Apple and Motorola introduced several incompatibilities into their PowerPC designs which made their instruction set incompatible with IBM's POWER architecture.
Eventually, Apple is either going to have to drop the usage of these extensions, or pay a lot of money to have them maintained as a separate architecture.
Whatever induced them to diverge in the first place? Was Motorla trying to lock them in as a customer?
...Nichia should endow a chair at a major research institute and arrange to have Nakamura granted tenure.
Nakamura's profession is scientific research. If relations have soured between Nichia and Nakamura to the extent that direct cooperation between them is no longer possible, then at the very least Nichia should arrange a setting where Nakamura can continue his research elsewhere.
Lots of companies endow chairs at major universities, and there are significant tax benefits for doing so. Nakamura also has obviously wasted a large part of his career on this pointless lawsuit, and might welcome such an opportunity to return to his passion.
Even if Nakamura has no interest in such an offer, the PR value for Nichia would be inestimable... right now their PR position seems very, very bad to me.
I've only used the free Sybase 11.0.3.3 server, but as I remember it has a 1932-byte limit per row. You get a warning if varchars exceed this length, and a hard error if fixed-length columns do.
Sybase and MS SQL Server used to be the same code. Did MS do away with the row limit?
First, the 8500 was designed by a drunk. This is not an easy machine to rip apart.
Second, after downloading Stuffit and trying to run it, I get an application error message. I have been unable to run XPostFacto because I can't even get Stuffit working.
Third, after moving some of the DIMMs around (thinking Stuffit's problem is RAM-related), the system won't boot at all.
Fourth, even if I did get Stuffit working, XPostFacto requires the existence of a MacOS 9 partition; you cannot natively install OSX only (it plays a trick and moves the root directory from the CD to disk filesystem). I bought a 9gig Micropolis SCSI drive to solve this problem, not that it's been much use.
This pile of junk is sitting on my kitchen floor until I have time to deal with it, which may be never.
My friend has a Linksys wireless base station and laptop; I gave him some minimal help in setting it up in "infrastructure" (unrestricted?) mode.
Unless there is some sort of "wireless sniffer" which can detect in-use MAC addresses, and also unless the wireless NIC can modify its own MAC, restricting the Linksys base station to a specific set of MACs should be sufficient for small-system security, correct?
It would be even better if I could take a MAC from an old NE2000 10Base2 ethernet card and use that MAC, since anyone trying to guess a MAC would probably use the ranges that have been assigned to the WAN manufacturers. It would be best if this could be done under Windows (in spite of my distaste for MS).
I'm just looking for a "hosts.deny" sort of security; I don't really need encryption (and I understand that wireless encryption has been broken anyway).
Pardon me for any technical errors; I'm clueless about wireless.
I don't know much about their clustering software, but I doubt that it was of the caliber of any commercial offerings of the old-school UNIX players.
From what I understand, Compaq Tru64 UNIX (formerly known as Digital UNIX, formerly known as Digital OSF/1) has the very best clustering capabilities in the industry. The native Tru64 filesystem, AdvFS, can be mounted by multiple UNIX systems at the same time, which eases cluster maintenance considerably. AdvFS is one of the important components of Tru64 that will be migrated to HP-UX (but this work is going very badly, from what I understand).
Supposedly, Oracle is releasing a clustering file system for Linux under the GPL, and it seems similar in capabilities to AdvFS. HP also has ported their MC/ServiceGuard software (the normal high-availability component of HP-UX) to Linux. With this kind of competetion, I can see why Turbolinux is hard-pressed in the clustering software arena.
However, Earth's poles have changed before...
on
Going Up?
·
· Score: 2
...and as far as I know, we don't know why. Perhaps it might be useful to find out before deployment?
Nah.
DB2? Will it scale?
on
Linuxworld Fun
·
· Score: 3, Interesting
While some may scoff at such a question, even the most basic DB2 documentation stresses the importance of keeping transactions short, due to limited resources for row-level locking and the dire effects of lock promotion on concurrency.
Conversely, Tom Kyte in his first book stresses that Oracle provides an unlimited number of row-level locks (by storing the locks on disk), and never promotes a lock.
Now, obviously, people have gotten DB2 to scale, since it powers some very large databases. I have an interest (and certifications) in both systems, but I can't help but wonder what sort of tricks must be played with the database to overcome concurrency issues with memory-based lock structures - does this require a 64-bit address space even for a moderately-sized db?
...and he is obviously ignoring the recent flap over the HP-Tru64 su vulnerability.
He said that he encourages those in the computer security field (but not anybody else) to run and attempt to crack industry software on their own computers (but not anybody else's) - ignoring the fact that this violates the DMCA - and then report any vulnerabilities to the government (as well as the manufacturer).
This seems like a tinly-veiled attempt to give the NSA a few more backdoors to me.
I vote for a 1-week courtesy notification period before a full, public disclosure - no matter who you are, or how much money you have.
PA-RISC binaries must run under an emulator on Itanium, at some performance loss over native binaries. This is a whole new platform, which HP is attempting to conceal from us.
Any business considering a move from PA-RISC to Itanium should also consider Sparc and Power since both of these established architectures have support, reliability, and scalability - points which are lacking or unproven on Itanium.
If AMD is successful in forcing Intel to adopt x86-64, great harm will be inflicted upon:
HP-UX
VMS
Tru64 (what is left of it that is rolling into HP-UX)
To a lesser extent, AIX-5L
While recent interviews with HP execs (on The Register) indicate that HP is taking some steps to "roll with the x86-64 punch," I sincerely doubt if HP can be convinced to port VMS to Opteron should it become necessary.
What is even more troubling for the Itanium is the fact that HP's compilers are faster than Intel's, but the HP compilers have not been released outside of HP-UX. The standoffish attitude of other ISVs (Dell, IBM, etc.) is not hard to understand given these circumstances.
You will also have noticed Microsoft's (now infamous) "leaked" memo on Windows-64 running on Opteron. Such a leak I believe has been carefully crafted to throw FUD upon all things Itanium. Furthermore, it is in Microsoft's best interests for Opteron to prevail, as such a victory will destroy not only DEC/Compaq's high end, but also HP (as much as HP-UX deserves to die, it should not fall to Microsoft).
If Intel and HP truly want Itanium to flourish, Intel must reduce the price immediately (to at least a SPEC-to-SPEC match with Athlon/Opteron, and possibly lower), and HP must release fast compilers under an open license.
If the Itanium market remains fragmented, AMD wins, and Microsoft's interests are advanced.
In an ideal game of dirty legal pool, with large or unlimited funding, what terrible trauma could someone like IBM (or even the MPAA) legally inflict on Forgent to shut the company down and acquire their patent, assuming that the patent stands?
I assume that the longevity of the instrument is much reduced when it is forced to rely upon solar power rather than nuclear batteries. This is a necessary evil, as it is just too dangerous to send this material into orbit with NASA's safety record.
Why am I glad that the Challenger blew up? Why should YOU be? Because the NEXT planned shuttle mission was going to carry LOTS of nuclear material.
Sounds like all that is necessary is an example of time-division multiplexing. Perhaps they could get somebody from AT&T who was doing this in the 60s.
I don't like the fact that ext3 is now included as a module. The default filesystem driver should be compiled as part of the kernel.
SGI's version of Red Hat is far preferable to Red Hat's own release for this reason.
Now, I must create and maintain an initrd on my IDE system (which was never required before), and I've also been in a crazy situation where attempting to mount an installed filesystem under ext3 caused and Oops, but changing fstab to ext2 was fine.
Down with Red Hat's use of ext3 as a module! Red Hat has never handled journaling in a reasonable manner.
Self-contained, proprietary p2p agents will be stomped by the RIIA et al. We need a core GNU distributed p2p agent. The agent should implement all downloads via stripped-down HTTP and perhaps all search functions as stripped-down LDAP. It should not be audio/video centric.
If the agent is released by some party as "GNU p2p" under the GPL, we might be able to get the FSF and also MIT to defend it, as well as the EFF.
Each of the remaining p2p players needs to endorse and convert to the new protocols before they get stomped by the RIIA. Kazaa and friends can still stamp their version with all their spyware and other "value-add."
As the agent matures to circumvent blocking techniques, the network moves as a whole.
If the thing is POSIX, this might be a good way to get a minimal Cygwin on a whole lot of systems.
The one thing that has been able to put a stop to Microsoft at this point is the GPL. Perhaps it could be useful against the RIIA - somebody should try and see. The value of the IP behind any of these systems is not that great - the GPL would be a fantastic curve-ball.
p.s. IANAL, nor am I a win32 programmer, so I really don't know what I'm talking about.
It is probably best that we get this all talked out now, and treat them as fairly as possible. However:
Have a whole bunch of guys show up with arms in plaster casts and signs on their necks with "Compaq," "Gateway," "Dell," "HP," "IBM," etc. Just mill around their booth.
How about a coffin with "Digital UNIX on Alpha - Rest In Peace - BRING THE MICROSOFT MURDERERS TO JUSTICE!"
I love Slashdot's "Bill the Borg" logo. How about a T-Shirt? Sold from a booth right next to them? Who has the booths right next to them? Will they play along?
How about a T-Shirt like this:
Windows NT Server $1,000 = Linux Free Microsoft Office $300 = Openoffice Free MS SQL Server $5000 = Sybase Free WHAT ARE YOU PEOPLE THINKING?!?!?!
Another T-Shirt: Microsoft DRM - We've got all your money, now we want your civil rights.
How about Thomas Pensfield Jackson as a keynote speaker?
There are lots of things that could be done to convince Microsoft that this is a really bad idea.
Exactly - RH62 requires an older OpenSSL for rpm dependence, but you have to install a newer version to get the SRPM to compile, which means temporarily removing the older OpenSSL RPM.
After you've built the RPMs, you remove the newer OpenSSH, put back the old one, and then when you try to install the OpenSSH RPMs you've built, rpm complains about OpenSSH dependencies, forcing an override with --force (or --nodeps, I can't remember).
I think 6.2 is still alive enough to warrant binaries at openssh.com. This is exactly the sort of thing that my Debian friends say is totally solved by apt.
MS is raising prices? Where do you see this. MS hasn't raised the price of Windows since Windows 95.
It's been fairly well-documented that Microsoft has been increasing corporate license fees.
There's been free office suites for awhile, but they're still doing poorly.
There have been several classes of free office components: a) commercial, closed source, b) KDE-QT based open-source. a) fails because there is no major backer and b) fails because KDE-QT apps lack the LGPL so industry has a hard time supporting them (AFAIK). Openoffice is the first real threat because it overcomes these problems.
Mozilla is good, but it's not an IE killer yet.
Who cares? Even if it were five times better than IE, most Windows users would never download it. Mozilla gives life to other platforms. This has never been about performance.
Lindows is practically whoring itself to get onto desktops. In a few months, we'll see if they are still "mainstream"
PCs may not be their core competence, but Wal-Mart has just proven itself willing to brave Microsoft's wrath (which the trial showed is a very real threat) to lower OS bundling costs. If Wal-Mart was willing to go this far, do you think they would hesitate to acquire Lindows as a company if they saw it as a reasonable contender?
MS hasn't yet been sentenced, so it's not yet complete.
They will be sentenced this year. Notice "2002 was the year" in my quote.
And even with a complete rewrite they still have all the concurency problems of memory-based row locks. What a stupid company. Still, for everything but power users, free Sybase Linux beats SQL Server anyday.
Slashdot Mirror
Do you need to run ad-aware every time you run kazaa?
My mistake; I based my post on this. Still, MacOS X won't run on any of the chips used in higher-end RS/6000s.
Apple and Motorola introduced several incompatibilities into their PowerPC designs which made their instruction set incompatible with IBM's POWER architecture.
Eventually, Apple is either going to have to drop the usage of these extensions, or pay a lot of money to have them maintained as a separate architecture.
Whatever induced them to diverge in the first place? Was Motorla trying to lock them in as a customer?
...Nichia should endow a chair at a major research institute and arrange to have Nakamura granted tenure.
Nakamura's profession is scientific research. If relations have soured between Nichia and Nakamura to the extent that direct cooperation between them is no longer possible, then at the very least Nichia should arrange a setting where Nakamura can continue his research elsewhere.
Lots of companies endow chairs at major universities, and there are significant tax benefits for doing so. Nakamura also has obviously wasted a large part of his career on this pointless lawsuit, and might welcome such an opportunity to return to his passion.
Even if Nakamura has no interest in such an offer, the PR value for Nichia would be inestimable... right now their PR position seems very, very bad to me.
Nichia, be a magnanimous victor.
It sounds like Novell has enough expertise to make a better vxfs than Veritas, and a better SAN/NAS device than EMC.
I wonder why they never capitalized on these markets.
I've only used the free Sybase 11.0.3.3 server, but as I remember it has a 1932-byte limit per row. You get a warning if varchars exceed this length, and a hard error if fixed-length columns do.
Sybase and MS SQL Server used to be the same code. Did MS do away with the row limit?
I bought an 8500 off ebay for around $75.
First, the 8500 was designed by a drunk. This is not an easy machine to rip apart.
Second, after downloading Stuffit and trying to run it, I get an application error message. I have been unable to run XPostFacto because I can't even get Stuffit working.
Third, after moving some of the DIMMs around (thinking Stuffit's problem is RAM-related), the system won't boot at all.
Fourth, even if I did get Stuffit working, XPostFacto requires the existence of a MacOS 9 partition; you cannot natively install OSX only (it plays a trick and moves the root directory from the CD to disk filesystem). I bought a 9gig Micropolis SCSI drive to solve this problem, not that it's been much use.
This pile of junk is sitting on my kitchen floor until I have time to deal with it, which may be never.
My friend has a Linksys wireless base station and laptop; I gave him some minimal help in setting it up in "infrastructure" (unrestricted?) mode.
Unless there is some sort of "wireless sniffer" which can detect in-use MAC addresses, and also unless the wireless NIC can modify its own MAC, restricting the Linksys base station to a specific set of MACs should be sufficient for small-system security, correct?
It would be even better if I could take a MAC from an old NE2000 10Base2 ethernet card and use that MAC, since anyone trying to guess a MAC would probably use the ranges that have been assigned to the WAN manufacturers. It would be best if this could be done under Windows (in spite of my distaste for MS).
I'm just looking for a "hosts.deny" sort of security; I don't really need encryption (and I understand that wireless encryption has been broken anyway).
Pardon me for any technical errors; I'm clueless about wireless.
This woman seems to wield a lot of power over both individual citizens and major corporations. I would like to know more about her.
I don't know much about their clustering software, but I doubt that it was of the caliber of any commercial offerings of the old-school UNIX players.
From what I understand, Compaq Tru64 UNIX (formerly known as Digital UNIX, formerly known as Digital OSF/1) has the very best clustering capabilities in the industry. The native Tru64 filesystem, AdvFS, can be mounted by multiple UNIX systems at the same time, which eases cluster maintenance considerably. AdvFS is one of the important components of Tru64 that will be migrated to HP-UX (but this work is going very badly, from what I understand).
Supposedly, Oracle is releasing a clustering file system for Linux under the GPL, and it seems similar in capabilities to AdvFS. HP also has ported their MC/ServiceGuard software (the normal high-availability component of HP-UX) to Linux. With this kind of competetion, I can see why Turbolinux is hard-pressed in the clustering software arena.
...and as far as I know, we don't know why. Perhaps it might be useful to find out before deployment?
Nah.
While some may scoff at such a question, even the most basic DB2 documentation stresses the importance of keeping transactions short, due to limited resources for row-level locking and the dire effects of lock promotion on concurrency.
Conversely, Tom Kyte in his first book stresses that Oracle provides an unlimited number of row-level locks (by storing the locks on disk), and never promotes a lock.
Now, obviously, people have gotten DB2 to scale, since it powers some very large databases. I have an interest (and certifications) in both systems, but I can't help but wonder what sort of tricks must be played with the database to overcome concurrency issues with memory-based lock structures - does this require a 64-bit address space even for a moderately-sized db?
He said that he encourages those in the computer security field (but not anybody else) to run and attempt to crack industry software on their own computers (but not anybody else's) - ignoring the fact that this violates the DMCA - and then report any vulnerabilities to the government (as well as the manufacturer).
This seems like a tinly-veiled attempt to give the NSA a few more backdoors to me.
I vote for a 1-week courtesy notification period before a full, public disclosure - no matter who you are, or how much money you have.
I'm running OpenSSH 3.4p1 on:
Do I need to rebuild these binaries? When will the OpenSSL audit be complete?
PA-RISC binaries must run under an emulator on Itanium, at some performance loss over native binaries. This is a whole new platform, which HP is attempting to conceal from us.
Any business considering a move from PA-RISC to Itanium should also consider Sparc and Power since both of these established architectures have support, reliability, and scalability - points which are lacking or unproven on Itanium.
If AMD is successful in forcing Intel to adopt x86-64, great harm will be inflicted upon:
While recent interviews with HP execs (on The Register) indicate that HP is taking some steps to "roll with the x86-64 punch," I sincerely doubt if HP can be convinced to port VMS to Opteron should it become necessary.
What is even more troubling for the Itanium is the fact that HP's compilers are faster than Intel's, but the HP compilers have not been released outside of HP-UX. The standoffish attitude of other ISVs (Dell, IBM, etc.) is not hard to understand given these circumstances.
You will also have noticed Microsoft's (now infamous) "leaked" memo on Windows-64 running on Opteron. Such a leak I believe has been carefully crafted to throw FUD upon all things Itanium. Furthermore, it is in Microsoft's best interests for Opteron to prevail, as such a victory will destroy not only DEC/Compaq's high end, but also HP (as much as HP-UX deserves to die, it should not fall to Microsoft).
If Intel and HP truly want Itanium to flourish, Intel must reduce the price immediately (to at least a SPEC-to-SPEC match with Athlon/Opteron, and possibly lower), and HP must release fast compilers under an open license.
If the Itanium market remains fragmented, AMD wins, and Microsoft's interests are advanced.
In an ideal game of dirty legal pool, with large or unlimited funding, what terrible trauma could someone like IBM (or even the MPAA) legally inflict on Forgent to shut the company down and acquire their patent, assuming that the patent stands?
I assume that the longevity of the instrument is much reduced when it is forced to rely upon solar power rather than nuclear batteries. This is a necessary evil, as it is just too dangerous to send this material into orbit with NASA's safety record.
Why am I glad that the Challenger blew up? Why should YOU be? Because the NEXT planned shuttle mission was going to carry LOTS of nuclear material.
Sounds like all that is necessary is an example of time-division multiplexing. Perhaps they could get somebody from AT&T who was doing this in the 60s.
I don't like the fact that ext3 is now included as a module. The default filesystem driver should be compiled as part of the kernel.
SGI's version of Red Hat is far preferable to Red Hat's own release for this reason.
Now, I must create and maintain an initrd on my IDE system (which was never required before), and I've also been in a crazy situation where attempting to mount an installed filesystem under ext3 caused and Oops, but changing fstab to ext2 was fine.
Down with Red Hat's use of ext3 as a module! Red Hat has never handled journaling in a reasonable manner.
The one thing that has been able to put a stop to Microsoft at this point is the GPL. Perhaps it could be useful against the RIIA - somebody should try and see. The value of the IP behind any of these systems is not that great - the GPL would be a fantastic curve-ball.
It is probably best that we get this all talked out now, and treat them as fairly as possible. However:
There are lots of things that could be done to convince Microsoft that this is a really bad idea.
Exactly - RH62 requires an older OpenSSL for rpm dependence, but you have to install a newer version to get the SRPM to compile, which means temporarily removing the older OpenSSL RPM.
After you've built the RPMs, you remove the newer OpenSSH, put back the old one, and then when you try to install the OpenSSH RPMs you've built, rpm complains about OpenSSH dependencies, forcing an override with --force (or --nodeps, I can't remember).
I think 6.2 is still alive enough to warrant binaries at openssh.com. This is exactly the sort of thing that my Debian friends say is totally solved by apt.
It's been fairly well-documented that Microsoft has been increasing corporate license fees.
There have been several classes of free office components: a) commercial, closed source, b) KDE-QT based open-source. a) fails because there is no major backer and b) fails because KDE-QT apps lack the LGPL so industry has a hard time supporting them (AFAIK). Openoffice is the first real threat because it overcomes these problems.
Who cares? Even if it were five times better than IE, most Windows users would never download it. Mozilla gives life to other platforms. This has never been about performance.
PCs may not be their core competence, but Wal-Mart has just proven itself willing to brave Microsoft's wrath (which the trial showed is a very real threat) to lower OS bundling costs. If Wal-Mart was willing to go this far, do you think they would hesitate to acquire Lindows as a company if they saw it as a reasonable contender?
They will be sentenced this year. Notice "2002 was the year" in my quote.
And even with a complete rewrite they still have all the concurency problems of memory-based row locks. What a stupid company. Still, for everything but power users, free Sybase Linux beats SQL Server anyday.
...rather than rebuilding. Generating new RPMs for 6.2 is a total nightmare.