One must not take for granted the vast powers of IPv6 auto-configurability. Even if the devices won't have any world-routable IPv6 addresses, one could have a device that hands out the addresses based on the devices' MAC addresses. Zero administration effort, consumer-friendly.
For some reason, I get a hard-on every time I think of wifi community networks and IPv6 within the same train of thought. They go perfectly hand-in-hand...associate with a network, and boom, the router nearest to you gives you a prefix delegation with which you can give yourself an address in. Seamless, especially for ad-hoc networks.
Sean Kennedy, The Fucking Man (sktfm.tv) wants his episodes/broadcasts mirrored! His material is very informative, as well as extremely entertaining. Such media survives best in the wild.
I think my earliest computer memory was when I saw a NeXTcube in a store near where I live. Of course, at the time, I had thought it was just a big Lego box or something:-/ This was about 1990--I was about four years old at the time.
As for my earliest memory, it was a funeral. Not the funeral itself, but I do recall how my parents were taking me to some shop to get my suit adjusted; we were delayed by a train. I think I was three.
Conversely, you would be surprised at how many people would have taken the great-grandparent post as fact if left unmoderated. The joke is only so funny up to a point.
Not that I would be surprised or anything to see RMS advocating the usage of the GNU/ prefix when referring to the firmware distributed with the newest batch of smart self-pleasure toys. Now, that's progress.
uClibc is a very complete glibc replacement; it uses the latest glibc headers to maintain this compatibility. Besides this, where it differs from diet libc is that diet-libc does not have a dynamic linker.
In short, do not even attempt to build GNOME or KDE against diet libc. Many parts of GNOME and KDE depend on a dynamic linker. So, your best chances would be with uClibc, as it already has a native gcc toolchain available, a native dynamic linker (though not for all platforms supported by uClibc), and has more 'diverse' features which diet libc may be missing.
This is the perfect place for open-source software in the government, compared to, say, branches of the armed forces.
Even though the transititon is probably more of a money issue than an advocacy one, the pyramid effect might just trickle down and help us out in the long run, if they're satisfied with Linux desktops. Who says the government won't pitch in?
Treat your computer like your property! Personally, I would remind the man that he is dealing with MY property, that EULA's are made to be disagreed with, and that I would no longer allow him to touch my property if he continues to install software without me agreeing.
This makes perfect sense to me, as there are great differences between my house and my computer. Fine, he can run a cable line through my wall; I don't necessarily feel like doing that part myself. But, nobody touches my computers.
Of course, the easiest way around this is just to be running some variant of Unix at the time. I think I scared (or baffled?) the poor man with my NeXTcube. He backed off and let me do the rest of the installation myself.
One of the small-town grocery stores nearby actually sells CD-Rs and CD-RWs at pretty decent prices, and place them next to the blank VHS tapes in the store. Seeing as to how they're becoming more ubiquitous, and devices like the Terapin VCD Recorder (at http://www.thinkgeek.com/electronics/video/57a6/) are starting to appear, perhaps CD-RWs could give VHS a run for its money, with comparable video and audio quality, as well as interoperability with a computer. For instance, you just missed Everybody Loves Raymond, so you hit KaZaA and somebody uploaded a VCD for you. So yeah, they've got their merits.
ls' large size is somewhat justified. Implementing ls, one would have to include code for parsing symbolic file modes (which is a considerable amount), terminal handling code (how else will your files be sorted in columns without a little knowledge of your terminal?)
Then again...it's not so justified; you're probably looking at GNU ls. At compile time, it internally links in libfetish.a (which contains tons of code unrelated to what ls does), and libintl.a (which isn't always needed, either).
Just reading this article makes me shit my pants and utterly pisses me off. I was almost crying after a lame sob story from a college guy not having a computer to get online with when he bought the last Broadband Adapter at our local game trade shop...ARRGH!
Are there ANY people out there who have retired their Dreamcast Broadband Adapters, and would be willing to donate to an aspiring coder? With a Broadband Adapter, I could do things not limited to novelty, such as:
Test my window manager on a somewhat
constrained platform (16MB of RAM - the
space consumed by the kernel and the initial
ramdisk).
Write an accelerated XFree86 driver using
the tile-based rendering capabilities of the
PowerVR graphics chip.
Eventually port KallistiOS' OpenGL
implementation over to Linux.
Use the Dreamcast as a self-hosted
development environment with an NFS mount to
my main box.
Use the Dreamcast as a simple, on-the-go
replacement for my main computer.
As one would imagine, accomplishing the above would be extremely slow, as much code would have to be transferred back and forth, and many binaries would be dumped to my box for storage.
I can think of more things if you'd like. And as soon as I'm done with the Broadband Adapter, I'll pass it onto someone else who may need it.
[The above has been a rant. Thank you for reading.]
I would kill to have the Broadband Adapter so I, myself, could continue my Dreamcast development. I no longer find it fun to wait for about an hour for the serial slave to upload code AND emulate a CD-ROM drive. And when I do not have time, I'm forced to go to the store and spend money on CD-Rs that would probably be turned into coasters throughout the development cycle.
Sure, the Dreamcasts are perhaps disposable, but the Broadband Adapters available certainly ARE NOT. There are people who have far better uses for them than to see them trickle away. Buy a uCsimm kit. They are much too small to even be noticed, and fit nicely above a ceiling tile. If one so desires, it could probably also be placed in a child's toy.
Without a POSIX-style/bin/sh present (I.e., accepts -c "commands"), any use of system() will be broken. Oh, and obviously, startup scripts, regular shell scripts, and perhaps many utilities would be impossible to implement without a shell. Basically, in that kind of situation, I would go with busybox, or just Kenneth Almquist's ash. Experiment and see what you would need.
I'm rather interested in taking that copy off of your hands. By the way, are you Dark Zero from k5? If so, I'll be glad to ask questions outside of Slashdot.
Re:You cannot deny GCC is the heart of free softwa
on
The Stallman Factor
·
· Score: 1
Indeed. My comments regarding gcc were more aimed at comments on this thread posted before mine. However, I must say you do carry a very good point--If it weren't for the GNU toolchain, I would have probably reinstalled Windows years ago to continue writing lame code for lame purposes. The GNU toolchain most definately showed me the light; because it is freely available, ubiquitous, and well established as a paramount in the world of compilers. Though some of the comments in this thread suggest that one would not be able to write code at all without it.
And time to address that little pet peeve which I tried to make clear in my previous post. RMS pushes the 'GNU/' prefix onto Linux distribution names--Because at the heart of those distributions are all of the GNU packages that make them tick. That's understandable. However, my previous comment was suggesting against RMS also pushing for the 'GNU/' prefix on the Linux kernel itself--Was Linux based on any ideas of any GNU projects? Most certainly not. Is the Linux kernel a GNU package? Most certainly not (my definition of a GNU package would be one developed in-house by the FSF). And finally reaching my previous point, I figure RMS wants the kernel itself to be known as GNU/Linux simply because GCC plays a major role in the development of the kernel. OF COURSE IT DOES! The text editor(s) used to write the damned kernel did, too. Hell, by that logic, all of our projects should be known as GNU/GCC/GDB/VI/Emacs/(insert name here).
I wonder why RMS isn't pushing for names like GNU/NetBSD at this point.
Re:You cannot deny GCC is the heart of free softwa
on
The Stallman Factor
·
· Score: 1
Do not get me wrong, I am very appreciative of what functionality the GNU C Compiler gives me to program quality software. However, I do deny any software I make as descendants of gcc--gcc is simply a utility for generating assembly code for many different types of machines. I did not base my software on any of gcc's methods of translating C code into assembly, or its functionality. Of course, I just as easily could have used lcc to generate assembly which would be passed to the assembler, as. By this logic, the very comment you read would be a derivative work of Netscape, which in turn is a derivative work of Motif, which was created based on ideas of MIT, and so on. And most certainly, I do not consider the software I write as derivatives of vi; I did not mimick the function of vi in any of my code (well, except a vi clone I played around with). Case in point, I respect all of the utilities in which make my own Open Source contributions possible; however, I could have just as easily used proprietary utilities to achieve the same product. And I actually find it easier to give kudos to the utilities that make my work possible by writing about them in forms longer than a 'GNU/' prefix.
Slashdot Mirror
Here's a screenshot inspired by this discussion.
Fair enough. :) Here it is. Had to jump through several hoops to get to it (such as HTTP version discrepencies, etc).
:)
Did ya think I was lying? And I didn't say it would be pretty, either.
Oh ho ho...you think you're so cute. Well, let's just see when I get a screenshot of your comment using WorldWideWeb!
Now, I didn't say it would be pretty.
One must not take for granted the vast powers of IPv6 auto-configurability. Even if the devices won't have any world-routable IPv6 addresses, one could have a device that hands out the addresses based on the devices' MAC addresses. Zero administration effort, consumer-friendly.
For some reason, I get a hard-on every time I think of wifi community networks and IPv6 within the same train of thought. They go perfectly hand-in-hand...associate with a network, and boom, the router nearest to you gives you a prefix delegation with which you can give yourself an address in. Seamless, especially for ad-hoc networks.
Am I alone in thinking this?
Sean Kennedy, The Fucking Man (sktfm.tv) wants his episodes/broadcasts mirrored! His material is very informative, as well as extremely entertaining. Such media survives best in the wild.
I think my earliest computer memory was when I saw a NeXTcube in a store near where I live. Of course, at the time, I had thought it was just a big Lego box or something :-/ This was about 1990--I was about four years old at the time.
As for my earliest memory, it was a funeral. Not the funeral itself, but I do recall how my parents were taking me to some shop to get my suit adjusted; we were delayed by a train. I think I was three.
Conversely, you would be surprised at how many people would have taken the great-grandparent post as fact if left unmoderated. The joke is only so funny up to a point.
Not that I would be surprised or anything to see RMS advocating the usage of the GNU/ prefix when referring to the firmware distributed with the newest batch of smart self-pleasure toys. Now, that's progress.
I'm seriously doubting that the Free Software Foundation lay any claim to uClibc or Busybox. So, sir, you are wrong in your troll.
uClibc is a very complete glibc replacement; it uses the latest glibc headers to maintain this compatibility. Besides this, where it differs from diet libc is that diet-libc does not have a dynamic linker.
In short, do not even attempt to build GNOME or KDE against diet libc. Many parts of GNOME and KDE depend on a dynamic linker. So, your best chances would be with uClibc, as it already has a native gcc toolchain available, a native dynamic linker (though not for all platforms supported by uClibc), and has more 'diverse' features which diet libc may be missing.
Those are a lot of processes...:-/
This is the perfect place for open-source software in the government, compared to, say, branches of the armed forces.
Even though the transititon is probably more of a money issue than an advocacy one, the pyramid effect might just trickle down and help us out in the long run, if they're satisfied with Linux desktops. Who says the government won't pitch in?
Treat your computer like your property! Personally, I would remind the man that he is dealing with MY property, that EULA's are made to be disagreed with, and that I would no longer allow him to touch my property if he continues to install software without me agreeing.
This makes perfect sense to me, as there are great differences between my house and my computer. Fine, he can run a cable line through my wall; I don't necessarily feel like doing that part myself. But, nobody touches my computers.
Of course, the easiest way around this is just to be running some variant of Unix at the time. I think I scared (or baffled?) the poor man with my NeXTcube. He backed off and let me do the rest of the installation myself.
One of the small-town grocery stores nearby actually sells CD-Rs and CD-RWs at pretty decent prices, and place them next to the blank VHS tapes in the store. Seeing as to how they're becoming more ubiquitous, and devices like the Terapin VCD Recorder (at http://www.thinkgeek.com/electronics/video/57a6/) are starting to appear, perhaps CD-RWs could give VHS a run for its money, with comparable video and audio quality, as well as interoperability with a computer. For instance, you just missed Everybody Loves Raymond, so you hit KaZaA and somebody uploaded a VCD for you. So yeah, they've got their merits.
ls' large size is somewhat justified. Implementing ls, one would have to include code for parsing symbolic file modes (which is a considerable amount), terminal handling code (how else will your files be sorted in columns without a little knowledge of your terminal?)
Then again...it's not so justified; you're probably looking at GNU ls. At compile time, it internally links in libfetish.a (which contains tons of code unrelated to what ls does), and libintl.a (which isn't always needed, either).
Blame GNU!
Thanks. That was a refreshing reply.
Just reading this article makes me shit my pants and utterly pisses me off. I was almost crying after a lame sob story from a college guy not having a computer to get online with when he bought the last Broadband Adapter at our local game trade shop...ARRGH!
Are there ANY people out there who have retired their Dreamcast Broadband Adapters, and would be willing to donate to an aspiring coder? With a Broadband Adapter, I could do things not limited to novelty, such as:
As one would imagine, accomplishing the above would be extremely slow, as much code would have to be transferred back and forth, and many binaries would be dumped to my box for storage.
I can think of more things if you'd like. And as soon as I'm done with the Broadband Adapter, I'll pass it onto someone else who may need it.
[The above has been a rant. Thank you for reading.]
I'm sorry. http://www.uclinux.org/
I would kill to have the Broadband Adapter so I, myself, could continue my Dreamcast development. I no longer find it fun to wait for about an hour for the serial slave to upload code AND emulate a CD-ROM drive. And when I do not have time, I'm forced to go to the store and spend money on CD-Rs that would probably be turned into coasters throughout the development cycle.
Sure, the Dreamcasts are perhaps disposable, but the Broadband Adapters available certainly ARE NOT. There are people who have far better uses for them than to see them trickle away. Buy a uCsimm kit. They are much too small to even be noticed, and fit nicely above a ceiling tile. If one so desires, it could probably also be placed in a child's toy.
"Dude, can I borrow your cell phone?"
Without a POSIX-style /bin/sh present (I.e., accepts -c "commands"), any use of system() will be broken. Oh, and obviously, startup scripts, regular shell scripts, and perhaps many utilities would be impossible to implement without a shell. Basically, in that kind of situation, I would go with busybox, or just Kenneth Almquist's ash. Experiment and see what you would need.
I'm rather interested in taking that copy off of your hands. By the way, are you Dark Zero from k5? If so, I'll be glad to ask questions outside of Slashdot.
And time to address that little pet peeve which I tried to make clear in my previous post. RMS pushes the 'GNU/' prefix onto Linux distribution names--Because at the heart of those distributions are all of the GNU packages that make them tick. That's understandable. However, my previous comment was suggesting against RMS also pushing for the 'GNU/' prefix on the Linux kernel itself--Was Linux based on any ideas of any GNU projects? Most certainly not. Is the Linux kernel a GNU package? Most certainly not (my definition of a GNU package would be one developed in-house by the FSF). And finally reaching my previous point, I figure RMS wants the kernel itself to be known as GNU/Linux simply because GCC plays a major role in the development of the kernel. OF COURSE IT DOES! The text editor(s) used to write the damned kernel did, too. Hell, by that logic, all of our projects should be known as GNU/GCC/GDB/VI/Emacs/(insert name here).
I wonder why RMS isn't pushing for names like GNU/NetBSD at this point.
Do not get me wrong, I am very appreciative of what functionality the GNU C Compiler gives me to program quality software. However, I do deny any software I make as descendants of gcc--gcc is simply a utility for generating assembly code for many different types of machines. I did not base my software on any of gcc's methods of translating C code into assembly, or its functionality. Of course, I just as easily could have used lcc to generate assembly which would be passed to the assembler, as. By this logic, the very comment you read would be a derivative work of Netscape, which in turn is a derivative work of Motif, which was created based on ideas of MIT, and so on. And most certainly, I do not consider the software I write as derivatives of vi; I did not mimick the function of vi in any of my code (well, except a vi clone I played around with). Case in point, I respect all of the utilities in which make my own Open Source contributions possible; however, I could have just as easily used proprietary utilities to achieve the same product. And I actually find it easier to give kudos to the utilities that make my work possible by writing about them in forms longer than a 'GNU/' prefix.
That's a lot of power for a kid's toy! Kind of reminds me of what Precursor (guy from endeffect.com) does with his Palm.