going back to human space) then the Ringworld is just as unsafe as ever. And btw, that ship is hardly unconspicuous.
I guess the main problem with having protectors in the first place is they're supposed to be a lot more intelligent than the author. Makes it kind of hard to predict their actions, especially in complex situations.
Physics is based on causality also - at several levels, too. Who'd have thought!
/. after all.
Here's a clue: a model of the Universe (or parts of it) is just that - a model. Meaning it describes the howaccurately enough, but does not explain the why . And guess what - causality is part of the reason.
But hey, don't let reason stand in the way of a good troll. This is
That will depend on your hardware. If Sun will control the license tight enough (Java-style, as they seem to imply) then ports to platforms not agreed by Sun will be forever-beta at best. Look at the bickering about Solaris on IA64; and in spite of their talks, I don't really see why they would regard Solaris on Power as more than a lab experiment (it's a competing hw platform after all, and Sun is selling hardware)
Also, there will be the issue of 'controlled innovation', Sun's way or the highway. This has good parts and bad parts, as does anything, but it will not fit everybody's teacup - just as Linux does not right now.
yes, but OO.o is not the only app out there.
Anyway, try the extendedPDF macro for OpenOffice - if you want something more like the Acrobat plugin for Word.
I'm not sure about how using a .forward file (or a procmail forwarding rule) is forging. I like to forward a copy of my mails to a web account when I'm on vacation just to make sure I can read them whether or not I have a (trusted machine with a) ssh client available (read: internet cafes). I guess it's time to change that procmail script then.