As a person in the research business, I'm not so sure that all reasearch and knowledge is good. Defining "research" is hard, and some people have "crossed the line" -- think of Nazi science research on people in prison camps. Clearly that was not good.
As for knowledge, is it ethical to use knowledge gained from those prison camp experiments? Almost everyone says no. This suggests that the statement "all knowledge is good" is not true; my off-the-cuff opinion is that it is an oversimplication with important exceptions.
Maybe we could say that Wisdom is good, and with wisdom, knowledge is good. But that depends on defining wisdom, and we'd probably end up with a tautology when we did that (wisdom == whatever it takes to make knowledge good). These aren't just semantic problems. There are records which suggest our species (I'm making an assumption here;-) has tried to nail down ideas like knowledge and truth for several millenia. It seems likely "we" have struggled with it for longer than we have records. This isn't simple stuff, and it's not taught in science, math, or computer courses.
-Paul Komarek
Re:Information about cases from a noise angle...
on
Choosing a Good Case
·
· Score: 2
Thanks for the Plycon link. When I was looking for Papst fans in the US two years ago, I gave up. You could find the noiser high flow Papst fans from Newark, but you couldn't find the quiet ones on this side of the Atlantic.
I wanted to add something to you your planning list: the power supply. The case you suggest is a good, high quality case, and if a PSU is included it is probably pretty good. In most cases, pun intended and necessary, to really keep the noise down one usually has to replace the stock PSU. For instance, Enermax makes makes some very nice dual-fan power supplies where the backboard fan can run slowly and be quiet, while the inside fan is faster and louder but not particularly audible outside the case. Be prepared to spend at least $50 on a good power supply.
In the last couple years I've tried the Chenbro Genie, Fong-Kai FK320, and the Chenbro Junior. The Fong-Kai came with the worst power supply, and the Antec PSU I ordered for the Genie became loud in a bit less than a year. The Genie and the Junior could probably run with only 120mm fans (except for those in the PSU, of course -- and the Genie can take a 92mm for the drive cage, if desired). The FK-320 needs a 92mm fan in back, which is the most important case fan. Overall the FK-320 is the noisest box. Both of these cases have sides that "slide" off forward or backwards, and both hang up terribly and are a real pain.
The Junior with an Enermax EG365-P-FC, a 120mm NMB on the back (could replace it with a Panasonic 120mm 'L1A or whatever -- I had this one on hand from Mouser Electronics) and also on the 80mm Alpha PAL8045 heatsink (with copper shim and Arctic Silver 3!;-) is my quietest machine by far. I am extremely pleased with it. I should mention that this machine is an Athlon XP 1900+ with a 10Krpm SCSI drive and ATI Radeon DDR 32 AIW, which I use for gaming and numerics (I'm a grad student). So far I've not come anywhere near heat problems to the best of my knowledge, even has Pittsburgh has (occassionally) skipped Spring and jumped to summer. If I run into trouble, I can always add a front fan. The sides lift off, starting with a handle-driven cam, and work perfectly.
The links I provided are to my favorite case and heat sink/fan vendors, respectively. Directron will let you choose which PSU you want for the case, has a terrific stock of cases, and has some cool case badges (includinge Tux!). 1CoolPC has a small, hand-picked selection heat sinks, fans, and related items; typically responds to email within an hour (including questions about heat sink performance and fan noise); and will refund your money *and* shipping if you don't like your purchase. Note that 1CoolPC's website is a little slow sometimes. To buy a similar setup to what I described above (excluding the CPU, disk, and video), from these vendors, will currently set you back about $240 including the Tux case badge, Arctic Silver, and shipping. That's about the same price as a fast CPU, or a normal CPU plus the motherboard. I believe a quiet, cool system is worth this amount of money.
It was the noise that pursueded me to scrap my lego-controlled webcam. The webcam wasn't a LEGO part, and was effectively a smooth sphere with a couple square bits on it (3com Big Picture NTSC). Between the harness and the focus mechanism (thank heavens of the current generation of squishy rubber LEGO tires), the device became pretty heavy. Add to that a third motor that is at least 10 years old and only turns at high speed, and the sound was terrible.
I did use my limit sensors to control some things, and dead-reckoning to keep the user from completely unscrewing the lense during focussing (and later bought rotation sensors to help that sort of thing).
I agree that LEGOs are expensive, in that you have to expend valuable money to acquire sufficient quantities of this valuable toy. But I have never seen a better option for physical prototyping. If you've ever
*) looked at the price of a CNC milling machine, or *) investigated the effort and money needed for working with plastics, or *) sought in vain for a house in Pittsburgh with a room to use as a wood shop (or fretted about using power tools at 9pm because you lived in an apartment) *) discovered just how expensive a linear bearing really is
then you know why LEGOs are worthwhile.
Sure you might have to work your idea around the blocks that the LEGO corporation provides (often using -Paul Komarek
No, it wasn't. And it should be clear that it wasn't. Unless you equate the Commander in Chief with the armed forces. There's a long tradition of making fun of presidents and Congress for the bizarre, often inexplicable, and all-to-often stupid things they do (or don't do). For instance, I might choose to call G. W. Bush the Bozo in Chief, without intending any criticism of the Pentagon.
There is no popular "approval ratings" metric for the armed forces. There is, on the other hand, an exteremely popular "approval ratings" metric of the President of the US. I think it is quite clear that the comment was directed at then-president G.H.W. Bush as if he were trying to improve his approval rating. Just like people dubbed the cruise missle attack then-president B. Clinton against (presumably) Osama Bin Laden the "Monica Missles", as if Clinton were trying to distract the public and the press from the never-ending Monica Lewinsky trial brought by Ken Starr.
I think Americans are still pretty proud of the US Armed Services. I think Americans haven't been proud of our government for many, many years.
It seems there are many comments of the form "now I have a reason to buy a GameCube". It will be interesting to see if Nintendo's sales really defy the normal trends, and surge instead of levelling off.
I hate to be a member of the XBox team. Imagine the XBox and the GameCube drag racing down the freeway, the XBox pretty impressed by its own top speed, when suddenly the Cube rolls down his window and says "Hey, buddy, how do get this thing out of second gear?" The Cube then laughs, peels out going from fourth to fifth, and smokes the poor XBox.
You should go outside in the Summer. How else will you get vitamin D (well, four cereal-bowls worth of milk will do it, too)? Save the video games for the Winter, and by Calcium Citrate vitamins.
I'm willing to wait for Nintendo. Just like I'm waiting for DVD region encoding to (legally) die before I by a DVD player.
Your previous post questioned if compiler names (or somesuch) should preface the names of the binaries they produce. I think everyone agrees that this is a dumb idea. Including Stallman. I write this in response to the first question you ask in a post attached to mine: "So, since IBM used to (and for all I know, still does) compile different parts of OS/2 with different compilers, they should have prepended the names of all the compilers to its name?" I simply wrote that building an OS around the Linux kernel basically requires building a GNU-based system.
I agree that the RMS quotes you provide are attacks on Torvald's commitment to Free software. What I refuted was your previous comment "He bitches and moans that Linus doesn't have enough ideological purity." The quotes you provide don't support your previous statement. RMS is clearly *not* bitching an moaning that Torvalds has too little ideological purity (an absolute statement). Instead, RMS complains that Torvalds message says "...'Non-free software is ok; I use it and develop it myself.'". That is, Torvalds' comments are contrary to the stated position of the GNU project (about which RMS does *not* complain); and because of his influence when Torvalds states these views, the GNU project's goals suffer.
As we both admit, we don't know for sure what would have happened without the linux kernel (me: FreeBSD wouldn't be much larger than it is today, because of its ISP-centric focus; you: FreeBSD would have grabbed most of the current GNU/Linux-using population, if I understand you correctly). However, I doubt RMS would be much less influential. He's been well-known for, geez, 30 years now? I don't believe his fame has increased tremendously in the last 5 years. In large part because most people still aren't paying attention to the contributions of the GNU project. On the other hand, RMS' infamy has probably grown in the last 5 years.;-)
But those who know RMS' history wouldn't be any less impressed with him had Torvalds not created Linux (the kernel). RMS matched an *entire* company's software output to guarantee a Free lisp machine implementation. He created gcc, gdb, emacs, and the GPL; the first two he made *by himself* (I'm not sure how much of the external contributions to TECO and such made it into the first GPL'd emacs). RMS received the McArthur Genius awared, and started the Free Software Foundation. And he still finds time to respond to his email these days.
I admit that being impressive and being famous is not the same thing. But I don't think RMS' fame has gained much from Torvalds' work. How many GNU/Linux-related articles in Business Week even mention RMS? I believe that RMS has earned his fame mostly by himself. And he's earned at least some of his infamy, too.;-)
Geesh, can nobody read carefully? This is the second time I'm responding to this issue.
I never said that the Linux kernel should be called GNU/Linux. As far as that goes, neither did Richard Stallman. I didn't even infer that. No careful reader could, in good faith, conclude from my comment that I felt that GNU should be prepended to things compiled with gcc.
Re-reading my previous comment (the grandparent of this one), I believe it is clear that I am arguing that
1) The kernel is not necessarily more important than the tools around it, as the great-grandparent comment suggested, and
2) Using the Linux kernel without the GNU tools is a waste of one's time, and may not be possible without modifications, and isn't worthwhile in any case.
I'm not sure we'd all be using *BSD. The free BSD projects have a different development model than Linux and the GNU project, with different priorities and structure. There is only *one* distrobution built around the FreeBSD kernel, for instance. The FreeBSD project is very centralized compared to the Linux and GNU models. The FreeBSD kernel doesn't yet support USB->serial adapters, despite Keyspan offering to help -- I assume this is because of lack of interest considering that they have very good support for other devices. It seems that the FreeBSD folks think that the (rather conservative, IMO) UFS filesystem is superior to ext2. The Linux folks seem to prefer ext2, worts, speed, and all.
Summing up the previous paragraph, it isn't at all clear to me that FreeBSD (or one of the others) would have emerged to fill the niche that GNU/Linux systems have. It seems likely that most of us would not have jumped on board the unix-on-my-PC train until something like GNU/Linux systems came along. Therefore, it isn't clear to me that we'd all be using *BSD.
"If Linus hadn't written a kernel, we'd all be running BSD, not the Hurd, because BSD would have been usable first." Heck, I remember installing 386BSD (or was that BSD386?) at the same time I tried SLS (an old GNU/Linux distro) in early 1993 (I didn't end up using either, because I didn't know what the heck I was doing and didn't commit to GNU/Linux until 1995 or 1996 when it became "easy" for "everyone" to join in =-).
Given the goals of RMS, it's not clear that he has any reason to appreciate Linux (the kernel). As an old, bad expression goes, to him the "Linux journey" has probably felt like watching your mother-in-law go over a cliff in your new Cadillac. As I argued above, it's not clear to me that one of the BSD forks would have had much effect on the eventual popularity of decentralized GNU-based systems. And the GPL is very popular in the BSD world. The *BSD folks use emacs and gcc, too. But their (non-GNU) filesystem utilities have screwed-up options, though (just kidding, just kidding).
RMS didn't bitch that Linus has too little ideological purity. He bitched about Linus' views and influence causing problems for the GNU project. I'm not completely sure there were no ad-hominem attacks as you suggest there were, and I'm not going to go back and double-check. I'll leave that as an exercise for some careful reader who managed to read my previous post correctly.
"I see... So, every program compiled on unix with fsf tools should be GNU/whatever?"
I didn't say that. Looking back at my post carefully, I see that I was responding to your assertion "it is possible to have a linux kernel running without any gnu software."
I was arguing that it is nigh on impossible to use the Linux the kernel without needing GNU tools at some point.
As far as GNU/ls goes, I'm not sure anyone would want their name attached to a command with that many options.;-)
Your argument that many people use the GNU tools, but only one group uses the kernel name Linux, suggest that the GNU tools are more important?
Your "theory" about using the Linux kernel without any GNU tools seems silly. First, you'll need to get a binary of the kernel without compilation using gcc or gmake -- note that Linux is distrbuted as source, and is very sensitive to which *version* of gcc compiles it. You can find binary version of the Linux kernel and its required user-space utilities, but did these reequire GNU tools to create? Probably. Furthermore, the user-space stuff will probably require *at least* the GNU project's glibc to run.
Once you have a non-GNU compiled kernel, you'll need a boot loader, or just dd it to the boot sector of a floppy using non-GNU dd, and use some non-GNU version of rdev that doesn't require something provided by the GNU project (like glibc).
And once your kernel boots, you're done! Now if only you had a shell, or init that doesn't depend on something from the GNU project.
Once you've done all that, you'll have a pathetic non-GNU system wrapped around Linux? In the process, I'll expect you'll have spent a lot of money, time, and sanity; and probably rewritten major portions of Linux and the kernel build system, not to mention the C library. Whee!
"it is possible to have a linux kernel running without any gnu software."
I'd like to see one. While you don't care what compiled it, it still needs to be compiled. Or did you forget that kernel.org distributes *source code*. And you need more than a compiler. You need gmake (almost certainly you need GNU's make), and you need some version of sed, and probably awk. It seems likely you'll need a shell, too. You can buy some of those, but who will replace gmake? Or has one of the propietary Unix vendors cloned all the features of gmake by now?
Oh, by the way, the kernel source comes as a compressed tarball. You'll need to decompress and untar it. You don't need gnu for this, but please point out which tools you'll use instead and how much they cost -- including your Solaris or Tru64 or HP-UX (or whatever) single-user license, please.
It also needs module tools (usually), and all kinds of other user-space utilities. If just one of these utilities is part of the GNU project(say, for instance, GRUB and LILO, though could always write the binary kernel you downloaded onto your MBR -- that might work), your kernel is going to be a real pain in the butt. Oh, and you'll need filesystems and filesystem tools, and a shell to log into, and a login program, and an init program, and you might want manual pages, too. Good luck finding all of this stuff without touching the GNU project.
Hell, even the *BSDs use a fair amount of GNU Project code. They don't use linux, though. Maybe the kernel *is* more easily replaced than the user-space utilities its programmers and users rely on?
Geesh, will people quit equating lines-of-code with what is most important? This nonsense was all over the place when the RMS originally made his GNU/Linux request. Thankfully the parent post doesn't dwell on the size of XFree86.
In particular, comparing XFree86 to the GNU project code is just silly. While XFree86 is *convenient*, it is certainly not central to the goals of the GNU project or the kernel we call linux.
The essential things to the GNU Project are things like bash, lilo, sed, awk, gcc, gdb, gmake, and emacs. The GNU project also needs a kernel, and right now they need the Free parts of the linux kernel to be complete. Linux needs things like gcc, gdb, gmake, lilo, bash, perl, sed, maybe awk (haven't seen it, but hard to imagine it's not in some script somewhere). Thus the GNU Project and the Linux (kernel) project are working *together*.
Furthermore, the GNU Project is a lot closer to finishing their kernel project, the HURD, than the Linux project is to finishing their own development toolchain. If one was to decide who could survive without the other, I expect only the GNU Project would come out standing.
Just how will you compile your Linux kernels without gcc and gmake and bash and sed? Using MSVC? Borland? The Portland group compilers? Compaq's cc in Tru64? I suppose the x86 folks could use one of Intel's compilers, but that doesn't solve gmake, bash, and sed dependencies. On the other hand, the HURD is (finally) working, even if only barely. If Torvald's decided to eschew the GNU Project entirely, I'd bet Alan Cox, among many others, would not stick around -- they'd either fork Linux, or work on the HURD.
I don't think RMS has ever accused Torvalds of hogging the "limelight". RMS isn't in this for ego or popularity. It seems clear from his comments that he wants the GNU name to be present so that people will know the GNU project exists, and hopefully go on to learn about its goals. RMS has more fame and credit than he probably cares for (consider just the LISP machine Battle Royale).
I don't think Stallman is content to believe that the "war" is already over. That's why he is still fighting. If he were a quitter, the GNU project would never have emerged. I wouldn't be surprised if people told him he had "already lost" when the lisp or emacs battles started.
2) If answer to 1) is yes, then are they using the parts they find illegal? Almost certainly not. Linux, the kernel, has many parts licensed by different people. Choosing only the parts with Free license, and among those, only the ones who comply with their license, seems entirely reasonable to me. In fact, you could call the resulting Free kernel a derivative work of the parts with a Free license, if you wanted. No need to throw the baby out with the bathwater.
In my opinion, your argument is valid but without premise. You would need to provided evidence that the FSF was using non-Free code. Once you've named a non-free part of the linux kernel, you'll need to show that the FSF is using that code.
Just because they downloaded a kernel from kernel.org with non-Free parts, doesn't mean they're using those parts. Especially since those parts are likely to be completely unnecessary device drivers. For instance, my desktop and gaming linux system shows the following at the top of lsmod's output (reformatted to fit the narrow text input box;-):
"Module Size Used by Not tainted"
The "not tainted" part is important to me. If I'm not misinterpreting it, it means that every module loaded (and I think also that everything compiled in) has a license tag which indicates a Free license. In particular, modules with no stated license will taint the kernel, creating false negatives. "not tainted" means "not tainted", every time. I guess I should really check whether "not tainted" necessarily means only Free software is included.
Given the area that they're covering, I doubt it. I would assume that porn in public spaces is not as popular as porn in your home. This wifi network appears to cover some popular public parks, and a *lot* of office buildings. It seems unlikely that the clerks or shoppers at Kaufman's are going be pirating music while riding the old wooden escalator. Especially given the really low bitrate an individual is likely to receive.
Allowing wifi on the trolley sounds like a great idea, too (at least for me;-). When I ride the trolly, there's at least another 15 people on board. Again, it's a public space. Same goes for the cafes that my ISP, Telerama, is covering. We're mostly talking about students who are working on research/homework, or addicted to the web (and coffee).
From what I remember, the Israeli pilots were prohibited from excessive running -- which definitely seems to disagree with what you are saying. If you are part of the Israeli Air Defense Force (I think that is what they call it), then I'm clearly in the wrong (and either my memory is screwed up, or the program was). I don't claim to be an expert; I just wanted to share some piece of (possibly wrong) trivia I ran into.
I've practiced the "grunting" thing before, just for fun. It makes your face turn red in a hurry!
It appears I should have been more clear. I thought the guy could not have been seriously objecting to this law. I've never lived in a state that didn't have this law (well, I've only lived in Washington and Pennsylvania). I think it's a damn good law.
When one is driving in a school zone, behind a bus, or in a residential neighborhood, one should be watching for kids. However, there are a lot of reckless, impatient drivers out there. Heck, I've even seen one driver ignore school crossing guards and go blasting through a crosswalk while kids were crossing. I think it's clear that such a law is needed in order to curb the many idiots behind the law. For those of us with brains and a heart, we need to follow the law, too.
In some places where drivers are particularly reckless, I've seen the schoolbus stop the bus across both lanes to stop people who ignore the fine. Fine by me. The schoolbus is in charge, as far as I'm concerned.
There are many factors which contribute to one's ability to sustain G-forces. While it is tempting to think that a very fit person could sustain more head-to-toe Gs, that is not necessarily true. If it is easy for the blood to flow out of the head (for instance, in an athlete whose exercise has promoted high levels of capilary and/or blood vessel growth -- note, I'm definitely not a doctor), then one can't sustain as many Gs. If the distance between the head and the neck is short, it makes it easier for the heart to achieve the pressure necessary to provide blood to the brain. I'm sure there are many other bizarre contributing factors.
As I mentioned, I'm talking out my rear-end to some degree. My source of information comes from some show I saw several years ago about the selection and training of fighter pilots in Israel. IIRC, they measured the heart/brain distance (or at least neck length) of candidates, and limited their pilots' cardiovascular exercise. Who would have guessed?
I don't even think there is a debate of which console is best for young kids. Get whatever Nintendo makes.
For adults with social skills that like to play games in the same room with friends, there's not much debate either. The gamecube still wins.
The other consoles seem to be targetted at the same audience as most PC video games. I think the description is something like "males between 15 and 35".
I think that Nintendo opened at $200 and MS at $300. It's been so long since the PS/2 came out that I have no idea what it opened at, and by now you'd have to take inflation into account to make a valid comparison.;-)
Early adopters always get screwed on price. It's their choice. The same is true in the PC market. Consider, for instance, the IDE hard drive market and the graphics card market.
Slashdot Mirror
"All research is good, all knowledge is good"
;-) has tried to nail down ideas like knowledge and truth for several millenia. It seems likely "we" have struggled with it for longer than we have records. This isn't simple stuff, and it's not taught in science, math, or computer courses.
As a person in the research business, I'm not so sure that all reasearch and knowledge is good. Defining "research" is hard, and some people have "crossed the line" -- think of Nazi science research on people in prison camps. Clearly that was not good.
As for knowledge, is it ethical to use knowledge gained from those prison camp experiments? Almost everyone says no. This suggests that the statement "all knowledge is good" is not true; my off-the-cuff opinion is that it is an oversimplication with important exceptions.
Maybe we could say that Wisdom is good, and with wisdom, knowledge is good. But that depends on defining wisdom, and we'd probably end up with a tautology when we did that (wisdom == whatever it takes to make knowledge good). These aren't just semantic problems. There are records which suggest our species (I'm making an assumption here
-Paul Komarek
Thanks for the Plycon link. When I was looking for Papst fans in the US two years ago, I gave up. You could find the noiser high flow Papst fans from Newark, but you couldn't find the quiet ones on this side of the Atlantic.
;-) is my quietest machine by far. I am extremely pleased with it. I should mention that this machine is an Athlon XP 1900+ with a 10Krpm SCSI drive and ATI Radeon DDR 32 AIW, which I use for gaming and numerics (I'm a grad student). So far I've not come anywhere near heat problems to the best of my knowledge, even has Pittsburgh has (occassionally) skipped Spring and jumped to summer. If I run into trouble, I can always add a front fan. The sides lift off, starting with a handle-driven cam, and work perfectly.
I wanted to add something to you your planning list: the power supply. The case you suggest is a good, high quality case, and if a PSU is included it is probably pretty good. In most cases, pun intended and necessary, to really keep the noise down one usually has to replace the stock PSU. For instance, Enermax makes makes some very nice dual-fan power supplies where the backboard fan can run slowly and be quiet, while the inside fan is faster and louder but not particularly audible outside the case. Be prepared to spend at least $50 on a good power supply.
In the last couple years I've tried the Chenbro Genie, Fong-Kai FK320, and the Chenbro Junior. The Fong-Kai came with the worst power supply, and the Antec PSU I ordered for the Genie became loud in a bit less than a year. The Genie and the Junior could probably run with only 120mm fans (except for those in the PSU, of course -- and the Genie can take a 92mm for the drive cage, if desired). The FK-320 needs a 92mm fan in back, which is the most important case fan. Overall the FK-320 is the noisest box. Both of these cases have sides that "slide" off forward or backwards, and both hang up terribly and are a real pain.
The Junior with an Enermax EG365-P-FC, a 120mm NMB on the back (could replace it with a Panasonic 120mm 'L1A or whatever -- I had this one on hand from Mouser Electronics) and also on the 80mm Alpha PAL8045 heatsink (with copper shim and Arctic Silver 3!
The links I provided are to my favorite case and heat sink/fan vendors, respectively. Directron will let you choose which PSU you want for the case, has a terrific stock of cases, and has some cool case badges (includinge Tux!). 1CoolPC has a small, hand-picked selection heat sinks, fans, and related items; typically responds to email within an hour (including questions about heat sink performance and fan noise); and will refund your money *and* shipping if you don't like your purchase. Note that 1CoolPC's website is a little slow sometimes. To buy a similar setup to what I described above (excluding the CPU, disk, and video), from these vendors, will currently set you back about $240 including the Tux case badge, Arctic Silver, and shipping. That's about the same price as a fast CPU, or a normal CPU plus the motherboard. I believe a quiet, cool system is worth this amount of money.
-Paul Komarek
Are you in the military, or a veteren?
-Paul Komarek
It was the noise that pursueded me to scrap my lego-controlled webcam. The webcam wasn't a LEGO part, and was effectively a smooth sphere with a couple square bits on it (3com Big Picture NTSC). Between the harness and the focus mechanism (thank heavens of the current generation of squishy rubber LEGO tires), the device became pretty heavy. Add to that a third motor that is at least 10 years old and only turns at high speed, and the sound was terrible.
I did use my limit sensors to control some things, and dead-reckoning to keep the user from completely unscrewing the lense during focussing (and later bought rotation sensors to help that sort of thing).
-Paul Komarek
I agree that LEGOs are expensive, in that you have to expend valuable money to acquire sufficient quantities of this valuable toy. But I have never seen a better option for physical prototyping. If you've ever
*) looked at the price of a CNC milling machine, or
*) investigated the effort and money needed for working with plastics, or
*) sought in vain for a house in Pittsburgh with a room to use as a wood shop (or fretted about using power tools at 9pm because you lived in an apartment)
*) discovered just how expensive a linear bearing really is
then you know why LEGOs are worthwhile.
Sure you might have to work your idea around the blocks that the LEGO corporation provides (often using
-Paul Komarek
No, it wasn't. And it should be clear that it wasn't. Unless you equate the Commander in Chief with the armed forces. There's a long tradition of making fun of presidents and Congress for the bizarre, often inexplicable, and all-to-often stupid things they do (or don't do). For instance, I might choose to call G. W. Bush the Bozo in Chief, without intending any criticism of the Pentagon.
There is no popular "approval ratings" metric for the armed forces. There is, on the other hand, an exteremely popular "approval ratings" metric of the President of the US. I think it is quite clear that the comment was directed at then-president G.H.W. Bush as if he were trying to improve his approval rating. Just like people dubbed the cruise missle attack then-president B. Clinton against (presumably) Osama Bin Laden the "Monica Missles", as if Clinton were trying to distract the public and the press from the never-ending Monica Lewinsky trial brought by Ken Starr.
I think Americans are still pretty proud of the US Armed Services. I think Americans haven't been proud of our government for many, many years.
-Paul Komarek
It seems there are many comments of the form "now I have a reason to buy a GameCube". It will be interesting to see if Nintendo's sales really defy the normal trends, and surge instead of levelling off.
I hate to be a member of the XBox team. Imagine the XBox and the GameCube drag racing down the freeway, the XBox pretty impressed by its own top speed, when suddenly the Cube rolls down his window and says "Hey, buddy, how do get this thing out of second gear?" The Cube then laughs, peels out going from fourth to fifth, and smokes the poor XBox.
-Paul Komarek
Peach kicks both their asses!
-Paul Komarek
You should go outside in the Summer. How else will you get vitamin D (well, four cereal-bowls worth of milk will do it, too)? Save the video games for the Winter, and by Calcium Citrate vitamins.
I'm willing to wait for Nintendo. Just like I'm waiting for DVD region encoding to (legally) die before I by a DVD player.
-Paul Komarek
Your previous post questioned if compiler names (or somesuch) should preface the names of the binaries they produce. I think everyone agrees that this is a dumb idea. Including Stallman. I write this in response to the first question you ask in a post attached to mine: "So, since IBM used to (and for all I know, still does) compile different parts of OS/2 with different compilers, they should have prepended the names of all the compilers to its name?" I simply wrote that building an OS around the Linux kernel basically requires building a GNU-based system.
;-)
;-)
I agree that the RMS quotes you provide are attacks on Torvald's commitment to Free software. What I refuted was your previous comment "He bitches and moans that Linus doesn't have enough ideological purity." The quotes you provide don't support your previous statement. RMS is clearly *not* bitching an moaning that Torvalds has too little ideological purity (an absolute statement). Instead, RMS complains that Torvalds message says "...'Non-free software is ok; I use it and develop it myself.'". That is, Torvalds' comments are contrary to the stated position of the GNU project (about which RMS does *not* complain); and because of his influence when Torvalds states these views, the GNU project's goals suffer.
As we both admit, we don't know for sure what would have happened without the linux kernel (me: FreeBSD wouldn't be much larger than it is today, because of its ISP-centric focus; you: FreeBSD would have grabbed most of the current GNU/Linux-using population, if I understand you correctly). However, I doubt RMS would be much less influential. He's been well-known for, geez, 30 years now? I don't believe his fame has increased tremendously in the last 5 years. In large part because most people still aren't paying attention to the contributions of the GNU project. On the other hand, RMS' infamy has probably grown in the last 5 years.
But those who know RMS' history wouldn't be any less impressed with him had Torvalds not created Linux (the kernel). RMS matched an *entire* company's software output to guarantee a Free lisp machine implementation. He created gcc, gdb, emacs, and the GPL; the first two he made *by himself* (I'm not sure how much of the external contributions to TECO and such made it into the first GPL'd emacs). RMS received the McArthur Genius awared, and started the Free Software Foundation. And he still finds time to respond to his email these days.
I admit that being impressive and being famous is not the same thing. But I don't think RMS' fame has gained much from Torvalds' work. How many GNU/Linux-related articles in Business Week even mention RMS? I believe that RMS has earned his fame mostly by himself. And he's earned at least some of his infamy, too.
-Paul Komarek
Geesh, can nobody read carefully? This is the second time I'm responding to this issue.
I never said that the Linux kernel should be called GNU/Linux. As far as that goes, neither did Richard Stallman. I didn't even infer that. No careful reader could, in good faith, conclude from my comment that I felt that GNU should be prepended to things compiled with gcc.
Re-reading my previous comment (the grandparent of this one), I believe it is clear that I am arguing that
1) The kernel is not necessarily more important than the tools around it, as the great-grandparent comment suggested, and
2) Using the Linux kernel without the GNU tools is a waste of one's time, and may not be possible without modifications, and isn't worthwhile in any case.
I'm not sure we'd all be using *BSD. The free BSD projects have a different development model than Linux and the GNU project, with different priorities and structure. There is only *one* distrobution built around the FreeBSD kernel, for instance. The FreeBSD project is very centralized compared to the Linux and GNU models. The FreeBSD kernel doesn't yet support USB->serial adapters, despite Keyspan offering to help -- I assume this is because of lack of interest considering that they have very good support for other devices. It seems that the FreeBSD folks think that the (rather conservative, IMO) UFS filesystem is superior to ext2. The Linux folks seem to prefer ext2, worts, speed, and all.
Summing up the previous paragraph, it isn't at all clear to me that FreeBSD (or one of the others) would have emerged to fill the niche that GNU/Linux systems have. It seems likely that most of us would not have jumped on board the unix-on-my-PC train until something like GNU/Linux systems came along. Therefore, it isn't clear to me that we'd all be using *BSD.
"If Linus hadn't written a kernel, we'd all be running BSD, not the Hurd, because BSD would have been usable first." Heck, I remember installing 386BSD (or was that BSD386?) at the same time I tried SLS (an old GNU/Linux distro) in early 1993 (I didn't end up using either, because I didn't know what the heck I was doing and didn't commit to GNU/Linux until 1995 or 1996 when it became "easy" for "everyone" to join in =-).
Given the goals of RMS, it's not clear that he has any reason to appreciate Linux (the kernel). As an old, bad expression goes, to him the "Linux journey" has probably felt like watching your mother-in-law go over a cliff in your new Cadillac. As I argued above, it's not clear to me that one of the BSD forks would have had much effect on the eventual popularity of decentralized GNU-based systems. And the GPL is very popular in the BSD world. The *BSD folks use emacs and gcc, too. But their (non-GNU) filesystem utilities have screwed-up options, though (just kidding, just kidding).
RMS didn't bitch that Linus has too little ideological purity. He bitched about Linus' views and influence causing problems for the GNU project. I'm not completely sure there were no ad-hominem attacks as you suggest there were, and I'm not going to go back and double-check. I'll leave that as an exercise for some careful reader who managed to read my previous post correctly.
-Paul Komarek
"I see... So, every program compiled on unix with fsf tools should be GNU/whatever?"
;-)
I didn't say that. Looking back at my post carefully, I see that I was responding to your assertion "it is possible to have a linux kernel running without any gnu software."
I was arguing that it is nigh on impossible to use the Linux the kernel without needing GNU tools at some point.
As far as GNU/ls goes, I'm not sure anyone would want their name attached to a command with that many options.
-Paul Komarek
w00t! beat you by 94 registrations! =-) However, your number is intrinsically cooler than mine (794).
There is no reason to reply to this silly outburst.
-Paul Komarek
Your argument that many people use the GNU tools, but only one group uses the kernel name Linux, suggest that the GNU tools are more important?
Your "theory" about using the Linux kernel without any GNU tools seems silly. First, you'll need to get a binary of the kernel without compilation using gcc or gmake -- note that Linux is distrbuted as source, and is very sensitive to which *version* of gcc compiles it. You can find binary version of the Linux kernel and its required user-space utilities, but did these reequire GNU tools to create? Probably. Furthermore, the user-space stuff will probably require *at least* the GNU project's glibc to run.
Once you have a non-GNU compiled kernel, you'll need a boot loader, or just dd it to the boot sector of a floppy using non-GNU dd, and use some non-GNU version of rdev that doesn't require something provided by the GNU project (like glibc).
And once your kernel boots, you're done! Now if only you had a shell, or init that doesn't depend on something from the GNU project.
Once you've done all that, you'll have a pathetic non-GNU system wrapped around Linux? In the process, I'll expect you'll have spent a lot of money, time, and sanity; and probably rewritten major portions of Linux and the kernel build system, not to mention the C library. Whee!
-Paul Komarek
"it is possible to have a linux kernel running without any gnu software."
I'd like to see one. While you don't care what compiled it, it still needs to be compiled. Or did you forget that kernel.org distributes *source code*. And you need more than a compiler. You need gmake (almost certainly you need GNU's make), and you need some version of sed, and probably awk. It seems likely you'll need a shell, too. You can buy some of those, but who will replace gmake? Or has one of the propietary Unix vendors cloned all the features of gmake by now?
Oh, by the way, the kernel source comes as a compressed tarball. You'll need to decompress and untar it. You don't need gnu for this, but please point out which tools you'll use instead and how much they cost -- including your Solaris or Tru64 or HP-UX (or whatever) single-user license, please.
It also needs module tools (usually), and all kinds of other user-space utilities. If just one of these utilities is part of the GNU project(say, for instance, GRUB and LILO, though could always write the binary kernel you downloaded onto your MBR -- that might work), your kernel is going to be a real pain in the butt. Oh, and you'll need filesystems and filesystem tools, and a shell to log into, and a login program, and an init program, and you might want manual pages, too. Good luck finding all of this stuff without touching the GNU project.
Hell, even the *BSDs use a fair amount of GNU Project code. They don't use linux, though. Maybe the kernel *is* more easily replaced than the user-space utilities its programmers and users rely on?
-Paul Komarek
Geesh, will people quit equating lines-of-code with what is most important? This nonsense was all over the place when the RMS originally made his GNU/Linux request. Thankfully the parent post doesn't dwell on the size of XFree86.
In particular, comparing XFree86 to the GNU project code is just silly. While XFree86 is *convenient*, it is certainly not central to the goals of the GNU project or the kernel we call linux.
The essential things to the GNU Project are things like bash, lilo, sed, awk, gcc, gdb, gmake, and emacs. The GNU project also needs a kernel, and right now they need the Free parts of the linux kernel to be complete. Linux needs things like gcc, gdb, gmake, lilo, bash, perl, sed, maybe awk (haven't seen it, but hard to imagine it's not in some script somewhere). Thus the GNU Project and the Linux (kernel) project are working *together*.
Furthermore, the GNU Project is a lot closer to finishing their kernel project, the HURD, than the Linux project is to finishing their own development toolchain. If one was to decide who could survive without the other, I expect only the GNU Project would come out standing.
Just how will you compile your Linux kernels without gcc and gmake and bash and sed? Using MSVC? Borland? The Portland group compilers? Compaq's cc in Tru64? I suppose the x86 folks could use one of Intel's compilers, but that doesn't solve gmake, bash, and sed dependencies. On the other hand, the HURD is (finally) working, even if only barely. If Torvald's decided to eschew the GNU Project entirely, I'd bet Alan Cox, among many others, would not stick around -- they'd either fork Linux, or work on the HURD.
I don't think RMS has ever accused Torvalds of hogging the "limelight". RMS isn't in this for ego or popularity. It seems clear from his comments that he wants the GNU name to be present so that people will know the GNU project exists, and hopefully go on to learn about its goals. RMS has more fame and credit than he probably cares for (consider just the LISP machine Battle Royale).
I don't think Stallman is content to believe that the "war" is already over. That's why he is still fighting. If he were a quitter, the GNU project would never have emerged. I wouldn't be surprised if people told him he had "already lost" when the lisp or emacs battles started.
-Paul Komarek
1) Is the FSF using linux, the kernel, at all?
;-):
2) If answer to 1) is yes, then are they using the parts they find illegal? Almost certainly not. Linux, the kernel, has many parts licensed by different people. Choosing only the parts with Free license, and among those, only the ones who comply with their license, seems entirely reasonable to me. In fact, you could call the resulting Free kernel a derivative work of the parts with a Free license, if you wanted. No need to throw the baby out with the bathwater.
In my opinion, your argument is valid but without premise. You would need to provided evidence that the FSF was using non-Free code. Once you've named a non-free part of the linux kernel, you'll need to show that the FSF is using that code.
Just because they downloaded a kernel from kernel.org with non-Free parts, doesn't mean they're using those parts. Especially since those parts are likely to be completely unnecessary device drivers. For instance, my desktop and gaming linux system shows the following at the top of lsmod's output (reformatted to fit the narrow text input box
"Module Size Used by Not tainted"
The "not tainted" part is important to me. If I'm not misinterpreting it, it means that every module loaded (and I think also that everything compiled in) has a license tag which indicates a Free license. In particular, modules with no stated license will taint the kernel, creating false negatives. "not tainted" means "not tainted", every time. I guess I should really check whether "not tainted" necessarily means only Free software is included.
-Paul Komarek
Given the area that they're covering, I doubt it. I would assume that porn in public spaces is not as popular as porn in your home. This wifi network appears to cover some popular public parks, and a *lot* of office buildings. It seems unlikely that the clerks or shoppers at Kaufman's are going be pirating music while riding the old wooden escalator. Especially given the really low bitrate an individual is likely to receive.
;-). When I ride the trolly, there's at least another 15 people on board. Again, it's a public space. Same goes for the cafes that my ISP, Telerama, is covering. We're mostly talking about students who are working on research/homework, or addicted to the web (and coffee).
Allowing wifi on the trolley sounds like a great idea, too (at least for me
-Paul Komarek
From what I remember, the Israeli pilots were prohibited from excessive running -- which definitely seems to disagree with what you are saying. If you are part of the Israeli Air Defense Force (I think that is what they call it), then I'm clearly in the wrong (and either my memory is screwed up, or the program was). I don't claim to be an expert; I just wanted to share some piece of (possibly wrong) trivia I ran into.
I've practiced the "grunting" thing before, just for fun. It makes your face turn red in a hurry!
-Paul Komarek
Agreed. I especially like your phrase "Microsoft is, inevitably, sucking hind teat". =-)
-Paul Komarek
It appears I should have been more clear. I thought the guy could not have been seriously objecting to this law. I've never lived in a state that didn't have this law (well, I've only lived in Washington and Pennsylvania). I think it's a damn good law.
When one is driving in a school zone, behind a bus, or in a residential neighborhood, one should be watching for kids. However, there are a lot of reckless, impatient drivers out there. Heck, I've even seen one driver ignore school crossing guards and go blasting through a crosswalk while kids were crossing. I think it's clear that such a law is needed in order to curb the many idiots behind the law. For those of us with brains and a heart, we need to follow the law, too.
In some places where drivers are particularly reckless, I've seen the schoolbus stop the bus across both lanes to stop people who ignore the fine. Fine by me. The schoolbus is in charge, as far as I'm concerned.
-Paul Komarek
There are many factors which contribute to one's ability to sustain G-forces. While it is tempting to think that a very fit person could sustain more head-to-toe Gs, that is not necessarily true. If it is easy for the blood to flow out of the head (for instance, in an athlete whose exercise has promoted high levels of capilary and/or blood vessel growth -- note, I'm definitely not a doctor), then one can't sustain as many Gs. If the distance between the head and the neck is short, it makes it easier for the heart to achieve the pressure necessary to provide blood to the brain. I'm sure there are many other bizarre contributing factors.
As I mentioned, I'm talking out my rear-end to some degree. My source of information comes from some show I saw several years ago about the selection and training of fighter pilots in Israel. IIRC, they measured the heart/brain distance (or at least neck length) of candidates, and limited their pilots' cardiovascular exercise. Who would have guessed?
-Paul Komarek
Somebody mod the parent up, this is a pretty funny post. Clearly this guy can't be serious.
-Paul Komarek
I don't even think there is a debate of which console is best for young kids. Get whatever Nintendo makes.
For adults with social skills that like to play games in the same room with friends, there's not much debate either. The gamecube still wins.
The other consoles seem to be targetted at the same audience as most PC video games. I think the description is something like "males between 15 and 35".
-Paul Komarek
I think that Nintendo opened at $200 and MS at $300. It's been so long since the PS/2 came out that I have no idea what it opened at, and by now you'd have to take inflation into account to make a valid comparison. ;-)
Early adopters always get screwed on price. It's their choice. The same is true in the PC market. Consider, for instance, the IDE hard drive market and the graphics card market.
-Paul Komarek