I will agree that KLyX isn't an applet for eye-candy window managers. However, that doesn't mean KLyX is the right way to go.
First of all, KLyX was based on an outdated version of LyX from the beginning, and it's developement has stalled compared to LyX. Secondly, while I personally find the appearance of Qt slightly nicer than that of XForms, I don't like it's license (For Qt1. Qt2 has mostly fixed this.).
Either way, that doesn't make it such a good thing. LyX is currently in the process of moving towards GUI independance, so that it can be used easily with mulitple GUI toolkits, including gtk and Qt. When you consider that LyX is much more featureful and stable than KLyX, that it seems to have greater active developement, and that it is moving towards a GUI independance that will render KLyX obsolete, it seems silly to pushy KLyX over the superior LyX.
Well, he is intending to do a review of KLyX, which is based on the LyX program, which is a WYSIWYM (What You See Is What You Mean) word processor that basically acts as a GUI front end for LaTeX.
Although, I do have to wonder, why KLyX instead of LyX? While I will admit, I prefer the Qt appearance over what's used in LyX, it's not enough to get me to switch. LyX seems to be, in my experience, a more stable and more featureful program.
If you want the power of LaTeX with the ease of a word processor, you really need to give LyX a try.
1) If you know enough to remove or shrink your Windows partitions and install Linux, will you not know enough to then remove it?
2) Why in the world would you go through the trouble of installing Linux if you wanted to remove it to install Windows?;-)
The only other situation I can think of is if you had a box that came with Linux preinstalled, and you wanted to put Windows on it...but, as it stands now, most machines are cheaper without an OS, or with Windows, than with Linux, simply due to greater competition among machines coming bundled with Windows.
After reading the reviews they posted for all of the Linux distributions, I think they need to change the title of this review. This wasn't a review of the distributions. This was a review of their installations, along with a few extra comments derived from the installation, the distributions' home pages, and their printed material thrown in for good measure.
The biggest problem I have with so many of these reviews, is how much they concentrate on the installation. Yes, I'll admit that getting Linux installed is the first hurdle, and often one of the more difficult ones. However, it's still only the first step, and it's not impossible. With a helpful friend or two, or an IRC channel, or even some good documentation, your average person can usually install Linux without too much trouble.
I have yet to see a review that even comments on upgrades. What happens when I want to upgrade my FooLinux box from version 4 to version 5? What is involved? How stable is the upgrade? What are the chances of it keeping my box down for an extended period of time? Does it require a reboot? Multiple reboots?
Not only that, but how about some reviewing of their utilities? Do they use some sort of package format? How compatible are the packages with other distro's using that same package format? How good are their package management utilties? Administrative utilities?
Better yet, reduce the number of passwords that you have to remember.
I've found that about 85% or more of the passwords I need to remember are login passwords. So, in an effort to cut that down, I began using the RSA authentication available with Secure Shell. This lets me use the same password (passphrase, actually) to access all of them, while also allowing me a very quick and easy way of changing my password, and the increased security that comes with requiring my private ssh identity along with my password.
1.) Does Debian have any system install automation like Redhat's kickstart?
This I can't answer for sure, because I haven't had occasion to try it, but I'm pretty sure I remember hearing about this capability for Debian.
2.) How good is the Debian package management?
To put it simply, Debian's package management is the best available anywhere. It will retrieve packages for you, handle dependancies, handle updates, and overall, just work amazingly. It basically does it all.;-)
For example, if I had a Red Hat 5.2 box, and I wanted to upgrade it to Red Hat 6.0, the most common upgrade path is to insert the 6.0 CD, reboot my computer, and tell it to upgrade the current install.
To upgrade a Debian 2.1 system to Debian 2.2, I change a few lines in/etc/apt/sources.list to point to the potato directories, and I type 'apt-get update; apt-get dist-upgrade'. Then, I sit back and wait for it to tell me that it's finished. That's it. No rebooting, no hassle, just sweet and worry free.;-)
I would like to use it with the 13 boxes as well...Right now if I need to update something on the boxes I make an rpm, put it in a certain directory, and it installs itself that night. Would something like that be possible in debian?
Yep, it's very possible, and very easy. You can either take the program you want to install and make a Debian package out of it (no more difficult than making an RPM, and many consider it much easier), or you can take a current Debian package for that program, get the source package, change it however you want to, build a new Debian binary (and/or source package), and 'dpkg -i ' to install it. Or, you can even set up your own custom Package listing on a local ftp/NFS server, and just 'apt-get install ' to have it download and install the program for you.
To give you one last example of how well Debian's package management is, I know someone who set things up so that when he attempted to upgrade a single package, Debian's package utility had to go out, get all the dependancies, plus those things the dependancies depended on, etc, and install it all in the proper order. He ended up having 457 packages updated, all at once, all with that one command. And it went without a single hitch or problem anywhere.
As a user of potato for a long time now, I've gotta say I think this is a great thing. I still have never seen an Operating System (or distro) as easy to administer, maintain, and upgrade, than a Debian box.
On another note, I noticed after a recent update that there is currently around 4500 software packages in the Debian repository. I wonder, is Debian now the largest linux Distro? Perhaps Linux support for DVD is coming none-too-soon, we might need it for Debian.;-)
It's very good, and then some. And I say this from experience, not just as a guess.
My school, the University of Nebraska at Omaha (UNO), was one of the first schools in the country to implement something like this, when we recently conducted our school government elections via online voting.
UNO was lucky, in that we already had much of the infrastructure in place. For the past 3 semesters, we've been able to register for classes online, check grades online, and for the past 2 semesters, pay for tuition via a credit card, online.
The authentication used for voting is the same as that required for the other online services. You have to login to a secure server using your student ID (Social Security Number) and your special 6 digit PIN number. This ensures that only registered students can vote, and that nobody can vote multiple times.
Another major advantage to online voting is the ability to distribute additional information. For example, we had biographies for all of the candidates online, so they could be viewed when voting. This helped with one of the most common voter complaints, lack of knowledge about the candidates and what they stand for.
We managed a 10% increase in voter turnout for the initial election, and an amazing 20% increase over that two weeks later in the Student President/Regent run-off election.
I was hoping to be able to provide links to more information about how we implemented things, but I'm afraid I can't find them at the moment. However, if anyone is interested in getting in touch with the people who handled the online voting, drop me an e-mail at topher@tconl.com. This really is a *good* thing, and well worth investigating further if you are considering it.
Does anyone know of a web site that has a list of browsers and what graphical formats they support?
I'd personally love to get rid of all my gifs, but while I prefer pngs to jpegs when I can, I've worked really hard to make my site universally viewable. I'd don't intend to give that up.
I guess I have to wonder whether we'd be better of converting most things to jpegs and waiting a little while longer befor going to pngs...
For those of you who enjoy good Anime, this is *definitely* worth checking out.
This is the man who created my all time favorite Anime, Kaze no Tani no Nausicaä (Nausicaa of the Valley of Wind). Not only is this a full length story (weighing in at right about 2 hours...I hate 45 minute 'movies' with no time for a real plot) but the depth, the characters, and the writing are all abosolutely amazing.
It's strange, but very few people in the US seem to be familiar with Nausicaa, despite the fact that I'm told it almost always places in the top 5 of every Japanese poll about the greatest Anime ever. I suppose part of the reason may be that no Enlglish dub was ever made (Technically, there was an English dub, it was called Warriors of the Wind. However, I refuse to count it, because it was so horribly chopped, mangled, and destroyed, it's not even worth watching. They cut out almost half an hour in the dubbing and editing, essentially everything that was important for the plot, leaving you with a few disjointed action scenes.)
If you'd like more information on Hayao Miyazaki and his Animes and Mangas, check out some of the following URLs:
There used to be three kinds of lies: Lies, Damn Lies, and Statistics. Now, there's a fourth type: Half-Truths.
To someone with little or no experience using and administering Windows NT or Linux, these may be believable. However, anyone who is intimate with multiple operating systems will no doubt have numerous questions afterreading this.
While I will admit, there are times where the Microsoft article makes valid points, most of it seems to be half truths, misinformation, myths, rumors, speculation, and opinion.
I don't have the time or interest to pick appart every single problem with this article, however, I'll give an example from the beginning of it. The Microsoft article starts with,
"First, it's worth noting that Linux is a UNIX-like operating system. Linux fundamentally relies on 30-year-old operating system technology and architecture."
In a strictly technical sense, this is in fact a true statement. However, it is only half true, if even that. It ignores the fact that through those 30 years of use, Unix has evolved and learned from the mistakes that it has made. The Unix that was originally created at Bell Labs is a far site from the modern Unix operating systems (including the Unix-like Operating Systems, such as Linux).
Linux has managed to learn from all that experience, while at the same time adapting to modern day usage. The fact that many of the ideas and theories that Linux/Unix is based are 30 years old is not a bad thing. It's a good thing.
This argument is in fact not even an NT vs. Unix argument. What it is, is an argument of Operating System theory and oppinion. The problem is, most of it is oppinion. It can't be proven. And much of what can be proven can be proven both ways. As Linus once said, "Microkernels might be a good idea, in theory. However, we aren't building a theoretical kernel. We're building a practical one that works." (These aren't exact words, but I think this captures the jist of it.)
However, you won't find that information in the article. If you want a lot of half truths, misleading statistics, and arguments of theory and philosophy, this is your article. If you want cold hard facts, then what are you wasting your time with articles for? You need to be talking to the people who work with these systems, or work with them yourself, and find out the truth first hand.
Lastly, remember that there are a thousand thousand uses for a computer out there. No single computer, or single OS, can handle all of them.
There is actually a very good reason that Linux has not traditionally been used on IRC servers. However, it is possible the reason is outdated.
The reason has to do with IRC servers needing one file descriptor for every user connecting to the server (along with a few others for miscellaneous purposes). Up until the 2.2.x kernel, Linux required extensive patching to handle more than 256 fd's per process. Not only that, but these patches were not well tested, nor were they great patches. While they allowed for additional file descriptors, they opened up numerous problems in that fact that the rest of the kernel wasn't ready to handle that many, especially not efficently.
I've followed and been involved with DALnet since late '95, and that's the reason there have been few (Note, 'few'. There have been some.) Linux IRC servers. Under low load (Read: When using a number of file descriptors that doesn't require the kernel to be patched and hacked) it's worked just fine. Under heavier load, other OS's (specifically FreeBSD, though NetBSD has occasionally made a strong showing as well) has been the OS of choice, simply because it has handled the file descriptor problem (along with a few other minor things) better than Linux.
Personally, I'd be very interested in seeing some testing done with the new 2.2.x kernel series. I'd have to guess it would manage *much* better, as I believe most of the patches to increase the number of file descriptors has been integrated and tested and fixed, so it's working properly.
To recap, running an IRC server with greater than 256 users is a Bad Idea, if you are using Linux and running a kernel prior to 2.2.x. Running an IRC server with up to around 1k users should be pretty good to go with a 2.2.x kernel. Above that, I don't feel qualified to say, as I've not tested it myself, yet.
If anyone has done any real testing with ircd on Linux 2.2.x kernels, I'd be very interested in hearing the results.;-)
I'd have to guess they just hadn't gotten to the one that you checked.
I happen to be a certifiable 'nobody' and I got this e-mail to. However, as yet, I've been unable to get to Network Solution's site, the connection is timing out.
Looks like they're prolly getting nailed with a less than pleasant response. Well deserved, however. When I signed up for a domain name, I *never* opted in to recieve any spam, advertisements, or security hole ridden web based e-mail crap.
I used to like InterNIC and Network Solutions, of course, that was a few years ago now. Back before they were losing their monopoly and acting like spoiled children about it.
Public keys are a part of PGP(or GPG). The way it works, when you use PGP you first generate a keypair. This consists of a private key, and a public key.
The private key you keep for yourself, and don't allow anyone else access to. This is what you use when signing something, or when decrypting something that is encrypted with your public key.
Your public key you can post on a website, publish to a keyserver, or even send via e-mail. This is what is used by other people to encrypt things. Something encrypted to your public key can only be decrypted by your private key.
I know this is a really basic explanation, but for information, check out http://www.pgp.com, http://www.gnupg.org, or do a search on your favorite search engine for PGP or public key cryptology.
MOSIX has actually been posted to slashdot before, and was the subject of a rather intense discussion.
They're project is a kernel module that requires kernel modifications, however, according to Linus's decision, this means they have to release the whole thing under the GPL if they decide to distribute the module. Last I heard, they hadn't decided what they were going to do about that, yet, as they wanted (needed?) to distrbitute it as binary only.
Why benchmarks? The claim made by the original poster was, the new record proved the stability and usefullness of the free unices. Now, ftp.cdrom.com runs FreeBSD. Obviously the new record can prove anything at all about FreeBSD only. It doesn't prove that Linux is _not_ just as capable either, but that's not the point made here.
The original poster was trying to make a point that seems to have gotten missed. You see, the original poster was saying that Free Unices are in fact capable of handling extremelly heavy load. FreeBSD just happens to be a free unix variant. Thus, this provides an example to saying, 'Free Unices can handle really heavy load.'
Everyone seems intent on picking them appart and treating them all as totally and completely seperate entities. However, that wasn't how the original post treated them.
If you put product "A" to a certain test, this proves absolutely nothing about product "B".
Perhaps not, but that isn't the comparison that was made. A much more apt comparison would be something like this:
You put product A, B, and C all together, and packaged them as product Foo. Now, Product B, part of Foo, does something really good. You now say, see, this is an example that shows that product Foo is really good.
Now, replace A with Linux, B with FreeBSD, and C with NetBSD. Now replace Foo with Free Unices. Perhaps this will clear up the intended point of the original poster?
I've heard lots of mixed reports on how far the Alpha port had progressed, though last I heard it was still fairly beta, but improving rapidly. The Sparc port though, last I heard, was pre-alpha still...
I love Linux as much or more than the next guy, and NetBSD sounds pretty cool, but how the heck does this record prove anything conclusively about NetBSD and Linux?
You misunderstand my point. I recently suggested at work that we use one of the various free Unices for a couple of servers. My suggestion was shot down, with the comment that none of the free Unices had ever been proven in a high stress, high load situation.
This is, in my oppinion, quite clearly an example where one of them has.
I love the free Unices. FreeBSD is stable as a rock on Intel hardware (though, unfortunately, not portable for crap yet). NetBSD has the stability of FreeBSD, along with the ability to run on damn near every single architecture available (even more than Linux). Linux just plain rocks, with it's stability, features, and amazingly fast evolution.
I also believe in using the right tool for the right task, and I often don't bother to differentiate between which one is better, or anything else. They're all Free Unices to me, each with their own strength and weeknesses.
This says a lot about FreeBSD, and the potential for Free Software in general. Don't make more out of it than there is to be made, though.
This is exactly my point. This is an example that shows very clearly that the Free Unices, and Free Software in general, *can* work, and *does* work. I'm not here to argue the specifics of each OS, or anything like that.
I look at this from the point of view that, when I show my boss evidence like this, all of the Free Unices win, and all of them become better recognised for their abilities by him.
It makes Free Software/Open Source advocates look intellectually dishonest.
I disagree. I see it as using a single example to prove a concept, as opposed to 'My OS is better than your OS.'
Anyone who ever claims that the free Unices aren't up to handling heavy load ought to see this.
I think this proves very conclusively that the free Unices (Linux, NetBSD, FreeBSD, etc) are all very capable, stable, powerful, and robust. I'd love to see a box running a commericial OS try to match this.;-)
Slashdot Mirror
I will agree that KLyX isn't an applet for eye-candy window managers. However, that doesn't mean KLyX is the right way to go.
First of all, KLyX was based on an outdated version of LyX from the beginning, and it's developement has stalled compared to LyX. Secondly, while I personally find the appearance of Qt slightly nicer than that of XForms, I don't like it's license (For Qt1. Qt2 has mostly fixed this.).
Either way, that doesn't make it such a good thing. LyX is currently in the process of moving towards GUI independance, so that it can be used easily with mulitple GUI toolkits, including gtk and Qt. When you consider that LyX is much more featureful and stable than KLyX, that it seems to have greater active developement, and that it is moving towards a GUI independance that will render KLyX obsolete, it seems silly to pushy KLyX over the superior LyX.
Well, he is intending to do a review of KLyX, which is based on the LyX program, which is a WYSIWYM (What You See Is What You Mean) word processor that basically acts as a GUI front end for LaTeX.
Although, I do have to wonder, why KLyX instead of LyX? While I will admit, I prefer the Qt appearance over what's used in LyX, it's not enough to get me to switch. LyX seems to be, in my experience, a more stable and more featureful program.
If you want the power of LaTeX with the ease of a word processor, you really need to give LyX a try.
And this brings up two interesting questions:
;-)
;-)
1) If you know enough to remove or shrink your Windows partitions and install Linux, will you not know enough to then remove it?
2) Why in the world would you go through the trouble of installing Linux if you wanted to remove it to install Windows?
The only other situation I can think of is if you had a box that came with Linux preinstalled, and you wanted to put Windows on it...but, as it stands now, most machines are cheaper without an OS, or with Windows, than with Linux, simply due to greater competition among machines coming bundled with Windows.
Oh, well. This was an amusing read, anyway.
After reading the reviews they posted for all of the Linux distributions, I think they need to change the title of this review. This wasn't a review of the distributions. This was a review of their installations, along with a few extra comments derived from the installation, the distributions' home pages, and their printed material thrown in for good measure.
;-)
The biggest problem I have with so many of these reviews, is how much they concentrate on the installation. Yes, I'll admit that getting Linux installed is the first hurdle, and often one of the more difficult ones. However, it's still only the first step, and it's not impossible. With a helpful friend or two, or an IRC channel, or even some good documentation, your average person can usually install Linux without too much trouble.
I have yet to see a review that even comments on upgrades. What happens when I want to upgrade my FooLinux box from version 4 to version 5? What is involved? How stable is the upgrade? What are the chances of it keeping my box down for an extended period of time? Does it require a reboot? Multiple reboots?
Not only that, but how about some reviewing of their utilities? Do they use some sort of package format? How compatible are the packages with other distro's using that same package format? How good are their package management utilties? Administrative utilities?
Perhaps someday someone will get it right.
Better yet, reduce the number of passwords that you have to remember.
I've found that about 85% or more of the passwords I need to remember are login passwords. So, in an effort to cut that down, I began using the RSA authentication available with Secure Shell. This lets me use the same password (passphrase, actually) to access all of them, while also allowing me a very quick and easy way of changing my password, and the increased security that comes with requiring my private ssh identity along with my password.
Check out the following site for full info on ISO images:
http://cdimage.debian.org/
1.) Does Debian have any system install automation like Redhat's kickstart?
;-)
/etc/apt/sources.list to point to the potato directories, and I type 'apt-get update; apt-get dist-upgrade'. Then, I sit back and wait for it to tell me that it's finished. That's it. No rebooting, no hassle, just sweet and worry free. ;-)
This I can't answer for sure, because I haven't had occasion to try it, but I'm pretty sure I remember hearing about this capability for Debian.
2.) How good is the Debian package management?
To put it simply, Debian's package management is the best available anywhere. It will retrieve packages for you, handle dependancies, handle updates, and overall, just work amazingly. It basically does it all.
For example, if I had a Red Hat 5.2 box, and I wanted to upgrade it to Red Hat 6.0, the most common upgrade path is to insert the 6.0 CD, reboot my computer, and tell it to upgrade the current install.
To upgrade a Debian 2.1 system to Debian 2.2, I change a few lines in
I would like to use it with the 13 boxes as well...Right now if I need to update something on the boxes I make an rpm, put it in a certain directory, and it installs itself that night. Would something like that be possible in debian?
Yep, it's very possible, and very easy. You can either take the program you want to install and make a Debian package out of it (no more difficult than making an RPM, and many consider it much easier), or you can take a current Debian package for that program, get the source package, change it however you want to, build a new Debian binary (and/or source package), and 'dpkg -i ' to install it. Or, you can even set up your own custom Package listing on a local ftp/NFS server, and just 'apt-get install ' to have it download and install the program for you.
To give you one last example of how well Debian's package management is, I know someone who set things up so that when he attempted to upgrade a single package, Debian's package utility had to go out, get all the dependancies, plus those things the dependancies depended on, etc, and install it all in the proper order. He ended up having 457 packages updated, all at once, all with that one command. And it went without a single hitch or problem anywhere.
As a user of potato for a long time now, I've gotta say I think this is a great thing. I still have never seen an Operating System (or distro) as easy to administer, maintain, and upgrade, than a Debian box.
;-)
On another note, I noticed after a recent update that there is currently around 4500 software packages in the Debian repository. I wonder, is Debian now the largest linux Distro? Perhaps Linux support for DVD is coming none-too-soon, we might need it for Debian.
It's very good, and then some. And I say this from experience, not just as a guess.
My school, the University of Nebraska at Omaha (UNO), was one of the first schools in the country to implement something like this, when we recently conducted our school government elections via online voting.
UNO was lucky, in that we already had much of the infrastructure in place. For the past 3 semesters, we've been able to register for classes online, check grades online, and for the past 2 semesters, pay for tuition via a credit card, online.
The authentication used for voting is the same as that required for the other online services. You have to login to a secure server using your student ID (Social Security Number) and your special 6 digit PIN number. This ensures that only registered students can vote, and that nobody can vote multiple times.
Another major advantage to online voting is the ability to distribute additional information. For example, we had biographies for all of the candidates online, so they could be viewed when voting. This helped with one of the most common voter complaints, lack of knowledge about the candidates and what they stand for.
We managed a 10% increase in voter turnout for the initial election, and an amazing 20% increase over that two weeks later in the Student President/Regent run-off election.
I was hoping to be able to provide links to more information about how we implemented things, but I'm afraid I can't find them at the moment. However, if anyone is interested in getting in touch with the people who handled the online voting, drop me an e-mail at topher@tconl.com. This really is a *good* thing, and well worth investigating further if you are considering it.
Does anyone know of a web site that has a list of browsers and what graphical formats they support?
I'd personally love to get rid of all my gifs, but while I prefer pngs to jpegs when I can, I've worked really hard to make my site universally viewable. I'd don't intend to give that up.
I guess I have to wonder whether we'd be better of converting most things to jpegs and waiting a little while longer befor going to pngs...
For those of you who enjoy good Anime, this is *definitely* worth checking out.
This is the man who created my all time favorite Anime, Kaze no Tani no Nausicaä (Nausicaa of the Valley of Wind). Not only is this a full length story (weighing in at right about 2 hours...I hate 45 minute 'movies' with no time for a real plot) but the depth, the characters, and the writing are all abosolutely amazing.
It's strange, but very few people in the US seem to be familiar with Nausicaa, despite the fact that I'm told it almost always places in the top 5 of every Japanese poll about the greatest Anime ever. I suppose part of the reason may be that no Enlglish dub was ever made (Technically, there was an English dub, it was called Warriors of the Wind. However, I refuse to count it, because it was so horribly chopped, mangled, and destroyed, it's not even worth watching. They cut out almost half an hour in the dubbing and editing, essentially everything that was important for the plot, leaving you with a few disjointed action scenes.)
If you'd like more information on Hayao Miyazaki and his Animes and Mangas, check out some of the following URLs:
http://www.nausicaa.net/
http://www.princess-mononoke.com/
http://www.ntv.co.jp/ghibli/
Find out for yourself why the old stigma 'cartoons are for kids' only goes in the US. This will show you how amazing an indepth a 'cartoon' can be.
There used to be three kinds of lies: Lies, Damn Lies, and Statistics. Now, there's a fourth type: Half-Truths.
To someone with little or no experience using and administering Windows NT or Linux, these may be believable. However, anyone who is intimate with
multiple operating systems will no doubt have numerous questions afterreading this.
While I will admit, there are times where the Microsoft article makes valid points, most of it seems to be half truths, misinformation, myths, rumors, speculation, and opinion.
I don't have the time or interest to pick appart every single problem with this article, however, I'll give an example from the beginning of it. The Microsoft article starts with,
"First, it's worth noting that Linux is a UNIX-like operating system. Linux fundamentally relies on 30-year-old operating system technology and architecture."
In a strictly technical sense, this is in fact a true statement. However, it is only half true, if even that. It ignores the fact that through those 30 years of use, Unix has evolved and learned from the mistakes that it has made. The Unix that was originally created at Bell Labs is a far site from the modern Unix operating systems (including the Unix-like Operating Systems, such as Linux).
Linux has managed to learn from all that experience, while at the same time adapting to modern day usage. The fact that many of the ideas and theories that Linux/Unix is based are 30 years old is not a bad thing. It's a good thing.
This argument is in fact not even an NT vs. Unix argument. What it is, is an argument of Operating System theory and oppinion. The problem is, most of it is oppinion. It can't be proven. And much of what can be proven can be proven both ways. As Linus once said, "Microkernels might be a good idea, in theory. However, we aren't building a theoretical kernel. We're building a practical one that works." (These aren't exact words, but I think this captures the jist of it.)
However, you won't find that information in the article. If you want a lot of half truths, misleading statistics, and arguments of theory and philosophy, this is your article. If you want cold hard facts, then what are you wasting your time with articles for? You need to be talking to the people who work with these systems, or work with them yourself, and find out the truth first hand.
Lastly, remember that there are a thousand thousand uses for a computer out there. No single computer, or single OS, can handle all of them.
There is actually a very good reason that Linux has not traditionally been used on IRC servers. However, it is possible the reason is outdated.
;-)
The reason has to do with IRC servers needing one file descriptor for every user connecting to the server (along with a few others for miscellaneous purposes). Up until the 2.2.x kernel, Linux required extensive patching to handle more than 256 fd's per process. Not only that, but these patches were not well tested, nor were they great patches. While they allowed for additional file descriptors, they opened up numerous problems in that fact that the rest of the kernel wasn't ready to handle that many, especially not efficently.
I've followed and been involved with DALnet since late '95, and that's the reason there have been few (Note, 'few'. There have been some.) Linux IRC servers. Under low load (Read: When using a number of file descriptors that doesn't require the kernel to be patched and hacked) it's worked just fine. Under heavier load, other OS's (specifically FreeBSD, though NetBSD has occasionally made a strong showing as well) has been the OS of choice, simply because it has handled the file descriptor problem (along with a few other minor things) better than Linux.
Personally, I'd be very interested in seeing some testing done with the new 2.2.x kernel series. I'd have to guess it would manage *much* better, as I believe most of the patches to increase the number of file descriptors has been integrated and tested and fixed, so it's working properly.
To recap, running an IRC server with greater than 256 users is a Bad Idea, if you are using Linux and running a kernel prior to 2.2.x. Running an IRC server with up to around 1k users should be pretty good to go with a 2.2.x kernel. Above that, I don't feel qualified to say, as I've not tested it myself, yet.
If anyone has done any real testing with ircd on Linux 2.2.x kernels, I'd be very interested in hearing the results.
I'd have to guess they just hadn't gotten to the one that you checked.
I happen to be a certifiable 'nobody' and I got this e-mail to. However, as yet, I've been unable to get to Network Solution's site, the connection is timing out.
Looks like they're prolly getting nailed with a less than pleasant response. Well deserved, however. When I signed up for a domain name, I *never* opted in to recieve any spam, advertisements, or security hole ridden web based e-mail crap.
I used to like InterNIC and Network Solutions, of course, that was a few years ago now. Back before they were losing their monopoly and acting like spoiled children about it.
What a shame.
This really looks like cool stuff, and a neat project.
'Course, as anyone else who only uses Windows for games, anything that advances gaming on Linux is a very Good Thing (tm).
I can't help but think a lot of people could use this to create some really amazing and fun things.
Public keys are a part of PGP(or GPG). The way it works, when you use PGP you first generate a keypair. This consists of a private key, and a public key.
The private key you keep for yourself, and don't allow anyone else access to. This is what you use when signing something, or when decrypting something that is encrypted with your public key.
Your public key you can post on a website, publish to a keyserver, or even send via e-mail. This is what is used by other people to encrypt things. Something encrypted to your public key can only be decrypted by your private key.
I know this is a really basic explanation, but for information, check out http://www.pgp.com, http://www.gnupg.org, or do a search on your favorite search engine for PGP or public key cryptology.
Anyone else notice the date on the ZDnet article that was linked as 'other articles'? Look close and check it out:
;-)
National Semiconductor unveils 'PC on a Chip'
April 6, 1998 9:20 AM PDT
Interesting.
MOSIX has actually been posted to slashdot before, and was the subject of a rather intense discussion.
They're project is a kernel module that requires kernel modifications, however, according to Linus's decision, this means they have to release the whole thing under the GPL if they decide to distribute the module. Last I heard, they hadn't decided what they were going to do about that, yet, as they wanted (needed?) to distrbitute it as binary only.
Couldn't you accoplish this, or pretty close, by using Telnet and a menu driven restricted shell?
;-)
Just give them access to a curses based file manager, and a text editor, or something similar? Just a thought.
I'm getting more and more pumped for this the closer the program comes to completion.
I can't wait until I have a small, fast, clean, lean web browser that still complies with most standards and offers the most important features.
It's great that we're getting back to what counts, losing some of the feature bloat and returning to a really good program.
Great job Mozilla team, keep up the excellent work!
The original poster was trying to make a point that seems to have gotten missed. You see, the original poster was saying that Free Unices are in fact capable of handling extremelly heavy load. FreeBSD just happens to be a free unix variant. Thus, this provides an example to saying, 'Free Unices can handle really heavy load.'
Everyone seems intent on picking them appart and treating them all as totally and completely seperate entities. However, that wasn't how the original post treated them.
If you put product "A" to a certain test, this proves absolutely nothing about product "B".
Perhaps not, but that isn't the comparison that was made. A much more apt comparison would be something like this:
Now, replace A with Linux, B with FreeBSD, and C with NetBSD. Now replace Foo with Free Unices. Perhaps this will clear up the intended point of the original poster?
It's now stable as a rock on Alpha's too :)
;-)
Is it? Neat.
I've heard lots of mixed reports on how far the Alpha port had progressed, though last I heard it was still fairly beta, but improving rapidly. The Sparc port though, last I heard, was pre-alpha still...
UPS Ground runs: $9.11
USPS Priority runs: $7.79
Those appear to be the cheapest.
I love Linux as much or more than the next guy, and NetBSD sounds pretty cool, but how the heck does this record prove anything conclusively about NetBSD and Linux?
You misunderstand my point. I recently suggested at work that we use one of the various free Unices for a couple of servers. My suggestion was shot down, with the comment that none of the free Unices had ever been proven in a high stress, high load situation.
This is, in my oppinion, quite clearly an example where one of them has.
I love the free Unices. FreeBSD is stable as a rock on Intel hardware (though, unfortunately, not portable for crap yet). NetBSD has the stability of FreeBSD, along with the ability to run on damn near every single architecture available (even more than Linux). Linux just plain rocks, with it's stability, features, and amazingly fast evolution.
I also believe in using the right tool for the right task, and I often don't bother to differentiate between which one is better, or anything else. They're all Free Unices to me, each with their own strength and weeknesses.
This says a lot about FreeBSD, and the potential for Free Software in general. Don't make more out of it than there is to be made, though.
This is exactly my point. This is an example that shows very clearly that the Free Unices, and Free Software in general, *can* work, and *does* work. I'm not here to argue the specifics of each OS, or anything like that.
I look at this from the point of view that, when I show my boss evidence like this, all of the Free Unices win, and all of them become better recognised for their abilities by him.
It makes Free Software/Open Source advocates look intellectually dishonest.
I disagree. I see it as using a single example to prove a concept, as opposed to 'My OS is better than your OS.'
Anyone who ever claims that the free Unices aren't up to handling heavy load ought to see this.
;-)
I think this proves very conclusively that the free Unices (Linux, NetBSD, FreeBSD, etc) are all very capable, stable, powerful, and robust. I'd love to see a box running a commericial OS try to match this.