Depends on whether or not you value your data. Things like Seti@home and open blender can pull it off because their tasks are cpu-intensive, public data and nobody cares about timeliness of results. Also their algorithms are very cpu-intensive compared to the amout of data sent round.
If you have large quantities of private data, and your algorithm doesnt distribute as well as the massively-distributed examples, then you have less choice. You may want a large supercluster with infiniband backbone and a few terabytes of RAID-5 network storage.
You give it away to somebody you trust, who locks down the cluster and gives you a private datalink, in exchange for CPU power you cannot afford yourself. And those big datacentres ought to have big private filestores for you to use too.
I dont know about pay-for-grid, but we host CERN-originated Large Hadron Collider simulations on our cluster, as do other sites all over Europe. There isnt the heed to build a single giant linux cluster, when you can use spare cycles from high performance across the entire continent.
Could you say to the govt "we will put your back door in, but have a status page saying what you are listening to?" That would be a way to make fun of them while being compliant? Or is it some box they stick in on the monitor port that you aren't allowed near, Carnivore style.
I dont understand how to fully comply either. How can you intercept packets sent to nodes on the same WLAN, packets that dont even go through the router? Would all single-lan packets have to be routed to some listener box, also known as a "single point of failure"?.
Maybe the attack on the unis is the first step before going after community WLANs, mesh networks and the like that are completely unmanageable right now. Scary stuff.
you meant to say "skype, the telco now owned by ebay, does not have any known back doors"
Remember how lotus notes's export encryption system used to include half the secret key in part of the transmission, so the nsa/fbi could pick it up, leaving only a, what, 64 bit key to break? In closed source code, putting in back doors for the government(s) is not only technically easy, it's the kind of thing that management think is a good idea.
Look at it from the MS perspective, their consumer is: Someone who buys/subscribes to windows, owns an xbox and a WMA-enabled music device/car radio.
Anything that does not permit people to make use of media across this entire DRM-managed ecosystem is anti consumer. that includes apple (ipod only, itunes mostly useless elsewhere,;'cept for limited CD burns, itunes bought videos not playable on DVD players.
So in terms of restrictiveness, the windows/WMA drm-managed ecosystem is broader and less 'restrictive' than Blu-ray, at least to the extent that Poland's communist government used to be somewhat less restrictive than the soviet union itself. Even so, it is bollocks.
Billg does have a point though: the only long term storage mech is HDD, though not, sadly, my 70GB work SCSI HDD, which failed last week. Raid-5 SATA-2 drive arrays, perhaps.
I agree with you, it is dangerous ground. I think MySQL's concern over short term $$ got in the way of thinking "How will this go down with the rest of the OSS community"
As the same time, I note key OSS projects -gcc, apache- have not abandoned support for their products on legacy SCO unix, simply because even though they despise the company, there is no reason to treat the users of it badly. After all, they support the tools on Windows...
I have other issues with mySQSL, and it aint this. No, Its the way they view any form of data use (even over the wire) as being GPL-contaminating.
I just checked the batteries on my hp nc6000 laptop and was disappointed to see they werent part of the recall. Disappointed? Yes, because being a year old, the batteries only go up to 90% and 95% charge respectively. Getting free replacements would have brought them up to spec; hey I'd have it done every year, although that bit "only use on AC power till your replacement comes in" is a bit inconvenient.
Incidentally, wearable computers have always had a problem with military use, despite the vision of the soldier with the head-up-display, and the problem was the batteries. A bullet entering the body after passing though a li-ion battery is more poisonous, plus the batteries themselves are a fire risk (do not shoot is implicit in the 'do not penetrate' bit of the disclaimer).
Its actually vagely possible in windows via DirectX -you cannot just copy the bits of a window, because the stuff goes straight to the screen. unless you know the right tricks (there are always tricks) stuff doesnt get captured. Try doing a screen shot of a DVD on windows to check.
What we could also do (but it would be f. hard), would be to change the entire key input mechanism of both Windows and X11. Windows assumes that users may be untrusted, but all apps a user has share the same rights, and have equal access to any window hosted under that 'root window'/window session. So you can find a window::FindWindow("iexplore","lloyds"); then use the computer-based-training API to intercept all keystrokes before they happen. Why should any app be allowed to intercept keystrokes? Surely that should be a high-privelege feature, not something on by default.
Someone could still do a phishing site that grabbed that passwords and did a near-real-time attack on the real site.
I stopped using my lloyds current account for various reasons (mainly to do with awful rates of interest). One other issue I had was trying to get a change of address authenticated remotely from the US. They absolutely refused to do it without auth -which was fine- but all they accepted was a photocopy of a passport. Any fake passport jpeg would have done.
What they ought to consider is giving out USB card readers for the chip-and-pin debit cards everyone has. not only could the card to auth (if it was built in), but they could use it to kickstart an infrastructure for secure online purchase, in which the card# wasnt sent over the wire.
ClearCase: the SCM system of wonder and fear. ClearQuest: defect tracking app from hell. RatRose: UML design app -very pricy, awful quality Rational Unified Process: a design methodology that used the apps.
There are some aspects of the RUP; a little design is a good thing. But a good test suite is better than an overdesigned app, IMO.
yeah, unit tests kick RUPs overbloated process into the wilderness
I cannot get over the idea that OSS projects have been suffering from a lack of the RUP. We have been making do with distributed SCM, email and wiki collaboration, bugzilla, xUnit testing and plaintext artifacts. Oh, and well documented code.
Now that we have the RUP, we can stop all that and do fancy UML pictures showing how use cases are implemented instead. I am so overjoyed,.
I actually got a tour of the studio when they were doing the filming for The Wrong Trousers, which is where I learned their dirty secret: There is more than one gromit.
They were filming different scenes in different parts of the studio, with different copies of Wallace and Gromit in each one. So they are truly expendable: cloned for the filming, discarded afterwards.
yeah, there is nothing to scare you more than a bugrep against some random distro, "debian unstable", which can be pretty broken sometimes (like touch() not working) or a fedora release.
The costs of replicating the environment mean that you probably cannot do it for more than a few distros, which leaves the end users to fend for themselves. Bad news for RedHat: SuSE is enough of a mainstream distro to merit the effort.
One mistake of redhat is that by giving "amateurs" nothing but fedora is that it has pushed away those amateurs who do write the apps RH needs, onto other distros like Suse or Ubuntu. The platform you develop on is always the best supported. A commercial vendor (like vmware) can afford copies of RHEL to test their product, but an OSS dev team? Unlikely. RedHat's pricing model has made redhat linux too expensive for OSS dev teams to keep around just for testing, unless they really care about the platform.
There are is XP Home, XP Pro, XP Media Centre, Windows XP Tablet plus the server side Windows Server 2003, that is one distro with various licensing rules. Oh, there is also XP third-world, that only runs on a PIII or lower, and wont even host in VMWare on a modern box because the CPUID opcode still returns the real CPU. When you domain XP Pro it changes a lot of behaviour; you need to test in both modes. Remember also that WinXP is a pretty bare bones system: no word processor, no DVD player or burning 'cept maybe in the media centre release...there is little variation between the distroes because there is not much in any of them.
In Longhorn we will seen many more versions, with a split between corporate ones (takes the corporate keys) and consumer ones, and a wider spread of functionality. yes, this will make testing that much harder for those people that still do windows software.
Yes, any Java GUI app written in swing crawls unless it has a lot of memory and a good CPU behind it. But that is because swing sucks, not Java itself.
Or, to put it differently, "having sucky java apps is optional". Computation and networking can be pretty nippy. Because threading is easier, you can also make use of multi-core systems quite easily.
What you cannot do in Java is go low-level when you need to issue cache prefetch opcodes, or directly address the SS2 parallel register set (Java1.4+ hands off all floating point operations to the SS2 registers/opcodes, but doesnt SIMD parallel ops).
At the same time, how often is top-tier performance really needed. For many programs, time to market with stable apps matters more. This is potentially Java's big weakness: successor languages (e.g. Ruby) that offer even better productivity, without making performance tangibly worse.
The Suse line used to be pretty good, but the 9.3 experience left me fairly unhappy about how well it had been tested. Actually that is unfair, the core was pretty good and works well on laptops, but they had thrown in some stuff (Open Office 2.0 beta) that I dont consider ready yet, and other stuff (Evolution) in releases that were broken (the exchange bridge was, anyway).
Do commercial distros have to deliberatly go the way of fedora, "unstable, not for production use" to force people to pay to switch to the paying version. And yes, I do object to this split between "hobbyists/amateurs" and professionals. Linux was built by amateurs; it was the professionals that gave us the win9x product line.
I think I have to (provisionally) disagree with your statement that the out-of-order stuff is there because of C/C++. It is my belief that the OOO stuff, which as been in the x86 since the P6 core, is there to get around the finite #of CPU registers in the x86 design, and the gulf between CPU speed and main memory latency. Unlike risc/IA64 systems, with lots of registers for the compilers to allocate, you only have a limited number of general purpose registers. The CPU gets around this with its internal register set, scheduling operations out of order but only pulling them back in to the public register set (and raising faults) in the original order.
Now, you could argue that IA64 is a more C/C++ compiler-centric box, as it assumes plenty of time up-front to compile down code, and a compiler written by competent people. Its a lot harder to do dynamic stuff on there.
I do agree with your MMU comments, though note that i286 up had a segmented memory model which did offer better memory protection, and a strong model of executable code versus plain data. And you could program it in C and later C++, as I recall from my windows 3.x programming days. But believe me, it was not a nice paradigm to code against, partly because segments were too small (64K), and partly because some things (pointer comparision) get so very, very hard. C/C++ code does normally assume that if you have two pointers char *p,*p1; then if p1!=p2 a write to p (*p=X) does not mean that *p2==X. That is, different pointers point to different places. In Win31, there was no such guarantee.
ok. we are going back into the dregs of my brain here; I started windows programming on windows 2.04 when it *really* sucked.
-I think WM_DRAWCLIPBOARD is sent for clipboard previews, which is pretty rare. There is some optional accessory, clipbook, which does it.
-DDE is built on windows messages too. It dates from win16, where you had shared memory, and could just use postmessage to send a pointer. That::GlobalAlloc(GMEM_SHARED) flag when you allocate memory says "I want memory 640K", because that funny Lotus-Intel-MS Extended-Memory-Sharing thing used DOS interrupts to map different bits of the 1024+ memory into the 640-1024K area. DDE and some other messages, when sent with::SendMessage() have the runtime marshall stuff they point to into cross-process shared memory.
-Network DDE is even worse. And it is still built in to winxp, albeit disabled. That clipbook app uses it again.
-I havent dared look at win64. The sole use I have for the win32 API calls that occupy part of my memory is to make up passwords that get past our It naming rules:::CreateFile("",2);
yes, it still uses messages to windows, as does most of win32 things, included COM.
I think what has happened is some app had a lock on the clipboard, and it is not releasing it. A crashed app should be dealt with, but an app that hasnt crashed? reboot time, I guess. Or I write the code to get teh clipboard owner and WM_DESTROY it, but that would go too near those clipboard APIs.
I run a 2-core HT Xeon system @ work, it looks like a 4-way CPU. This is great for finding race condition bugs in my software -and other peoples.
my core troublespot appears to be cups, which will spin at 100% of a CPU within a few hours of starting. So I have to restart cups every morning. This is so, irritating. I suppose i could just get cron to reset it for me, but still.
Whereas on windows, I havent had to reboot my laptop since, what, yesterday, when the clipboard stopped working. I didnt even know the clipboard could stop working, but no, you can suddenly stop being able to cut and paste. Trust me to be the one to find out.
I've done keyboard controller hacking before now; the core in a laptop is a little 16 bit RISC engine that shares the bios flash. With a hacked KBC you can intercept keystrokes and the entire main memory image looks 100% normal, because it is. Your friend the keyboard has been 0wned instead...
yeah, we had to revert to C/C++ and x86 assember for our nuclear powered flying hospital, the one that was to circle above major cities to provide budget medical care.
With hindsight, trying to implement the entire fly-by-wire control system on a JSP page served up on the same low-end server/tomcat runtime that hosted the customer-accessible front end and the reactor GUI was a bit of a design error.
Maybe it is not so much not recognising it, as having native libraries or other stuff that needs to be recompiled into 64 bit mode.
its a bit like the maxosX to macosx86 migration: All java apps will work automatically on the x86 os, but as Java itself will be a native x86 app, the PPC emulation will not kick in for any native libraries -these will all have to be rebuilt for MacOS/x86.
Nb, java1.5.05 came out last week. Swing apps on linux appear to have better dialog box support (i.e. key entry works more reliably), but i got two SIGSEGV crashes in four hours -after a full system reboot. So I wouldn't rush to upgrade your docs.
Slashdot Mirror
Depends on whether or not you value your data. Things like Seti@home and open blender can pull it off because their tasks are cpu-intensive, public data and nobody cares about timeliness of results. Also their algorithms are very cpu-intensive compared to the amout of data sent round.
If you have large quantities of private data, and your algorithm doesnt distribute as well as the massively-distributed examples, then you have less choice. You may want a large supercluster with infiniband backbone and a few terabytes of RAID-5 network storage.
You give it away to somebody you trust, who locks down the cluster and gives you a private datalink, in exchange for CPU power you cannot afford yourself. And those big datacentres ought to have big private filestores for you to use too.
I dont know about pay-for-grid, but we host CERN-originated Large Hadron Collider simulations on our cluster, as do other sites all over Europe. There isnt the heed to build a single giant linux cluster, when you can use spare cycles from high performance across the entire continent.
-steve
How secret do CALEA warrants have to be?
Could you say to the govt "we will put your back door in, but have a status page saying what you are listening to?" That would be a way to make fun of them while being compliant? Or is it some box they stick in on the monitor port that you aren't allowed near, Carnivore style.
I dont understand how to fully comply either. How can you intercept packets sent to nodes on the same WLAN, packets that dont even go through the router? Would all single-lan packets have to be routed to some listener box, also known as a "single point of failure"?.
Maybe the attack on the unis is the first step before going after community WLANs, mesh networks and the like that are completely unmanageable right now. Scary stuff.
> Skype can't be tapped
you meant to say "skype, the telco now owned by ebay, does not have any known back doors"
Remember how lotus notes's export encryption system used to include half the secret key in part of the transmission, so the nsa/fbi could pick it up, leaving only a, what, 64 bit key to break? In closed source code, putting in back doors for the government(s) is not only technically easy, it's the kind of thing that management think is a good idea.
Look at it from the MS perspective, their consumer is: Someone who buys/subscribes to windows, owns an xbox and a WMA-enabled music device/car radio.
;'cept for limited CD burns, itunes bought videos not playable on DVD players.
Anything that does not permit people to make use of media across this entire DRM-managed ecosystem is anti consumer. that includes apple (ipod only, itunes mostly useless elsewhere,
So in terms of restrictiveness, the windows/WMA drm-managed ecosystem is broader and less 'restrictive' than Blu-ray, at least to the extent that Poland's communist government used to be somewhat less restrictive than the soviet union itself. Even so, it is bollocks.
Billg does have a point though: the only long term storage mech is HDD, though not, sadly, my 70GB work SCSI HDD, which failed last week. Raid-5 SATA-2 drive arrays, perhaps.
I agree with you, it is dangerous ground. I think MySQL's concern over short term $$ got in the way of thinking "How will this go down with the rest of the OSS community"
As the same time, I note key OSS projects -gcc, apache- have not abandoned support for their products on legacy SCO unix, simply because even though they despise the company, there is no reason to treat the users of it badly. After all, they support the tools on Windows...
I have other issues with mySQSL, and it aint this. No, Its the way they view any form of data use (even over the wire) as being GPL-contaminating.
aah you worry too much.
I just checked the batteries on my hp nc6000 laptop and was disappointed to see they werent part of the recall. Disappointed? Yes, because being a year old, the batteries only go up to 90% and 95% charge respectively. Getting free replacements would have brought them up to spec; hey I'd have it done every year, although that bit "only use on AC power till your replacement comes in" is a bit inconvenient.
Incidentally, wearable computers have always had a problem with military use, despite the vision of the soldier with the head-up-display, and the problem was the batteries. A bullet entering the body after passing though a li-ion battery is more poisonous, plus the batteries themselves are a fire risk (do not shoot is implicit in the 'do not penetrate' bit of the disclaimer).
Its actually vagely possible in windows via DirectX -you cannot just copy the bits of a window, because the stuff goes straight to the screen. unless you know the right tricks (there are always tricks) stuff doesnt get captured. Try doing a screen shot of a DVD on windows to check.
::FindWindow("iexplore","lloyds"); then use the computer-based-training API to intercept all keystrokes before they happen. Why should any app be allowed to intercept keystrokes? Surely that should be a high-privelege feature, not something on by default.
What we could also do (but it would be f. hard), would be to change the entire key input mechanism of both Windows and X11. Windows assumes that users may be untrusted, but all apps a user has share the same rights, and have equal access to any window hosted under that 'root window'/window session. So you can find a window
Someone could still do a phishing site that grabbed that passwords and did a near-real-time attack on the real site.
I stopped using my lloyds current account for various reasons (mainly to do with awful rates of interest). One other issue I had was trying to get a change of address authenticated remotely from the US. They absolutely refused to do it without auth -which was fine- but all they accepted was a photocopy of a passport. Any fake passport jpeg would have done.
What they ought to consider is giving out USB card readers for the chip-and-pin debit cards everyone has. not only could the card to auth (if it was built in), but they could use it to kickstart an infrastructure for secure online purchase, in which the card# wasnt sent over the wire.
Yes, Rational Software used to sell
ClearCase: the SCM system of wonder and fear.
ClearQuest: defect tracking app from hell.
RatRose: UML design app -very pricy, awful quality
Rational Unified Process: a design methodology that used the apps.
There are some aspects of the RUP; a little design is a good thing. But a good test suite is better than an overdesigned app, IMO.
yeah, unit tests kick RUPs overbloated process into the wilderness
I cannot get over the idea that OSS projects have been suffering from a lack of the RUP. We have been making do with distributed SCM, email and wiki collaboration, bugzilla, xUnit testing and plaintext artifacts. Oh, and well documented code.
Now that we have the RUP, we can stop all that and do fancy UML pictures showing how use cases are implemented instead. I am so overjoyed,.
I actually got a tour of the studio when they were doing the filming for The Wrong Trousers, which is where I learned their dirty secret: There is more than one gromit.
They were filming different scenes in different parts of the studio, with different copies of Wallace and Gromit in each one. So they are truly expendable: cloned for the filming, discarded afterwards.
yeah, there is nothing to scare you more than a bugrep against some random distro, "debian unstable", which can be pretty broken sometimes (like touch() not working) or a fedora release.
The costs of replicating the environment mean that you probably cannot do it for more than a few distros, which leaves the end users to fend for themselves. Bad news for RedHat: SuSE is enough of a mainstream distro to merit the effort.
One mistake of redhat is that by giving "amateurs" nothing but fedora is that it has pushed away those amateurs who do write the apps RH needs, onto other distros like Suse or Ubuntu. The platform you develop on is always the best supported. A commercial vendor (like vmware) can afford copies of RHEL to test their product, but an OSS dev team? Unlikely. RedHat's pricing model has made redhat linux too expensive for OSS dev teams to keep around just for testing, unless they really care about the platform.
There are is XP Home, XP Pro, XP Media Centre, Windows XP Tablet plus the server side Windows Server 2003, that is one distro with various licensing rules. Oh, there is also XP third-world, that only runs on a PIII or lower, and wont even host in VMWare on a modern box because the CPUID opcode still returns the real CPU. When you domain XP Pro it changes a lot of behaviour; you need to test in both modes. Remember also that WinXP is a pretty bare bones system: no word processor, no DVD player or burning 'cept maybe in the media centre release...there is little variation between the distroes because there is not much in any of them.
In Longhorn we will seen many more versions, with a split between corporate ones (takes the corporate keys) and consumer ones, and a wider spread of functionality. yes, this will make testing that much harder for those people that still do windows software.
Yes, any Java GUI app written in swing crawls unless it has a lot of memory and a good CPU behind it. But that is because swing sucks, not Java itself.
Or, to put it differently, "having sucky java apps is optional". Computation and networking can be pretty nippy. Because threading is easier, you can also make use of multi-core systems quite easily.
What you cannot do in Java is go low-level when you need to issue cache prefetch opcodes, or directly address the SS2 parallel register set (Java1.4+ hands off all floating point operations to the SS2 registers/opcodes, but doesnt SIMD parallel ops).
At the same time, how often is top-tier performance really needed. For many programs, time to market with stable apps matters more. This is potentially Java's big weakness: successor languages (e.g. Ruby) that offer even better productivity, without making performance tangibly worse.
The Suse line used to be pretty good, but the 9.3 experience left me fairly unhappy about how well it had been tested. Actually that is unfair, the core was pretty good and works well on laptops, but they had thrown in some stuff (Open Office 2.0 beta) that I dont consider ready yet, and other stuff (Evolution) in releases that were broken (the exchange bridge was, anyway).
Do commercial distros have to deliberatly go the way of fedora, "unstable, not for production use" to force people to pay to switch to the paying version. And yes, I do object to this split between "hobbyists/amateurs" and professionals. Linux was built by amateurs; it was the professionals that gave us the win9x product line.
I think I have to (provisionally) disagree with your statement that the out-of-order stuff is there because of C/C++. It is my belief that the OOO stuff, which as been in the x86 since the P6 core, is there to get around the finite #of CPU registers in the x86 design, and the gulf between CPU speed and main memory latency. Unlike risc/IA64 systems, with lots of registers for the compilers to allocate, you only have a limited number of general purpose registers. The CPU gets around this with its internal register set, scheduling operations out of order but only pulling them back in to the public register set (and raising faults) in the original order.
Now, you could argue that IA64 is a more C/C++ compiler-centric box, as it assumes plenty of time up-front to compile down code, and a compiler written by competent people. Its a lot harder to do dynamic stuff on there.
I do agree with your MMU comments, though note that i286 up had a segmented memory model which did offer better memory protection, and a strong model of executable code versus plain data. And you could program it in C and later C++, as I recall from my windows 3.x programming days. But believe me, it was not a nice paradigm to code against, partly because segments were too small (64K), and partly because some things (pointer comparision) get so very, very hard. C/C++ code does normally assume that if you have two pointers char *p,*p1; then if p1!=p2 a write to p (*p=X) does not mean that *p2==X. That is, different pointers point to different places. In Win31, there was no such guarantee.
ok. we are going back into the dregs of my brain here; I started windows programming on windows 2.04 when it *really* sucked.
::GlobalAlloc(GMEM_SHARED) flag when you allocate memory says "I want memory 640K", because that funny Lotus-Intel-MS Extended-Memory-Sharing thing used DOS interrupts to map different bits of the 1024+ memory into the 640-1024K area. DDE and some other messages, when sent with ::SendMessage() have the runtime marshall stuff they point to into cross-process shared memory.
::CreateFile("",2);
-I think WM_DRAWCLIPBOARD is sent for clipboard previews, which is pretty rare. There is some optional accessory, clipbook, which does it.
-DDE is built on windows messages too. It dates from win16, where you had shared memory, and could just use postmessage to send a pointer. That
-Network DDE is even worse. And it is still built in to winxp, albeit disabled. That clipbook app uses it again.
-I havent dared look at win64. The sole use I have for the win32 API calls that occupy part of my memory is to make up passwords that get past our It naming rules:
yes, it still uses messages to windows, as does most of win32 things, included COM.
I think what has happened is some app had a lock on the clipboard, and it is not releasing it. A crashed app should be dealt with, but an app that hasnt crashed? reboot time, I guess. Or I write the code to get teh clipboard owner and WM_DESTROY it, but that would go too near those clipboard APIs.
yes -if some storage device doesnt seamless transfer data, the only word you can use is "broken".
It'd be like a socket write() operation that only wrote approved packets.
I run a 2-core HT Xeon system @ work, it looks like a 4-way CPU. This is great for finding race condition bugs in my software -and other peoples.
my core troublespot appears to be cups, which will spin at 100% of a CPU within a few hours of starting. So I have to restart cups every morning. This is so, irritating. I suppose i could just get cron to reset it for me, but still.
Whereas on windows, I havent had to reboot my laptop since, what, yesterday, when the clipboard stopped working. I didnt even know the clipboard could stop working, but no, you can suddenly stop being able to cut and paste. Trust me to be the one to find out.
good point.
I've done keyboard controller hacking before now; the core in a laptop is a little 16 bit RISC engine that shares the bios flash. With a hacked KBC you can intercept keystrokes and the entire main memory image looks 100% normal, because it is. Your friend the keyboard has been 0wned instead...
Yes, although I thought SD memory had this kind of "feature" too, as did some of the enhanced memory sticks from sony.
But remember we consumers have been crying out for a way to move our music around freely and securely. Oh, wait a minute, I have that, its called scp.
yeah, we had to revert to C/C++ and x86 assember for our nuclear powered flying hospital, the one that was to circle above major cities to provide budget medical care.
With hindsight, trying to implement the entire fly-by-wire control system on a JSP page served up on the same low-end server/tomcat runtime that hosted the customer-accessible front end and the reactor GUI was a bit of a design error.
-steve
Maybe it is not so much not recognising it, as having native libraries or other stuff that needs to be recompiled into 64 bit mode.
its a bit like the maxosX to macosx86 migration: All java apps will work automatically on the x86 os, but as Java itself will be a native x86 app, the PPC emulation will not kick in for any native libraries -these will all have to be rebuilt for MacOS/x86.
Nb, java1.5.05 came out last week. Swing apps on linux appear to have better dialog box support (i.e. key entry works more reliably), but i got two SIGSEGV crashes in four hours -after a full system reboot. So I wouldn't rush to upgrade your docs.