Does anyone here know more about this "New client-to-server protocol" they speak of?
They needed to change the protocol to support such nifty features as proper error codes (SQLSTATE I belive) and better transaction control. I'm hoping that we'll see multi-stage transactions in 7.5.
Older clients can still talk to newer servers, they just won't support the new features.
ERServer was released open source months ago. Check out GBorg for more information.
And why do YOU need raw disk access? The PostgreSQL developers belive (and rightly I think) that the operating system can do the caching better than they can. Why re-write the wheel? Operating systems have came a long way in the last 10 or 12 years. Days of slow access to the disk are long gone.
They make no pretense of being widely cross-platform, and do things like this, which break basic 'rules' of the UNIX system, for tweaky/performance reasons.
Huh???
No, I would say that FreeBSD is the 'Stable over features'BSD version. The interface has not changed, your shell scripts, mounts, etc, etc will work just fine.
BTW, this was done for anything but performance. If anything, this will hurt performance of those binaries (but then again, how often do you really really use those binaries anyway???). It was done to allow more functionality(sp) for the system.
See your post for how to misconstrue copyright law.
Sure that one copy is under the LGPL. But as copyright holder, I am NOT constrained by that license. I can create a BSD, public domain or even take it private.
As long as I either get copyright assignments, OR do not take contributions at all, I can do whatever I want with my code.
You have to use a hack to install OS X on any pre-G3 Mac, no matter what processor upgrade card is stuck inside it. Anything but public betas will detect the older hardware during the initial CD boot and politely explain that you have incompatible hardware.
Sorry, a friend told me that his 7300-G3 upgrade installed OS X 10.2 just fine...
The only thing I've got here is 7300-200's and have installed OS X 10.1.4 (the only OS X disk I have) using Xpostfacto.
Yes, I do know that my hardware is unsupported on anything after 9.1.X, but I'm about to nuke the OS X install and just run 9.X. OS X is too much of a disk hog for the 2 2GB SCSI disks the 7300 has in it.:)
Not by default. I've got two 7300-200 here and the only way I was able to install OS X was using a hack to boot from 9.1 first. OS X will run fine a PPC604, Apple just does not want to support older hardware. It's a different story if you install a G3/G4 upgrade card first.
OS X will not install onto anything besides G3 (and not even all of those), G4 and G5s', without a hack.
Mac OS X has about as much in common with Mach as with FreeBSD, if not less. The kernel (check out Darwin) looks very LITTLE like Mach. No single servers, and it's not a true microkernel. What they did was take a BSD single server, hardwire the connections and optimise it. Then add years of hacking...:)
Oh and to add flamebait. If Mac OS X is not FreeBSD based, then why call Linux GNU/Linux? What matters most? The kernel or the userland...:)
So who else has contributed broken software to a 1.0 release?
I'm the reason why the mitsumi CD-ROM driver was broken, and of course Rod had just cut the gold master (and back then it was a major pain to make masters) and could not update the sources.
After about 20 patches, he just just gave me a commit bit...:)
Wrong question. Being that there is no (AFAIK) compiler that even conforms to the existing standards, the only change in the language in the next 4 or 5 years would be to make this happen.
Well both gcc 3.X and MS.NET come real close it. In both of them I can take any of the examples in "The C++ Programming Langauge" or from the standard and they just work. They both have bugs, but I would say they are about 95% of the way there. They also cover atleast 95% of the available machines.
BTW, did you notice the date on the "Portable C++ Guide"? It's from back in 1998. (Updated 2 years ago) Over three quarters of the stuff is way out of date. e.g. C++ comments are now legal C. Who still uses Visual C++ 1.5? Who still uses a CFront based compiler? Does any modern OS?
Well then, there must be a whole lot of crappy C and C++ programmers and no crappy Java, Python, Ruby, Smalltalk, Lisp programmers.
I can't speak for C programmers, but for C++, not using the standard library is one of the major sins.
There are plenty of programmers that use only a little bit or none of the standard library. One of Stroustrup's primary examples is for inputing a string and it has ZERO chance of a memory leak. It uses Standard string and vectors (both of which do automatic memory allocation). For a good book on learning C++ lookup Koenig and Moo's "Accelerated C++".
In well written C++, the chances of a buffer overrun or memory leak is almost zero. The problem is finding well written code...:(
Actually, if you read the IBM contract addendum, it does actually say it is irrevocable. Yes it does. It just means that the license does not have to be paid for. Here is the full clause from earlier:
1. No Addition Royalty. Upon payment to SCO of the consideration in the section entitled "Consideration", IBM will have the irrevocable, fully paid-up, perpetual right to exercise all of its rights under the Related Agreements beginning January 1, 1995 at no additional royalty fee.
Now look at the bold text, what DOES the Related Agreements say? That is the question. That addendum relates soley to the royalties. If the Related Agreements say they can issue UNIX licenses except on days of the full moon, it means just that. They can issue licenses everyday with no royalties, EXCEPT on days of the full moon. If they do so anyway, they are in violation.
For a cheap amp, you can buy 1 watt 2.4ghz amp IC's for very cheap. Here is a site that has a schematic and board layout for it.
And very illegal in the US. Your not allowed to build your own gear unless your a Amateur radio operator operating in the Amateur bands. I'm willing to bet that Canada has the same kind of regs...
And don't forget about being obnoxiusly anti-pagan. When a Wiccan coven got permission to hold circle on a base, he went off the deep end. You'll find his rants on Google...
It depends. As long as you don't bother with one of the primary users of the band you should be ok. Or not. It depends.
If you start kicking it up too much (or try to use a power amp instead of just an antenna), you might catch the notice of someone. What happens then depends on who it is. If it's a Ham, she might just tell you about the problem and drop it. On the other hand a commericial company will most likely tell the FCC and then you get a nice little letter in the mail after the van has drove by a couple of times.
It's definitely possible to write C++ code that doesn't do this crap.
Yeap, sure is. No buffer overruns, no mem leaks, no dangling pointers, etc, etc. All possiable with STD C++; If you add the Boost and ACE libs then you have just about everything needed to do system work. (Check out Koenig and Moo for a good way to learn C++, INSTEAD of trying to learn C first) Wether you want OO, generic or functional programming, C++ can do it...:)
Thats why I use Perl and C++, langauges that don't try to tell ME how to program, they just let me get the work done.
BSDI agreed to substitute a port of the University of California's in a release which became known as 4.4 BSD(Lite) for BSD/386.
Just as a side note, BSDI WAS based on 4.4BSD-Lite. As was 386BSD (which birthed both FreeBSD and NetBSD). After the settlement, BSDI, FreeBSD and NetBSD all restarted development using 4.4BSD-Lite2 per the agreement. For FreeBSD this was the dividing line between the 1.X and 2.X series.
With the advent of the release of 32V, people could once again distribute FreeBSD 1.X (why, I don't know!)
Someone correct me if I'm wrong, but I think Linux back in the very early 90's used the Minix filesystem prior to the ext series. I never did figure out why they created a whole new filesystem from scratch instead of just grabbing the BSD ufs (that just about everyone else used at the time).
Except the Linux compatibility layer will be illegal thanks to SCO:(
Well it would be if there were any real Linux kernel source code in the layer, but there is not. It's mostly a simple translation between Linux system calls and BSD system calls. Thats why the Linux compat module is NOT GPL'ed...
About the only place this applies is to trademarks.
The most that will usally happen is that the courts will rule that you can't get money out of someone if you've sat around too long.
But selective copyright/patent enforcement? I can do that all day long. What was that in the news about Red Hat saying they would not enforce their patents? Same thing. They can go after the big players and leave the small fish alone and the courts won't say a word...
Slashdot Mirror
Does anyone here know more about this "New client-to-server protocol" they speak of?
They needed to change the protocol to support such nifty features as proper error codes (SQLSTATE I belive) and better transaction control. I'm hoping that we'll see multi-stage transactions in 7.5.
Older clients can still talk to newer servers, they just won't support the new features.
Huh?
ERServer was released open source months ago. Check out GBorg for more information.
And why do YOU need raw disk access? The PostgreSQL developers belive (and rightly I think) that the operating system can do the caching better than they can. Why re-write the wheel? Operating systems have came a long way in the last 10 or 12 years. Days of slow access to the disk are long gone.
BWP
They make no pretense of being widely cross-platform, and do things like this, which break basic 'rules' of the UNIX system, for tweaky/performance reasons.
Huh???
No, I would say that FreeBSD is the 'Stable over features'BSD version. The interface has not changed, your shell scripts, mounts, etc, etc will work just fine.
BTW, this was done for anything but performance. If anything, this will hurt performance of those binaries (but then again, how often do you really really use those binaries anyway???). It was done to allow more functionality(sp) for the system.
BWP
See your post for how to misconstrue copyright law.
Sure that one copy is under the LGPL. But as copyright holder, I am NOT constrained by that license. I can create a BSD, public domain or even take it private.
As long as I either get copyright assignments, OR do not take contributions at all, I can do whatever I want with my code.
BWP
You have to use a hack to install OS X on any pre-G3 Mac, no matter what processor upgrade card is stuck inside it. Anything but public betas will detect the older hardware during the initial CD boot and politely explain that you have incompatible hardware.
Sorry, a friend told me that his 7300-G3 upgrade installed OS X 10.2 just fine...
The only thing I've got here is 7300-200's and have installed OS X 10.1.4 (the only OS X disk I have) using Xpostfacto.
Yes, I do know that my hardware is unsupported on anything after 9.1.X, but I'm about to nuke the OS X install and just run 9.X. OS X is too much of a disk hog for the 2 2GB SCSI disks the 7300 has in it.:)
BWP
Not by default. I've got two 7300-200 here and the only way I was able to install OS X was using a hack to boot from 9.1 first.
OS X will run fine a PPC604, Apple just does not want to support older hardware. It's a different story if you install a G3/G4 upgrade card first.
OS X will not install onto anything besides G3 (and not even all of those), G4 and G5s', without a hack.
BWP
Mac OS X is based on Mach.
Mac OS X has about as much in common with Mach as with FreeBSD, if not less. The kernel (check out Darwin) looks very LITTLE like Mach. No single servers, and it's not a true microkernel. What they did was take a BSD single server, hardwire the connections and optimise it. Then add years of hacking...:)
Oh and to add flamebait. If Mac OS X is not FreeBSD based, then why call Linux GNU/Linux? What matters most? The kernel or the userland...:)
BWP
So who else has contributed broken software to a 1.0 release?
I'm the reason why the mitsumi CD-ROM driver was broken, and of course Rod had just cut the gold master (and back then it was a major pain to make masters) and could not update the sources.
After about 20 patches, he just just gave me a commit bit...:)
BWP
We used to charge $200 for replacing that fuse too, was probably the most lucrative repair we did.
Sounds like Pac Bell designed it that way...:)
BWP
I also worked in a couple of shops in my IT infancy. I killed a number of motherboards swapping keyboards and mice between computers.
Sounds like they might NOT have been ps/2 periphs... The old AT keyboards could fry, but frying a serial mouse? That would take some doing.
BWP
Why remove it? It's not under the GPL...:)
Sure, the Linux libraries are GPL, but the emulation/translator is not.
BWP
Wrong question. Being that there is no (AFAIK) compiler that even conforms to the existing standards, the only change in the language in the next 4 or 5 years would be to make this happen.
.NET come real close it. In both of them I can take any of the examples in "The C++ Programming Langauge" or from the standard and they just work. They both have bugs, but I would say they are about 95% of the way there. They also cover atleast 95% of the available machines.
Well both gcc 3.X and MS
BTW, did you notice the date on the "Portable C++ Guide"?
It's from back in 1998. (Updated 2 years ago) Over three quarters of the stuff is way out of date.
e.g. C++ comments are now legal C.
Who still uses Visual C++ 1.5?
Who still uses a CFront based compiler? Does any modern OS?
BWP
Well then, there must be a whole lot of crappy C and C++ programmers and no crappy Java, Python, Ruby, Smalltalk, Lisp programmers.
:(
I can't speak for C programmers, but for C++, not using the standard library is one of the major sins.
There are plenty of programmers that use only a little bit or none of the standard library. One of Stroustrup's primary examples is for inputing a string and it has ZERO chance of a memory leak. It uses Standard string and vectors (both of which do automatic memory allocation). For a good book on learning C++ lookup Koenig and Moo's "Accelerated C++".
In well written C++, the chances of a buffer overrun or memory leak is almost zero. The problem is finding well written code...
BWP
Actually, if you read the IBM contract addendum, it does actually say it is irrevocable. Yes it does. It just means that the license does not have to be paid for. Here is the full clause from earlier:
1. No Addition Royalty. Upon payment to SCO of the consideration in the section entitled "Consideration", IBM will have the irrevocable, fully paid-up, perpetual right to exercise all of its rights under the Related Agreements beginning January 1, 1995 at no additional royalty fee.
Now look at the bold text, what DOES the Related Agreements say? That is the question. That addendum relates soley to the royalties. If the Related Agreements say they can issue UNIX licenses except on days of the full moon, it means just that. They can issue licenses everyday with no royalties, EXCEPT on days of the full moon. If they do so anyway, they are in violation.
BWP
For a cheap amp, you can buy 1 watt 2.4ghz amp IC's for very cheap. Here is a site that has a schematic and board layout for it.
And very illegal in the US. Your not allowed to build your own gear unless your a Amateur radio operator operating in the Amateur bands. I'm willing to bet that Canada has the same kind of regs...
BWP
And don't forget about being obnoxiusly anti-pagan. When a Wiccan coven got permission to hold circle on a base, he went off the deep end. You'll find his rants on Google...
BWP
It depends. As long as you don't bother with one of the primary users of the band you should be ok. Or not. It depends.
If you start kicking it up too much (or try to use a power amp instead of just an antenna), you might catch the notice of someone. What happens then depends on who it is. If it's a Ham, she might just tell you about the problem and drop it. On the other hand a commericial company will most likely tell the FCC and then you get a nice little letter in the mail after the van has drove by a couple of times.
BWP (aka N5VMF/7)
It's definitely possible to write C++ code that doesn't do this crap.
Yeap, sure is. No buffer overruns, no mem leaks, no dangling pointers, etc, etc. All possiable with STD C++; If you add the Boost and ACE libs then you have just about everything needed to do system work. (Check out Koenig and Moo for a good way to learn C++, INSTEAD of trying to learn C first) Wether you want OO, generic or functional programming, C++ can do it...:)
Thats why I use Perl and C++, langauges that don't try to tell ME how to program, they just let me get the work done.
BWP
As a C++ programmer, this doesn't make sense to me. C++ supports both prefix and postfix operators. Any idea what the author was talking about?
Just the fact that the prefix operators don't require a temp variable; so if you just need the side effect, use prefix.
BTW, you don't have to do this, but it's just about idomatic for C++.
BWP
BSDI agreed to substitute a port of the University of California's in a release which became known as 4.4 BSD(Lite) for BSD/386.
Just as a side note, BSDI WAS based on 4.4BSD-Lite. As was 386BSD (which birthed both FreeBSD and NetBSD). After the settlement, BSDI, FreeBSD and NetBSD all restarted development using 4.4BSD-Lite2 per the agreement. For FreeBSD this was the dividing line between the 1.X and 2.X series.
With the advent of the release of 32V, people could once again distribute FreeBSD 1.X (why, I don't know!)
BWP
Huh?
Someone correct me if I'm wrong, but I think Linux back in the very early 90's used the Minix filesystem prior to the ext series. I never did figure out why they created a whole new filesystem from scratch instead of just grabbing the BSD ufs (that just about everyone else used at the time).
BWP
Except the Linux compatibility layer will be illegal thanks to SCO :(
Well it would be if there were any real Linux kernel source code in the layer, but there is not. It's mostly a simple translation between Linux system calls and BSD system calls. Thats why the Linux compat module is NOT GPL'ed...
BWP
AFAIK FreeBSD forked NetBSD which is geared towards embedded devices (and other obscure hardware ;-)
Huh? Wrong BSD sorry. FreeBSD and NetBSD were both based on 386BSD.
OpenBSD is the one that forked from NetBSD.
BWP
Huh?
About the only place this applies is to trademarks.
The most that will usally happen is that the courts will rule that you can't get money out of someone if you've sat around too long.
But selective copyright/patent enforcement? I can do that all day long. What was that in the news about Red Hat saying they would not enforce their patents? Same thing. They can go after the big players and leave the small fish alone and the courts won't say a word...
BWP
Yeah, and the only problem is that if you lose, you get to pay THEIR lawyers bills...
Not fun..
BWP