Yeah, I guess the conclusion one can draw is you can develop better, more diverse, and stronger products with an open source model than a closed one. This general rule applies for products from the dark side as well as from the good guys.
I'll agree with that USPS is the best way to ship packages. For me they've always been more reliable, and faster than UPS, and signifcantly less expensive that FedEx. Recently, the fact that they offered USPS shipping was the sole reason I bought Civ:CTP from Linux Central instead of Linux Mall. I'll almost always choose the USPS shipping vendor over the UPS shipping only one.
Personally I lean towards Books&Bytes (www.bytes.com) for my technical books. It was better back in their brick and mortar days when you could page through books, but I'd doubt they'd spam people or do other annoying things. They've never been flashy or a hot market property or anything, just a book retailer that only sells computer books and has been doing so for about 12 years.
Almost all of this guy's points have been addressed. They are solved problems. I'd like him to install a system with a configuation like mine, use KDE and autorun. Then you have an interface that addresses his complaints, plus all your cds automatically mount and unmount (and do so without making you wait for extended periods of time, or giving you error messages that intimidate the user, as Windows does). As for Windows, it's UI is quite inconsistent. See the "UI hall of shame" (don't remember the URL) to see some examples.
This is true, but I don't see the problem. Let's take my system for example. I have 4 PCI cards in the system. The network card's driver is in the kernel. The SCSI card's driver is in the kernel. The Sound Card's driver is in the kernel. The Graphics Card's drive is in user space. Why? Beats the hell out of me. Sure the graphics card driver could take down the kernel, but so could any of the other ones.
As was pointed out before NUMA allows for many more processors than your avarage Sparc. Furthermore, as was also pointed out, it's not really dependent on a certain type of chips. It sounds like the future direction of NUMA will involve IA-64 and PPC chips.
First of all you're quite ignorant as to how different systems work. No kernel that I know of has a GUI in it. That includes NT's. NT has a component of the kernel called GDI which provides very limited graphics in the kernel. That makes things a lot faster, and really isn't the reason for instability. The GGI project in the linux world has similar goals, and IMHO, they are on the right track. Now the problem with NT is the windowing system. I don't have my roomate's copy of "Inside Windows NT" with me to quote the dlls and exes, but basically they way that it works is that the system boots, the GUI starts. However, if the GUI dies, the startup processes tell the kernel to die, because the assumption is that you're pretty fucked without that GUI. It would be like init under linux monitoring X after X starts and shutting down the machine if it finds X isn't running. This is a very poor design, but it has nothing to do with the GDI layer in the NT kernel.
Re:This doesn't say they're supporting linux to me
on
RS/6000 Linux Box
·
· Score: 1
I didn't say it was satire. I said people misread and misinterpreted what was said in both cases.
This doesn't say they're supporting linux to me.
on
RS/6000 Linux Box
·
· Score: 1
Although IBM may support linux on the RS6k, this article sure doesn't say that. "Linux-oriented bundled solutions" is usually market speak for "we'll try to make the machines play nicely with your existing and future linux systems" not "we'll offer linux with tha machine". I think joins the BeOS/Amiga article in the category of "we misread that" posts. Two in one day.
You give Bill Gates too much credit. Despite the official line, MS is not a bunch of inovators under the command of the great Bill Gates. It is a company that reacts well to the moves of others. If Bill Gates was never born and Steve Jobs was never born, and Xerox execs still didn't know what they were sitting on, someone else would have come along and stolen the technology.
Almsot everyone pays for comercial compilers. GCC and Cygnus make up a very small part of the market. The optimizations of GCC on say the Sparc architecture suck. Outside of the world of free unices, people pay for compilers most of the time. Sometimes they come bundled with the OS (this practice predates gcc) and sometimes they don't.
Re:My thoughts on corporations using BSD code
on
Feature:GPL vs BSD
·
· Score: 1
Actually the NT microkernel is supposed to be pretty decent (at least according to a number of people I knew who worked on different parts of NT). It's the Win32 API and various other OS pieces above the microkernel that is shit. Therefore, if they tacked their various bits of ugly code on top of BSD UNIX, little would change.
Of course, in the world of commercial unices, there's CDE. Kernel code for various commercial unices is pretty decent, but CDE is awful code (IMHO). At least in the unix world, KDE is becoming more and more accepted, and it's code is pretty damn nice (the X server's code isn't too bad either)
Let's say linux was released under the BSD license. Let's say MS decided to make a closed version of linux. Let's say most people did not know any better than to think MS invented it. So what?
Would any of these people contribute in any meaningful way to the linux codebase anyways? The only difference it would make is that the OS on most people's desk wouldn't suck, and MS would make money. That doesn't bother me.
Furthermore, may I point out that free versions of BSD UNIX still are alive and kicking despite proprietary versions by Sun (Solaris still has a large chunck on BSD code, even though it's now pretty SysV, however, old SunOS was BSD to the core), and BSDI. Fragmentation amongst the free versions has been much more of an issue than the existence of proprietary versions
Well, it depends on how you define freedom. With a BSD license no one can take away the code you released under the BSD license. However, they can add on it and make the add-on code non-free. I consider the BSD license more free, personally, because it does not impose restrictions upon the way others release their own code.
Actually, linux supported the BeBox two years ago (I haven't checked up on the port since). However, it did not support SMP (which kind of is important on a BeBox).
It may be the exact opposite. Some of Be's newsletters have said they were interested in getting into the "internet device" market. Furthermore, they have, with the permission of the actual developers, incorporated some linux code into parts of the OS (some of Be's ethernet drivers, actually are wrappers around Donald Becker's ethernet drivers), and I believe they were incorporating some more linux code in (not sure where, not sure when).
GNOME has acheived some nice things, pushed the idea of pervasive corba (even if they haven't acheived it themselves). However, out of all the free software projects I've come accross, GNOME is perhaps the most offensive with its attitude of 1.0 expiriments, instability, flamage and FUD.
GNOME should not be considered an idicator of a modern linux system, but rather an exception to the way things are and should be done.
Give me any package from any distribution and I'll get it working on any other distribution in an hour or so, tops.
Furthermore, distributions package the code. They don't write the code. The code is the same accross all distribs, they just are package a bit different.
Complete code forking (aka when two branches form, they never converge again, and they become increasingly incompatable) is not an issue of the BSD license. It is a result of the BSD culture. Mind you, I consider there to be far worse things in the world than forking.
Furthermore, there has been only one fork in linux systems in recent memory, the libc5 vs. glibc debate, and that will end at some point (hopefully soon). The kernel has not forked. I run a system that only marginally resembles it's original distribution (a ditrib named Nomad written by a couple of guys who lived in the same dorm as me). A lot of the system has gotten changed sine the Nomad install. You know what though, I can still run damn near every package out of Slackware, Redhat, SuSE, ect. That's not forking in my opinion.
So if you don't like it come up with your own certification mark. Anonymous coward certified if you will.
The OSI mark is only as strong as people's trust in the OSI. If you don't like ESR and/or other OSI people, then don't worry about their seal of approval. If people respect their opinions, then the mark means something.
This is absolutely false. IMHO, the BSD license is more free (as in speech, not beer) that the GPL. I can do anything I want with the source, including making non-free products from it. However, I cannot restrict others freedoms by not allowing them to develop and distribute their own free code.
This is an old argument though (and after watching Eric Raymond and Theo DeRaadt battle it out there is little more I can add to either camp). Both licenses are clearly open source, IMHO. Others have given me access to their code to modify, improve, and redstribute, and squabbles aside, that's what matters.
Slashdot Mirror
Yeah, I guess the conclusion one can draw is you can develop better, more diverse, and stronger products with an open source model than a closed one. This general rule applies for products from the dark side as well as from the good guys.
I'll agree with that USPS is the best way to ship packages. For me they've always been more reliable, and faster than UPS, and signifcantly less expensive that FedEx. Recently, the fact that they offered USPS shipping was the sole reason I bought Civ:CTP from Linux Central instead of Linux Mall. I'll almost always choose the USPS shipping vendor over the UPS shipping only one.
Personally I lean towards Books&Bytes (www.bytes.com) for my technical books. It was better back in their brick and mortar days when you could page through books, but I'd doubt they'd spam people or do other annoying things. They've never been flashy or a hot market property or anything, just a book retailer that only sells computer books and has been doing so for about 12 years.
Almost all of this guy's points have been addressed. They are solved problems. I'd like him to install a system with a configuation like mine, use KDE and autorun. Then you have an interface that addresses his complaints, plus all your cds automatically mount and unmount (and do so without making you wait for extended periods of time, or giving you error messages that intimidate the user, as Windows does). As for Windows, it's UI is quite inconsistent. See the "UI hall of shame" (don't remember the URL) to see some examples.
This is true, but I don't see the problem. Let's take my system for example. I have 4 PCI cards in the system. The network card's driver is in the kernel. The SCSI card's driver is in the kernel. The Sound Card's driver is in the kernel. The Graphics Card's drive is in user space. Why? Beats the hell out of me. Sure the graphics card driver could take down the kernel, but so could any of the other ones.
As was pointed out before NUMA allows for many more processors than your avarage Sparc. Furthermore, as was also pointed out, it's not really dependent on a certain type of chips. It sounds like the future direction of NUMA will involve IA-64 and PPC chips.
First of all you're quite ignorant as to how different systems work. No kernel that I know of has a GUI in it. That includes NT's. NT has a component of the kernel called GDI which provides very limited graphics in the kernel. That makes things a lot faster, and really isn't the reason for instability. The GGI project in the linux world has similar goals, and IMHO, they are on the right track. Now the problem with NT is the windowing system. I don't have my roomate's copy of "Inside Windows NT" with me to quote the dlls and exes, but basically they way that it works is that the system boots, the GUI starts. However, if the GUI dies, the startup processes tell the kernel to die, because the assumption is that you're pretty fucked without that GUI. It would be like init under linux monitoring X after X starts and shutting down the machine if it finds X isn't running. This is a very poor design, but it has nothing to do with the GDI layer in the NT kernel.
I didn't say it was satire. I said people misread and misinterpreted what was said in both cases.
Although IBM may support linux on the RS6k, this article sure doesn't say that. "Linux-oriented bundled solutions" is usually market speak for "we'll try to make the machines play nicely with your existing and future linux systems" not "we'll offer linux with tha machine". I think joins the BeOS/Amiga article in the category of "we misread that" posts. Two in one day.
You give Bill Gates too much credit. Despite the official line, MS is not a bunch of inovators under the command of the great Bill Gates. It is a company that reacts well to the moves of others. If Bill Gates was never born and Steve Jobs was never born, and Xerox execs still didn't know what they were sitting on, someone else would have come along and stolen the technology.
There is no porting involved. The posix complient layer of BeOS is quite passable. It's usually a simple recompile.
Almsot everyone pays for comercial compilers. GCC and Cygnus make up a very small part of the market. The optimizations of GCC on say the Sparc architecture suck. Outside of the world of free unices, people pay for compilers most of the time. Sometimes they come bundled with the OS (this practice predates gcc) and sometimes they don't.
Actually the NT microkernel is supposed to be pretty decent (at least according to a number of people I knew who worked on different parts of NT). It's the Win32 API and various other OS pieces above the microkernel that is shit. Therefore, if they tacked their various bits of ugly code on top of BSD UNIX, little would change.
Of course, in the world of commercial unices, there's CDE. Kernel code for various commercial unices is pretty decent, but CDE is awful code (IMHO). At least in the unix world, KDE is becoming more and more accepted, and it's code is pretty damn nice (the X server's code isn't too bad either)
So what?
Let's say linux was released under the BSD license. Let's say MS decided to make a closed version of linux. Let's say most people did not know any better than to think MS invented it. So what?
Would any of these people contribute in any meaningful way to the linux codebase anyways? The only difference it would make is that the OS on most people's desk wouldn't suck, and MS would make money. That doesn't bother me.
Furthermore, may I point out that free versions of BSD UNIX still are alive and kicking despite proprietary versions by Sun (Solaris still has a large chunck on BSD code, even though it's now pretty SysV, however, old SunOS was BSD to the core), and BSDI. Fragmentation amongst the free versions has been much more of an issue than the existence of proprietary versions
Well, it depends on how you define freedom. With a BSD license no one can take away the code you released under the BSD license. However, they can add on it and make the add-on code non-free. I consider the BSD license more free, personally, because it does not impose restrictions upon the way others release their own code.
Actually, linux supported the BeBox two years ago (I haven't checked up on the port since). However, it did not support SMP (which kind of is important on a BeBox).
No,
Pentium class CPU: $10-30
Hard disk: $10 (I bought a 650 MB IDE off a roomate for 5, I figure new it might fetch 10).
That drops the cost to $42-62.
It may be the exact opposite. Some of Be's newsletters have said they were interested in getting into the "internet device" market. Furthermore, they have, with the permission of the actual developers, incorporated some linux code into parts of the OS (some of Be's ethernet drivers, actually are wrappers around Donald Becker's ethernet drivers), and I believe they were incorporating some more linux code in (not sure where, not sure when).
"Expirimental" 1.0 software. That's just wrong.
GNOME has acheived some nice things, pushed the idea of pervasive corba (even if they haven't acheived it themselves). However, out of all the free software projects I've come accross, GNOME is perhaps the most offensive with its attitude of 1.0 expiriments, instability, flamage and FUD.
GNOME should not be considered an idicator of a modern linux system, but rather an exception to the way things are and should be done.
Distribution != system.
Give me any package from any distribution and I'll get it working on any other distribution in an hour or so, tops.
Furthermore, distributions package the code. They don't write the code. The code is the same accross all distribs, they just are package a bit different.
Complete code forking (aka when two branches form, they never converge again, and they become increasingly incompatable) is not an issue of the BSD license. It is a result of the BSD culture. Mind you, I consider there to be far worse things in the world than forking.
Furthermore, there has been only one fork in linux systems in recent memory, the libc5 vs. glibc debate, and that will end at some point (hopefully soon). The kernel has not forked. I run a system that only marginally resembles it's original distribution (a ditrib named Nomad written by a couple of guys who lived in the same dorm as me). A lot of the system has gotten changed sine the Nomad install. You know what though, I can still run damn near every package out of Slackware, Redhat, SuSE, ect. That's not forking in my opinion.
Well to each their own opinion. I like to be able to do whatever I want with my own code, and I prefer not to restict what others do with my code.
If you find the need to put additional restrictions on your code as a crutch because you think free software cannot compete, that's your decission.
So if you don't like it come up with your own certification mark. Anonymous coward certified if you will.
The OSI mark is only as strong as people's trust in the OSI. If you don't like ESR and/or other OSI people, then don't worry about their seal of approval. If people respect their opinions, then the mark means something.
Only if you release it under a new license and find the need to call it OSI certified.
I think most open source hackers have their favorite license (usually either GPL or BSD) and release all their code under it.
This is absolutely false. IMHO, the BSD license is more free (as in speech, not beer) that the GPL. I can do anything I want with the source, including making non-free products from it. However, I cannot restrict others freedoms by not allowing them to develop and distribute their own free code.
This is an old argument though (and after watching Eric Raymond and Theo DeRaadt battle it out there is little more I can add to either camp). Both licenses are clearly open source, IMHO. Others have given me access to their code to modify, improve, and redstribute, and squabbles aside, that's what matters.