Probably the largest complaint about X (and certainly one of the most plainly obvious) is that every program looks like it was developed in a vacuum.
This is bad.
I'm not saying that every program should be cookie-cutter, but consistency is a virtue. Plus, don't forget that interfaces continue to evolve, and if you've used a standard toolkit, it is likely to evolve as well. Your program can probably take advantage of that evolution without needing to be manually updated.
IMHO, there are a few reasons to make a new toolkit:
1) There isn't a good one available (Motif probably started because of this)
2) None of those available do what you need (Probably why Qt started). Nowadays, this argument doesn't hold water; with C++ toolkits, you can extend the toolkit easily, and with open-source toolkits, you can modify the source and benefit the whole community.
3) None of those available have an acceptable license (Gtk).
There are also a few reasons NOT to make a new toolkit:
1) It would be cool / fun. Instead, concentrate on making your app cool / fun.
2) It would be easy (as easy as using an existing toolkit? I doubt it). It's much more important for it to be easy for the user.
3) You want to be famous / change the world. Sorry bub, it ain't gonna happen.
Speaking from the standpoint of both a developer and a user, my advice is: stick with standards!
Bladeenc works great for me. IMHO I think it has the best sound quality and decent speed.
On another topic, though, has anyone tried Xing? I heard that they had the fastest encoder.
-TomK
CPU Gurus, Not Language Gurus
on
Quickie Sunday
·
· Score: 1
I really thought that CPU Gurus might be an interesting read, since I am a "software guy" but have some interest in hardware.
Unfortunately, it is painfully obvious that english is not their native language. It is so bad I couldn't stand to spend more than 5 minutes struggling to decipher it.
Hope they get someone who knows english someday, then it might actually be an interesting site.
Who cares how *long* it takes to install an OS. That is completely uninteresting to me. Whether you are using Linux or NT you should never have to reinstall, so it is worth taking the extra time to select packages that you really want.
I think it is rediculous to spend so much effort making the install faster. It is much more useful to make the install smarter; i.e. by coming up with logical grouping of options and automatic dependency selection, for example.
-Tom
Mirror sites for screenshots?
on
Some KDE news
·
· Score: 1
www.kde.org seems slashdotted..
-Tom
Why not use this for Stampede Linux?
on
GCC 2.95 Released
·
· Score: 1
Would there be any issues with Stampede Linux (or RedHat, etc) building the binaries in their distro with Code Fusion?
'Course, they couldn't distribute Code Fusion itself, but if it is fully compatible with gcc-2.95, that wouldn't seem to be much of a problem. Just that binaries the user compiles themself won't be as fast.
You make the assumption that Amiga will require modifications to the Linux kernel. While it is true that the GPL would require them to release kernel changes to the community, it is not true that they would have to release all of their software, since presumeably their UI, utilities, etc will all be user-land. No GPL restrictions there (unless they did something stupid like link against GPL'd libraries).
Many people (yourself included) point to Emacs as an example of this, but I disagree with that. Emacs really doesn't do all that much, by itself - rather, it is extensible. By coding elisp packages, Emacs gains the features that a user needs. I believe this can be one of the best ways to combat bloat, if done correctly. For example, I use Emacs to edit code and interact with my version control system, but I don't use it to browse the web, play games, or do any of the other things that it -could- do. So, I just don't load those elisp packages. This keeps Emacs relatively small and fast on my system.
The "One program does it all" philosophy -does- cause feature bloat when all functionality is compiled into the program rather than implemented as extensions to it. (i.e. MS Word, Excel, etc)
(5) The need to support all hardware under the sun
This is what the OS is for.. If your program isn't OS-level (i.e. a driver, etc) then it shouldn't support any hardware, specifically. If it needs to, then there is something wrong with the underlying OS, IMHO. Having the OS support hardware with a generic interface to applications reduces bloat. (Rant: this is why I think GGI, for example, is actually bloat-reducing, not bloat-inducing.)
BTW, drivers for specific hardware should also be modules that are loaded only if you have that hardware (ala the Linux Kernel).
Anyway, just my $2. (inflation, ya know.. 2 cents isn't what it used to be)
Don't forget about Citrix's Winframe / Metaframe products, that allow you to use a somewhat X-like client/server model with MS Windows apps. (Although the definitions of client/server are a bit different than what is generally used in the X world, the concept is the same.)
IANAA (I am not an aviator), so I don't know whether this thing will fly or not. But, for it to be practical, it will need one more thing.. Completely computerized control. The real problem with the idea of flying cars is the one truth that has been proven through the ages: people are stupid.
The whole reason for the FAA regulations is because people are fallible (or stupid). If these things were controlled by computer (with sufficient backup systems) and just directed by humans, the FAA might allow "home use". But until you can just get in and say "take me to work" and have the car take care of navigation, this thing would just be too dangerous in the hands of idiots.
I rarely see Linux uptime less than a month. Where are you looking? Perhaps the system you are looking at is running experimental software. If it is, then you have no reason to complain.
The only things that I have ever seen crash Linux are X and GGI. GGI is experimental - play with fire and you get burned. And, last time I checked, the other OS's you list use X too (with the possible exception of OpenVMS, I've never used it so I can't comment there).
In some cases, I agree (emacs, gnome). In 95% of cases, I disagree (bash, sed, grep, gcc,...). There is a difference in my mind between "value added" and "bloat". I believe that you get more functionality with GNU utilities than with their *BSD counterparts. And when talking about such small utilities, size really doesn't matter anyway. It would be interesting to compare speed of GNU utils vs. BSD utils in a standardized environment.
Even in those cases where the utilities are bloated, I still use them due to lack of acceptable alternatives (including those from our friends in Redmond). There simply isn't any other editor that can do what emacs can - even die-hard vi fans will admit that. (What they won't admit, though, is that an editor should do those things.)
It was also pointed out in this thread that GNU utilities strive for ease-of-use. This is something that is sorely overlooked by *BSD advocates. Your OS might be the greatest thing since sliced bread but if I (as an end-user) can't be productive with it, you can shove it. In my experience, Linux's ease-of-use is only rivaled among Unixen by Solaris.
I have never seen a Linux distribution that came with full source that could be rebuilt with one command
So what? If your distribution was built correctly in the first place, why rebuild it? Again, this goes back to ease-of-use. For an academic or experimental OS, compiling every program yourself might be acceptable. For a production OS, it is not. Productivity is king, and you lose productivity when you must spend time compiling every program you wish to install.
On my Linux 2.2.x systems, it fails to properly unmount its drives.
I have never heard of this happening to anyone before. This sounds like a symptom of incorrect configuration. Do you really believe that "Linux is simply broken" in such an important area?
I also enjoy it when it simply quits responding to IP packets for a while.
Once again, I have never heard of anything like this happening. I have used Linux for 4 years and have seen nothing but excellent network performance.
I never see BSD systems broken into.
Because it isn't worth the time to crack a system that nobody uses. Linux is gaining in popularity; that means more eyes looking for vulnerabilities. It also means more hands fixing them.
And why do so many of the Linux commands ship without documentation or manpages?
Which commands would those be? Why don't you email the author of the command, and whoever produced the distribution, and ask for documentation? I'll bet in most cases, they would answer "it's already there, look in/usr/doc."
The fact of the matter is, price being considered equal, OS's are judged by the marketplace, and the marketplace is saying that Linux is better, due to a nice combination of technical excellence and mass-market appeal. *BSD advocates really seem sore about this, because they don't seem to realize the second part of the equation. Technical excellence isn't enough.
You might want to check out Arla, which is a free AFS client implementation. They also include an experimental AFS server.
About two years ago I tried the Transarc Linux AFS client. It worked but was a real pain in the a**. It was a kernel module distributed in binary form only, and always for a very old kernel (which, incidentally, didn't support all of my hardware). Perhaps this has changed.
I just don't see it. I tried M4 on Linux and it's still about half the speed of IE4 rendering freshmeat. Plus it seems really unresponsive trying to click the back/forward buttons. Also, I thought it did progressive rendering? It still seems to have to load all of freshmeat before it will display any of it.
Yes, it's faster than Netscape 4.0, but not "blazingly fast".
More "features" = more bloat = more delay. "Let's throw in a chat client, but wait, how about we start from scratch with that too?"
These people should be shot for their wanton misuse of development resources. Every day without a Mozilla release pushes me closer to switching back to NT, just for IE. (Yes, IE sucks, but its a damn sight better than Netscape.)
This might be possible if all APIs were well documented in the first place, but who says they are? Most of you who develop must realize that development and documentation proceed at different paces, with the former usually outpacing the latter by a longshot.
#2: Can't release anything without first (or simultaneously) releasing specs
..And you thought MS products were late to ship now?? BTW, what about those divisions of MS where such a thing simply isn't relevant (games division, etc)? Really, this goes right along with #1. If #1 is OK (and it might be) then there's nothing to hold back #2.
#3: Can't certify products as "MS approved"
OK, then how about "Bill Gates approved"? Or, "this is known to probably work OK with Windows"? What stops MS from coming up with some other catchy slogan? I wonder how much it really means to the average consumer that their program is "certified for windows" rather that just "made to work with windows". I'm guessing that such a restriction accomplishes nothing at all.
Overall, I applaud RMS for coming up with some good ideas for suitable punishment for MS but I think they are just that - ideas. They still need to be refined before they could seriously be considered.
"As long as he doesn't make everyone call it GNU/Windows" -- anonymous coworker. (But it shows how RMS manages to lose respect from otherwise open-minded people through his rediculous adamance)
Is it only a derivative if I cut & paste the source from a GPL program into my own? This would seem silly since I could just compile the GPL code into a library and call the functions from my proprietary code.
Therefore, it seems to be a derivative if I am linking against a library with GPL'd code. If so, what if I just take the code that I want and compile it into a binary which listens on a socket for instructions from my proprietary code, and responds? I could re-release my modified "listener" under the GPL but still have a proprietary program reaping the benefits.
Therefore, it seems that it might even be a derivative if I use the functionality of the GPL'd code, without using the code itself.. In which case, if an OS was distributed under the GPL, wouldn't a program be considered a derivative even if it uses system calls? All of a sudden, the whole system must be distributed under the GPL.
Lets make things more complicated: assume I have two libraries which perform identically from an executable's point of view (i.e. Motif & LessTif), of which one is distributed under a proprietary license and the other under the GPL. I write a proprietary program and link it to Motif. I sell it to you. You don't have Motif, so you link it to LessTif and find it works flawlessly. Is my program now a "derivative" of LessTif? Does it now fall under the GPL? Can the user distribute the program under the GPL?
One thing to note: I don't know exactly how the GPL works - it might address these concerns. But even the fact that I don't know the answers to these questions (and probably few others do either) makes the GPL a very unattractive thing to get tangled up in.
Don't forget that had he done this, he would have been paying an up-front rental cost on the discs for the first 48 hours. So he would have effectively had to rent each movie twice in order to see it once (once when buying it, once again later when he's ready to watch it).
Yes, but if the "GNU System" didn't happen to exist when Linus developed his kernel, there probably wouldn't be anything called a "Linux OS Distribution".
If you remember, there was (and continues to be) no "GNU System" (unless you are talking about the Hurd). There are "GNU tools" which work in the "Linux" operating system. When RMS says there is such a thing as a "GNU System" he is taking credit away from the people who wrote the Linux kernel (in particular, Linus).
"GNU/Linux" chafes me, but I absolutely take offense at RMS trying to remove reference to Linus' (significant!) contribution.
Every time I read an account of RMS's thoughts, my belief that he is a crackpot, is strengthened.
"The so-called Linux system", as he puts it, is not "so-called", but it actually exists. I know that is hard to comprehend for someone who failed to complete his own OS, but why must he constantly malign Linus' (and hundreds of other developers') accomplishments??
RMS claims he believes in freedom, but apparently that doesn't include freedom of speech, for Linus call an OS kernel which he designed, whatever he wants. Not to mention that the GPL is more restrictive in some respects than some commercial licenses.
God bless GNU, for without them, Linux would have taken longer to develop, but I'd wager money that it still would have been developed. Or how about the BSDs? Complete, fully functional, free operating systems without all of the GNU BS^H^Hbaggage.
At the company I work for (Citrix), we have a machine that dispenses pizza, hot ham & cheese sandwiches, french fries, grilled chicken sandwiches, or chicken nuggets..
They aren't half bad, either.. though the portions are a bit small. All of them are prepared in about 2 mins.
I have Katz's articles filtered out, now how can I filter articles about Katz's articles?
I have nothing against Katz except that his articles waste my time and screen space, and are wholly irrelevant and uninteresting.
Want a real app benchmark? You got it.
on
iMac Linux
·
· Score: 1
How is that a fair comparison? Photoshop was written for the Macintosh first, and therefore it is understandable that it takes better advantage of MacOS than Windows. Not to mention that running anything under MacOS or Windows is pretty irrelevant to the question that I had.. that was, how do they compare running the same application on the same OS? I am far more interested in how fast Netscape renders Slashdot, than in how fast Photoshop does a transform. Or, perhaps, you could show me the numbers comparing gimp on Linux/PPC vs. gimp on Linux/x86. BTW, everyone says the same thing: "Linux on my PPC screams".. but what does that mean? Linux on my PII/400 "screams" also.. I want numbers!
These machines perform just fine..?
on
iMac Linux
·
· Score: 1
I've used an iMac with MacOS. It was an exercise in frustration. Its speed left MUCH to be desired. Compared to even a Pentium 233, it was not as responsive (even though it might have outperformed a P233 on a straight CPU benchmark). Plus, MacOS is total crap, not even providing decent multitasking.
So, my question is, how does this thing stack up against the CPUs from Chipzilla (Intel) when running a common OS (Linux)? Also, how is its graphics performance compared to, say, a Matrox Millenium II? And don't forget disk access speed..
Remember that benchmarks are more than just raw CPU speed. I want to see benchmarks of real-world apps..
Speaking of which, is there a way to script a graphical benchmark in X? (i.e. launch program X, scroll around, move the window, etc)
Slashdot Mirror
Probably the largest complaint about X (and certainly one of the most plainly obvious) is that every program looks like it was developed in a vacuum.
This is bad.
I'm not saying that every program should be cookie-cutter, but consistency is a virtue. Plus, don't forget that interfaces continue to evolve, and if you've used a standard toolkit, it is likely to evolve as well. Your program can probably take advantage of that evolution without needing to be manually updated.
IMHO, there are a few reasons to make a new toolkit:
1) There isn't a good one available (Motif probably started because of this)
2) None of those available do what you need (Probably why Qt started). Nowadays, this argument doesn't hold water; with C++ toolkits, you can extend the toolkit easily, and with open-source toolkits, you can modify the source and benefit the whole community.
3) None of those available have an acceptable license (Gtk).
There are also a few reasons NOT to make a new toolkit:
1) It would be cool / fun. Instead, concentrate on making your app cool / fun.
2) It would be easy (as easy as using an existing toolkit? I doubt it). It's much more important for it to be easy for the user.
3) You want to be famous / change the world. Sorry bub, it ain't gonna happen.
Speaking from the standpoint of both a developer and a user, my advice is: stick with standards!
Over and out.
-TomK
Bladeenc works great for me. IMHO I think it has the best sound quality and decent speed.
On another topic, though, has anyone tried Xing? I heard that they had the fastest encoder.
-TomK
I really thought that CPU Gurus might be an interesting read, since I am a "software guy" but have some interest in hardware.
Unfortunately, it is painfully obvious that english is not their native language. It is so bad I couldn't stand to spend more than 5 minutes struggling to decipher it.
Hope they get someone who knows english someday, then it might actually be an interesting site.
-Tom
Who cares how *long* it takes to install an OS. That is completely uninteresting to me. Whether you are using Linux or NT you should never have to reinstall, so it is worth taking the extra time to select packages that you really want.
I think it is rediculous to spend so much effort making the install faster. It is much more useful to make the install smarter; i.e. by coming up with logical grouping of options and automatic dependency selection, for example.
-Tom
www.kde.org seems slashdotted..
-Tom
Would there be any issues with Stampede Linux (or RedHat, etc) building the binaries in their distro with Code Fusion?
'Course, they couldn't distribute Code Fusion itself, but if it is fully compatible with gcc-2.95, that wouldn't seem to be much of a problem. Just that binaries the user compiles themself won't be as fast.
-Tom
You make the assumption that Amiga will require modifications to the Linux kernel. While it is true that the GPL would require them to release kernel changes to the community, it is not true that they would have to release all of their software, since presumeably their UI, utilities, etc will all be user-land. No GPL restrictions there (unless they did something stupid like link against GPL'd libraries).
-Tom
(1) One-program-does-it-all philosophy
Many people (yourself included) point to Emacs as an example of this, but I disagree with that. Emacs really doesn't do all that much, by itself - rather, it is extensible. By coding elisp packages, Emacs gains the features that a user needs. I believe this can be one of the best ways to combat bloat, if done correctly. For example, I use Emacs to edit code and interact with my version control system, but I don't use it to browse the web, play games, or do any of the other things that it -could- do. So, I just don't load those elisp packages. This keeps Emacs relatively small and fast on my system.
The "One program does it all" philosophy -does- cause feature bloat when all functionality is compiled into the program rather than implemented as extensions to it. (i.e. MS Word, Excel, etc)
(5) The need to support all hardware under the sun
This is what the OS is for.. If your program isn't OS-level (i.e. a driver, etc) then it shouldn't support any hardware, specifically. If it needs to, then there is something wrong with the underlying OS, IMHO. Having the OS support hardware with a generic interface to applications reduces bloat. (Rant: this is why I think GGI, for example, is actually bloat-reducing, not bloat-inducing.)
BTW, drivers for specific hardware should also be modules that are loaded only if you have that hardware (ala the Linux Kernel).
Anyway, just my $2. (inflation, ya know.. 2 cents isn't what it used to be)
I don't believe a graphical install must necessarily install X. The installer itself doesn't need to dictate what is installed.
As an example, doesn't RedHat use cpio during install? I know that you can choose not to install cpio, at least on older versions of RH.
Worst case scenario: remove X when you are done installing. Simple.
BTW, it doesn't look like Caldera gives you the option, however.. not terribly configurable as far as which packages end up getting installed.
-Tom
..shameless plug..
Don't forget about Citrix's Winframe / Metaframe products, that allow you to use a somewhat X-like client/server model with MS Windows apps. (Although the definitions of client/server are a bit different than what is generally used in the X world, the concept is the same.)
-Tom
IANAA (I am not an aviator), so I don't know whether this thing will fly or not. But, for it to be practical, it will need one more thing.. Completely computerized control. The real problem with the idea of flying cars is the one truth that has been proven through the ages: people are stupid.
The whole reason for the FAA regulations is because people are fallible (or stupid). If these things were controlled by computer (with sufficient backup systems) and just directed by humans, the FAA might allow "home use". But until you can just get in and say "take me to work" and have the car take care of navigation, this thing would just be too dangerous in the hands of idiots.
-Tom
I rarely see a Linux uptime greater than a month
...). There is a difference in my mind between "value added" and "bloat". I believe that you get more functionality with GNU utilities than with their *BSD counterparts. And when talking about such small utilities, size really doesn't matter anyway. It would be interesting to compare speed of GNU utils vs. BSD utils in a standardized environment.
/usr/doc."
I rarely see Linux uptime less than a month. Where are you looking? Perhaps the system you are looking at is running experimental software. If it is, then you have no reason to complain.
The only things that I have ever seen crash Linux are X and GGI. GGI is experimental - play with fire and you get burned. And, last time I checked, the other OS's you list use X too (with the possible exception of OpenVMS, I've never used it so I can't comment there).
Linux, specifically GNU, utilities redefine bloat.
In some cases, I agree (emacs, gnome). In 95% of cases, I disagree (bash, sed, grep, gcc,
Even in those cases where the utilities are bloated, I still use them due to lack of acceptable alternatives (including those from our friends in Redmond). There simply isn't any other editor that can do what emacs can - even die-hard vi fans will admit that. (What they won't admit, though, is that an editor should do those things.)
It was also pointed out in this thread that GNU utilities strive for ease-of-use. This is something that is sorely overlooked by *BSD advocates. Your OS might be the greatest thing since sliced bread but if I (as an end-user) can't be productive with it, you can shove it. In my experience, Linux's ease-of-use is only rivaled among Unixen by Solaris.
I have never seen a Linux distribution that came with full source that could be rebuilt with one command
So what? If your distribution was built correctly in the first place, why rebuild it? Again, this goes back to ease-of-use. For an academic or experimental OS, compiling every program yourself might be acceptable. For a production OS, it is not. Productivity is king, and you lose productivity when you must spend time compiling every program you wish to install.
On my Linux 2.2.x systems, it fails to properly unmount its drives.
I have never heard of this happening to anyone before. This sounds like a symptom of incorrect configuration. Do you really believe that "Linux is simply broken" in such an important area?
I also enjoy it when it simply quits responding to IP packets for a while.
Once again, I have never heard of anything like this happening. I have used Linux for 4 years and have seen nothing but excellent network performance.
I never see BSD systems broken into.
Because it isn't worth the time to crack a system that nobody uses. Linux is gaining in popularity; that means more eyes looking for vulnerabilities. It also means more hands fixing them.
And why do so many of the Linux commands ship without documentation or manpages?
Which commands would those be? Why don't you email the author of the command, and whoever produced the distribution, and ask for documentation? I'll bet in most cases, they would answer "it's already there, look in
The fact of the matter is, price being considered equal, OS's are judged by the marketplace, and the marketplace is saying that Linux is better, due to a nice combination of technical excellence and mass-market appeal. *BSD advocates really seem sore about this, because they don't seem to realize the second part of the equation. Technical excellence isn't enough.
-Tom
You might want to check out Arla, which is a free AFS client implementation. They also include an experimental AFS server.
About two years ago I tried the Transarc Linux AFS client. It worked but was a real pain in the a**. It was a kernel module distributed in binary form only, and always for a very old kernel (which, incidentally, didn't support all of my hardware). Perhaps this has changed.
-Tom
2. The rendering engine is blazing fast.
I just don't see it. I tried M4 on Linux and it's still about half the speed of IE4 rendering freshmeat. Plus it seems really unresponsive trying to click the back/forward buttons. Also, I thought it did progressive rendering? It still seems to have to load all of freshmeat before it will display any of it.
Yes, it's faster than Netscape 4.0, but not "blazingly fast".
More "features" = more bloat = more delay. "Let's throw in a chat client, but wait, how about we start from scratch with that too?"
These people should be shot for their wanton misuse of development resources. Every day without a Mozilla release pushes me closer to switching back to NT, just for IE. (Yes, IE sucks, but its a damn sight better than Netscape.)
Sigh... where's opera for linux?
-Tom
"Help me Opera, you're my only hope"
They probably include the glibc 2.0.6 runtime library with the system, and then set LD_LIBRARY_PATH before running SO.
Just a guess, though, I haven't seen the distro.
-Tom
#1: release APIs
This might be possible if all APIs were well documented in the first place, but who says they are? Most of you who develop must realize that development and documentation proceed at different paces, with the former usually outpacing the latter by a longshot.
#2: Can't release anything without first (or simultaneously) releasing specs
..And you thought MS products were late to ship now?? BTW, what about those divisions of MS where such a thing simply isn't relevant (games division, etc)? Really, this goes right along with #1. If #1 is OK (and it might be) then there's nothing to hold back #2.
#3: Can't certify products as "MS approved"
OK, then how about "Bill Gates approved"? Or, "this is known to probably work OK with Windows"? What stops MS from coming up with some other catchy slogan? I wonder how much it really means to the average consumer that their program is "certified for windows" rather that just "made to work with windows". I'm guessing that such a restriction accomplishes nothing at all.
Overall, I applaud RMS for coming up with some good ideas for suitable punishment for MS but I think they are just that - ideas. They still need to be refined before they could seriously be considered.
"As long as he doesn't make everyone call it GNU/Windows" -- anonymous coworker. (But it shows how RMS manages to lose respect from otherwise open-minded people through his rediculous adamance)
My question is, what constitutes a "derivative"?
Is it only a derivative if I cut & paste the source from a GPL program into my own? This would seem silly since I could just compile the GPL code into a library and call the functions from my proprietary code.
Therefore, it seems to be a derivative if I am linking against a library with GPL'd code. If so, what if I just take the code that I want and compile it into a binary which listens on a socket for instructions from my proprietary code, and responds? I could re-release my modified "listener" under the GPL but still have a proprietary program reaping the benefits.
Therefore, it seems that it might even be a derivative if I use the functionality of the GPL'd code, without using the code itself.. In which case, if an OS was distributed under the GPL, wouldn't a program be considered a derivative even if it uses system calls? All of a sudden, the whole system must be distributed under the GPL.
Lets make things more complicated: assume I have two libraries which perform identically from an executable's point of view (i.e. Motif & LessTif), of which one is distributed under a proprietary license and the other under the GPL. I write a proprietary program and link it to Motif. I sell it to you. You don't have Motif, so you link it to LessTif and find it works flawlessly. Is my program now a "derivative" of LessTif? Does it now fall under the GPL? Can the user distribute the program under the GPL?
One thing to note: I don't know exactly how the GPL works - it might address these concerns. But even the fact that I don't know the answers to these questions (and probably few others do either) makes the GPL a very unattractive thing to get tangled up in.
Don't forget that had he done this, he would have been paying an up-front rental cost on the discs for the first 48 hours. So he would have effectively had to rent each movie twice in order to see it once (once when buying it, once again later when he's ready to watch it).
Yes, but if the "GNU System" didn't happen to exist when Linus developed his kernel, there probably wouldn't be anything called a "Linux OS Distribution".
If you remember, there was (and continues to be) no "GNU System" (unless you are talking about the Hurd). There are "GNU tools" which work in the "Linux" operating system. When RMS says there is such a thing as a "GNU System" he is taking credit away from the people who wrote the Linux kernel (in particular, Linus).
"GNU/Linux" chafes me, but I absolutely take offense at RMS trying to remove reference to Linus' (significant!) contribution.
Every time I read an account of RMS's thoughts, my belief that he is a crackpot, is strengthened.
"The so-called Linux system", as he puts it, is not "so-called", but it actually exists. I know that is hard to comprehend for someone who failed to complete his own OS, but why must he constantly malign Linus' (and hundreds of other developers') accomplishments??
RMS claims he believes in freedom, but apparently that doesn't include freedom of speech, for Linus call an OS kernel which he designed, whatever he wants. Not to mention that the GPL is more restrictive in some respects than some commercial licenses.
God bless GNU, for without them, Linux would have taken longer to develop, but I'd wager money that it still would have been developed. Or how about the BSDs? Complete, fully functional, free operating systems without all of the GNU BS^H^Hbaggage.
In short: RMS, grow up.
At the company I work for (Citrix), we have a machine that dispenses pizza, hot ham & cheese sandwiches, french fries, grilled chicken sandwiches, or chicken nuggets..
They aren't half bad, either.. though the portions are a bit small. All of them are prepared in about 2 mins.
I have Katz's articles filtered out, now how can I filter articles about Katz's articles?
I have nothing against Katz except that his articles waste my time and screen space, and are wholly irrelevant and uninteresting.
How is that a fair comparison? Photoshop was written for the Macintosh first, and therefore it is understandable that it takes better advantage of MacOS than Windows. Not to mention that running anything under MacOS or Windows is pretty irrelevant to the question that I had.. that was, how do they compare running the same application on the same OS? I am far more interested in how fast Netscape renders Slashdot, than in how fast Photoshop does a transform. Or, perhaps, you could show me the numbers comparing gimp on Linux/PPC vs. gimp on Linux/x86. BTW, everyone says the same thing: "Linux on my PPC screams".. but what does that mean? Linux on my PII/400 "screams" also.. I want numbers!
I've used an iMac with MacOS. It was an exercise in frustration. Its speed left MUCH to be desired. Compared to even a Pentium 233, it was not as responsive (even though it might have outperformed a P233 on a straight CPU benchmark). Plus, MacOS is total crap, not even providing decent multitasking.
So, my question is, how does this thing stack up against the CPUs from Chipzilla (Intel) when running a common OS (Linux)? Also, how is its graphics performance compared to, say, a Matrox Millenium II? And don't forget disk access speed..
Remember that benchmarks are more than just raw CPU speed. I want to see benchmarks of real-world apps..
Speaking of which, is there a way to script a graphical benchmark in X? (i.e. launch program X, scroll around, move the window, etc)