I agree that 's a mess. The internic won't register it for you, but I believe that yes, you could have a 3rd-level domain with backspaces in it. or dots (how about....hotmail.com?). I think that what section 3.1 is saying is that you should follow the conventions that are appropriate - ie., the limitations of/etc/hosts, or the rules in the various mail RFCs.
RFC 1035 states, in section 2.3.1 that labels should "start with a letter, end with a letter or digit, and have as interior characters only letters, digits, and hyphen." but it also says in section 3.1 that domains may use any ASCII characters. In fact, you are even allowed to have "."s in your domain.
So I believe that the all-numbers is merely a convention (and not a requirement) and also an InterNIC restriction.
Erhm, the reason the we say *nix instead of Unix is not to avoid trademark infringement, it's to get many of the unixes (unices?) out there, such as XENIX, Freenix (what Linux used to be called), etc. The Asterisk is a wildcard in a large variety of languages, including bash, regular expressions, and the DOS command line. In fact, *nix is not a good regular expression because of all the Unixes that don't follow it (irix, aix, linux, Solaris, etc.)
Trademarks only mean you can't sell a product and call it UNIX. You're still allowed to say the word UNIX to your heart's content, as long as a reasonable person wouldn't confuse you for the person who owns the trademark.
That's all fine; but I think that using free software instead of proprietary software is not just a technical decision; it engages other issues as well.
For example, closed-source solutions hide what is going on. There is no real way to tell if Windows NT is sending your E-mail back to the NSA.
But beyond any practical advantages of free software, there is also a moral issue at stake. If we really valure the freedom that free software gives us, then mabye someday Microsoft will come out with a better OS than Windows. (I know, work with me on this) I would say that if there is a Free Software solution that does the job, then go with it.
Undernet dosen't have this problem (nick theft), because they maintain a server -you take a nick, it's yours as long as you keep using it. EFnet, on the other hand, takes the position that nicks are ephemeral anyway, and no one owns them.
As for the network being fragile, this is due primarily to its distributed nature and the amount of DoS it recieves. If you wanted to you could create an IRC network w/ only one server, which would be pretty stable as long as the server didn't crash. IRC servers recieve lots and lots of denial of service attacks to the servers themselves, and also their routers, DNS servers etc. But this is a consequence of the kind of people on IRC, not the protocol itself.
What I want to see is if someone can provide a good reason why the IRC protocol can't serve as a standard IM Protocol.
Well, but this is a defect in the client, not the protocal. It would be very easy to have your client say "oh-oh", make toast, or SYN flood Paris whenever you recieve a MSC.
Domain suffixes don't mean anything anymore. It could be hosted on a dedicated server in Exodus' colo center, and be entirely immune to the slashdot effect.
For that matter, Slashdot itself is a.org and is entirely immune.
They appear (on the same page) to have applied for a patent on a light that tells how its doing. Nevermind that old mainframes used to have a bunch of those lights (OT: IMO the media wants to bring the blinkenlights back) but also Sony's AIBO flashes different colored lights depending on its mood.
So there. Not that the patent office reads slashdot.
Something that's not been made clear to me is why IRC won't work as an instant messaging platform. Near as I can see, it has everything required - file transfer (DCC), buddy lists (notify), chat (duh), private messaging (duh), and a lot of other stuff as well.
So why dosen't everyone just make an IRC-based instant messenger?
OK, the process your C code goes through to get to machine code looks something like this:
C code -> Preprocessed C code -> assembly code -> machine code -> linked code
A description of the processes:
C code. This is obvious. Then we feed it through the preprocessor, which deals with all those nasty #defines and #includes. In a small program, this can increase the size of the file by several orders of magnitude.
After preprocessing, we have preprocessed C code (surprise!) which we then feed to the compiler. The compiler takes all that C code, runs some optimizations on it, tilts it sideways (and lets the dirt fall out), pokes at it, and turns it into assembly code.
Assembly code is processor-specific. Naturally some programs start here (like the current topic). Assembly code deals with the raw hardware (ie., registers etc.) but is human-readable, to use the term loosely. It is still in ASCII, and contains one processor instruction per line. This is also the last step that allows comments.
Feeding the assembly code through an assembler yields binary code. This is raw machine code that the processor feeds on. If you are running it inside an operating system, typically it will be encased with some other materials to yield object code.
Finally, if you are on an operating system and want to make an executable (or library), you link your object code. This notes which shared libraries are required, as well as including the structures for an executable (exe, a.out, elf, etc.) For the most part, all code is linked except for bootstrap code.
A true Klingon sysadmin can stand up to ANY slashdot article! You have insulted my family honor by saying that my server will fail before slashdot! Prepare to die!
True, but if you were running a counterfiting operation, then technically your victims should get the money, not the government. So mabye we should just make microsoft send everyone who ever bought windows a $25 rebate?
Name your editor, there is probably an ispell module for it. I know that vi, emacs, and pico all have this, which probably accounts for at least 95% of linux users' text editors.
This would be disastrous for our economy. Are you aware how much microsoft stock there is out there? billions and billions. Much of this stock is owned by the middle class; is it really fair to them to take their assets in this manner?
How can you say that? Long distance service has gotten more compeditive recently, not less. Witness the drop in rates. AT&T is presently offering 7 cents a minute, 24x7. It was only a few months ago they were offering 15.
No no, I believe moderators select a category AND a direction, and that the two are not associated. so you could mark something (-1, insightful) just as you could mark it (+1, offtopic)
I agree that 's a mess. The internic won't register it for you, but I believe that yes, you could have a 3rd-level domain with backspaces in it. or dots (how about ....hotmail.com?). I think that what section 3.1 is saying is that you should follow the conventions that are appropriate - ie., the limitations of /etc/hosts, or the rules in the various mail RFCs.
So I believe that the all-numbers is merely a convention (and not a requirement) and also an InterNIC restriction.
Erhm, the reason the we say *nix instead of Unix is not to avoid trademark infringement, it's to get many of the unixes (unices?) out there, such as XENIX, Freenix (what Linux used to be called), etc. The Asterisk is a wildcard in a large variety of languages, including bash, regular expressions, and the DOS command line. In fact, *nix is not a good regular expression because of all the Unixes that don't follow it (irix, aix, linux, Solaris, etc.)
Trademarks only mean you can't sell a product and call it UNIX. You're still allowed to say the word UNIX to your heart's content, as long as a reasonable person wouldn't confuse you for the person who owns the trademark.
So I guess you block all mail from 1800flowers.com? There's no restriction on using numbers in domain names.
Allright, I'll bite. What OS other than Windows NT ends with a capital T?
That's all fine; but I think that using free software instead of proprietary software is not just a technical decision; it engages other issues as well.
For example, closed-source solutions hide what is going on. There is no real way to tell if Windows NT is sending your E-mail back to the NSA.
But beyond any practical advantages of free software, there is also a moral issue at stake. If we really valure the freedom that free software gives us, then mabye someday Microsoft will come out with a better OS than Windows. (I know, work with me on this) I would say that if there is a Free Software solution that does the job, then go with it.
Actually, the law does not just require open source, it explicitly requires Free Software. Richard Stallman has a webpage on the differences.
Don't bet on it. I got the redhat letter, and my only contribution was making a couple entries into their bugzilla.
Undernet dosen't have this problem (nick theft), because they maintain a server -you take a nick, it's yours as long as you keep using it. EFnet, on the other hand, takes the position that nicks are ephemeral anyway, and no one owns them.
As for the network being fragile, this is due primarily to its distributed nature and the amount of DoS it recieves. If you wanted to you could create an IRC network w/ only one server, which would be pretty stable as long as the server didn't crash. IRC servers recieve lots and lots of denial of service attacks to the servers themselves, and also their routers, DNS servers etc. But this is a consequence of the kind of people on IRC, not the protocol itself.
What I want to see is if someone can provide a good reason why the IRC protocol can't serve as a standard IM Protocol.
Well, but this is a defect in the client, not the protocal. It would be very easy to have your client say "oh-oh", make toast, or SYN flood Paris whenever you recieve a MSC.
Domain suffixes don't mean anything anymore. It could be hosted on a dedicated server in Exodus' colo center, and be entirely immune to the slashdot effect.
.org and is entirely immune.
For that matter, Slashdot itself is a
It's worse.
It's much worse.
They appear (on the same page) to have applied for a patent on a light that tells how its doing. Nevermind that old mainframes used to have a bunch of those lights (OT: IMO the media wants to bring the blinkenlights back) but also Sony's AIBO flashes different colored lights depending on its mood.
So there. Not that the patent office reads slashdot.
But, if you only used it for instant messaging (read: keep out of channels), and kept your nick secret, then who's going to find your IP?
Something that's not been made clear to me is why IRC won't work as an instant messaging platform. Near as I can see, it has everything required - file transfer (DCC), buddy lists (notify), chat (duh), private messaging (duh), and a lot of other stuff as well.
So why dosen't everyone just make an IRC-based instant messenger?
OK, the process your C code goes through to get to machine code looks something like this:
C code -> Preprocessed C code -> assembly code -> machine code -> linked code
A description of the processes:
C code. This is obvious. Then we feed it through
the preprocessor, which deals with all those nasty #defines and #includes. In a small program, this can increase the size of the file by several orders of magnitude.
After preprocessing, we have preprocessed C code (surprise!) which we then feed to the compiler. The compiler takes all that C code, runs some optimizations on it, tilts it sideways (and lets the dirt fall out), pokes at it, and turns it into assembly code.
Assembly code is processor-specific. Naturally some programs start here (like the current topic). Assembly code deals with the raw hardware (ie., registers etc.) but is human-readable, to use the term loosely. It is still in ASCII, and contains one processor instruction per line. This is also the last step that allows comments.
Feeding the assembly code through an assembler yields binary code. This is raw machine code that the processor feeds on. If you are running it inside an operating system, typically it will be encased with some other materials to yield object code.
Finally, if you are on an operating system and want to make an executable (or library), you link your object code. This notes which shared libraries are required, as well as including the structures for an executable (exe, a.out, elf, etc.) For the most part, all code is linked except for bootstrap code.
A true Klingon sysadmin can stand up to ANY slashdot article! You have insulted my family honor by saying that my server will fail before slashdot! Prepare to die!
Yeah, yeah (-1 offtopic/troll)
Probably slashdot has a script that automatically marks the first post -1 offtopic. ;)
True, but if you were running a counterfiting operation, then technically your victims should get the money, not the government. So mabye we should just make microsoft send everyone who ever bought windows a $25 rebate?
Name your editor, there is probably an ispell module for it. I know that vi, emacs, and pico all have this, which probably accounts for at least 95% of linux users' text editors.
This would be disastrous for our economy. Are you aware how much microsoft stock there is out there? billions and billions. Much of this stock is owned by the middle class; is it really fair to them to take their assets in this manner?
Anti-trust violation is a civil offense. That means no jail term for anyone.
Besides, what Microsoft has done is a lot more subtle and complex that stealing a VCR.
It can be found here.
How can you say that? Long distance service has gotten more compeditive recently, not less. Witness the drop in rates. AT&T is presently offering 7 cents a minute, 24x7. It was only a few months ago they were offering 15.
No no, I believe moderators select a category AND a direction, and that the two are not associated. so you could mark something (-1, insightful) just as you could mark it (+1, offtopic)
But, I don't really see how it's flamebait.
Um, where do I volounteer to play Quake on the test processors? I can test 4 at a time that way! ;-)