If you are looking for a complete distribution where tight control is kept, you want Slackware. All the applications that come with it are tested, and it does have a great deal of applications. Also, there is active support for it by people who go and package up little tarballs for it (Slackware uses tar.gz format, so no evil RPMs around).
Debian is looser, as it's backed by a consortium:-) (I've always wanted to say that), and seems to have fairly nice support.
Red Hat, et all, is for people who like buggy Linux. :-P
Slackware-current on ftp.cdrom.com is the first Slackware to do Glibc (4.0 was 2.2 Kern + Libc5), and it's as stable in its current state as Red Hat ever was. Of course, I'll use my hybrid Slackware 3.6/4.0 system as my server for a long time before doing any ugprades on that, but the glibc Slackware is awsome for workstations. VMware with Win95OSR2 in it completes my working experience.
If you have an int overflow, it just resets the int to the lower/higher bytes, depending on endiannes.
A string is different. A string is a HUMAN data object, not a computer one. Having a fixed-length is practically required because, you guessed it, computers can't encode this data in a really good way (Pascal's int then string array idea is fine, until the string array is > an int).
The C language was designed to be extensible, and lightweight; Linux is like this, too. Why should we have dynamic strings in the language? If you want dynamic strings, write a C++ class or a C function library to deal with such. It can be done.
But things like FTP commands, and directorys are naturally of a fixed string length. Why use a dynamic string to work with it?
On top of that, how can you get data, then malloc the data? You'd have to redesign the function to read into a buffer of some kind (an infinite buffer? Oooh, maybe a fixed-length one), and then copy to a buffer of appropriate length. You'd also have to grow (realloc) the buffer as needed. This is a lot of work that is not needed 99% of the time when dealing with strings. I don't need 1%er language functionality, that just adds bloat and slows code.
Besides, why should I have dynamic string when the content is static? Waste of CPU time, et all.
I hate people who come from basic/pascal, and then complain because they think of strings in rather a different light then normal C prorgammers do. Please, if you use a language, think in that language when working with it.
Here's a little problem I've noticed (including relating to the recent ProFTPD root exploit).
People think they can get away with strcpy, or sprintf, or similar. This is wrong. You should ALWAYS verify the amount of data copied, wether it be to a fixed sized buffer, or a malloced region.
strncpy, and snprintf are very, very good ways to secure your code from the start.
But this is often disregarded! Agh! Pascal and Basic make people soft about how they handle strings, because they encode length in them and use their buffers in a way that seems logical at first, but is very holey when it comes to actually implementing things.
Strings in C != hard, if you can accept the ideas of pointers, string library functions (I like the abstraction), and general good coding techniques:-)
>Most probably this will generate a thread of (but MY system has been up for XXX days). Well "that don't impress me much" unless it's > 5 years.
Well, what can I say.
IIRC, 5 years ago Linux was 1.2.0-ishly new. Maybe 1.0 was still the stable tree. Eitherway, you'd be NUTS to run that. Once we have a super-good kernel (2.2 is coolio, 2.4 will be very nice for this. Perahps 3.0 will be The One for this), then we can stick it in a corner running a super important job forever...
The problem is that in 5 years, much much much has happened. While it may be cool to be able to upgrade stuff modularly, you do have to change the kernel. Even if we manage to make the entire networking/IP stack modular to the point where we can upgrade it while it's on, it'd probably be a gross hack. You do have to replace/upgrade the root filesystem disk *sometime*:-)
Besides, I like using redundant clusters. Who needs uptime if you can put 5 systems on the project? You should check out the Shuttle's cluster design.
First, thank you for not writing a blatant flame/troll. Another thank you for letting me exploit my reference books on Software Design to share this post with you:-) (I'll be drawing from "The C++ Programming Language, third Editior" 1997, Bjarne Stroustrup)
You say things were "done right the first time." That may very well be true, but the thing I've found has always been best in terms of software development is openness. Share ideas, share content, think think think, and then do it all over. The design, implementation, and organization of software is a very critical factor as to how well the program is, or how it will be perceived.
I think Linus has struck a good balance with the way Linux is currently managed. Overall design and structure are the two major ways a program can go bad. If the overall design is lacking, you get corner cutting ("we'll just ship this one, and fix the problem in the next release"). If the structure is overempahised, you get a situations where programs are delayed by constant program reorganizations ("but this new structure is much better than the old one; people will want to wait for it.").
From what I have read here, which is by no means definitive, some BSD users think the progress is too slow, and that to get the benefits, they must run the current tree. That is bad. BSD is a bit too much on structure for my taste.
Microsoft, of course (must mock them), seems to have *both* problems, at the same time. How so? The various subsystems (like input, video output, etc) seem to have been "done enough to ship, we'll fix the problems later," whereas with things like the NT kernel change over, etc, it seems to be a "time to redo this and that and this," but this is mainly because it wasn't done right the first time, and they've been just bolting new things on to an inadaquate base.
Linux, of course, has struck a nice balance (MY OPINION, NOT NECCESARILY YOURS). Each stable tree doesn't change things too too much (as is proper), ie: 2.0.0 kernel and 2.0.38 kernel run same programs (or should), unless overwise stopped by a bug. The development series is where people get to throw around ideas, and hack things up. This is a good way to get development, but it is tempered by the fact that improvements go to maintainers, go to the overall sub-tree maintainer, go to someone like Alan Cox, and then hit Linus, who can just read the diff, see how it works in with his unifiying vision of the kernel, and provide feedback.
Good design of any software involves the flow of ideas back and forth between the design and implementation. Companies like Microsoft seem to use the "waterfall" implementation, where the designers make rigid specs to send to the mere "coders," who must then slot their code into the well defined slots. However, their slots are changed over time, and the programmers can't really give feedback back to the designers without a lot of hassle. Eventually, they just start over and let the old project slam into the ground, stopped by its own uselessness and cruft. They are actively trying to ditch the Win9x kernel because of this (yay), and use the NT kernel (starting to accumulate cruft). They also have WinCE (new stuff is usually OK, but I don't trust MS anymore).
FreeBSD is good software. It runs stable servers, and seems to be doing well for itself. But I like the design process of Linux, and I really like reading the Alan Cox diaries;-)
I had Slackwaree 3.6 on my system here. How did I upgrade?
Simple, I went to kernelnotes.org, read "the wonderful world of 2.2," and checked what software upgrades would be required. Except for some shell utils (which were rather old, but upgraded easily), my Slackware system had software that was either up to date, or past the min requirements. Installed the new kernel in one day (and I was a lot less experienced with Linux than I am now).
I'm assuming a FreeBSD upgrade should be the same. The kernel should be modular enough you can just drop in a new one.
I'd love to use OpenBSD here sometime on an internal machine to learn about BSDes in general, and OpenBSD specifically, but Linux is my love. I develop for it, and enjoy it.
The GNU Hurd, of course, is another thing I'd love to play with and learn about, and I hope that the way some of you BSD people seem to turn into ravening flamers whenever an [opposing BSD camp | Linux] is mentioned, doesn't also affect that line of development.
We're all friends here. I like Unixen systems, you like Unixen systens. BSD is based on Berkely hacker code, Linux is based on Finnish hacker code, and GNU Hurd is based on that crazy Mach thing that Steve Jobs is somehow associated with;-)
>10) FreeBSD is faster Not on my hardware that it doesn't support.
>9) FreeBSD is more secure 1 problem on Bugtraq for my current Slackware Linux server VS several for FreeBSD (OpenBSD is cool, though. I think the philosophy, and fact it is based in my home town just rule).
>8) FreeBSD has better networking Better at what networking now? TCP/IP, Apple Talk, IPv6, firewalling, etc, etc? Opinions don't count as objectivity...
>7) FreeBSD has a better written kernel Which is, of course, an opinion statement.
>6) FreeBSD runs more major Web Sites And can continue to do so, away from my hardware.
>5) FreeBSD is more efficient At making zealots who hurt the freesoftware movement. You aren't all bad, but the post this in reply to is.
>4) FreeBSD is free from such useless efforts as WINE and Gnome. Cool, I wrote an emulator for SNES! Since that one guy said an emulator for Windows functionality was useless, I won't port this there, either. No software is not good software.
As for GNOME, why on earth wouldn't you want a cool GUI to work in? GIMP for console doesn't work too well, budski.
>3) FreeBSD is more stable Opinion statement #3. My Slackware Linux server is as stable as the power company will let it be. This is why I am buying a UPS:-)
>2) FreeBSD is BSD based Humans are carbon based. There are a lot of asshole humans. Go figure.
>1) FreeBSD has not been contaimated by the billionaires at the cathedral RedHat. Er, Slackware Linux is not RedHat. If you're crazy to avoid Linux because of RedHat, I guess I should be crazy and avoid *BSD because of a certain ravening FreeBSD user.
You are REALLY funny, though. Thanks for reminding me that I should be kind, and courtious to all computer users, regardless of their "sacreligous" operating system choices. I might still be a bit curt to Mac users, though;-)
Compiles most places. Wonderful little thing takes regexps (gif2png *.gif).
Now, to make a program that scans through all HTML files replacing gif with PNG... Or, you could just add an Apache URL translator rule and not edit the HTML at all:-)
The spec is finalised, which is cool. Now we just have to wait for some cool hacker to put it in Mozilla. Screw IE, it can't even show PNGs semi-properly, except on Win32 platforms (Mac, etc, ports are b0rked).
Carnegie Mellon is the best university for cool AI projects. MIT is just a school for engineering weenies who like making things with their hands, CMU is where people go to make new versions of Eliza:-)
Seriously, our wicked beastly connections must have killed this server dead-ish..
If those poor souls only had 64mb of ram (horrors), they are doomed.
My little server is a P200 w/ 96mb of ram.. If I knew, or at least had forewarning I'd be Slashdotted, I'd quickly swap cards/HDs with my most powerful machine (K6-2 263 w/ 128mb of ram, although I might get the ram up to 160mb or more;-)..
Methinks they'll also need the wake-on kernel patch to make Apache more efficient if we do this again.
I have a Slackware 2.3 CD from 1995. It's a modified version in a 4CD set (it has the RedHat mother's day release, ewww). The cool thing is that it is using Kernel 1.2.8 and the ELF beta libs (also has kernel sources from first to 1.2.13 and 1.3.15). A nice piece of history.
The libc4 to libc5/ELF transition took over a year. Why did the libc5/glib transistion take longer? Seems it's only been a year and a half. I'm glad of that time because it means a) I can use Slackware 4.9 for a stable 2.2.x/libc5 setup (like my current server), and have the 5.0 (possibly unstable, but good) glib on the desktop.
If you haven't noticed, most versions of the glib to the point in time were't quite good enough, at least when we are talking about uptimes of months or years.
You'll notice that, yes, it's the same author. The first and second post were posted within minutes of each other. Could it be, they do indeed appear to originate from the same person.
Anyway, my evil-bastard-modded Slackware 3.6 (now on Kernel 2.2.9 w/ Andreas patch 3 + devfs, as well as Cyrus, and some Slackware 4.0 packages) server will remain the same, but I am quite eager to get glibc2 on my desktop machine:-)
Wired's Journalist integrity (the lies they say)
on
Wired on Slashdot
·
· Score: 1
"In that sense, the site is no different from the tradition of alternative weekly newspapers published in just about every large American city -- papers that praise a few holy cows and lambaste regular targets. "
Casini was loaded with 75 pounds of plutonium. The only people to raise a ruccuss when NASA lauched it (on a rocket with a 1 in 12 failure rate), were the alternative news media. The "Tradditional" media often either glosses over, for _FAILS_ to cover some very serious subject material. Hell, Rob knew about Casini, and not because of CNN or Wired.
Has anyone here ever heard of Project Censored? It's a wonderful project for journalism majors at a US university in California (can't remember which one). Each year, they pick and review the 10 least published but potentially very important stories. Remember echelon? Reported in alternative news papers. Cancer causing agents in makeup and skins lotions in 10x the amounts they showed up in meats like Bacon int he 1970s (anyone remember that scare?), only alternative news covered that.
Alternative news may seem to have an agenda, but it doesn't. These papers are usually put out by University students, or other people who care about getting the word around about potentially important stories.
Just look up "Project Censored" in AltaVista, it will take you to what you want.
I'm tempted to add cnn.com and wired.com to my Internet Junkbuster filter, or just axe off access with IPchains.
The netmask is a per-computer think. It basically controls the broadcast address your computer will use for that subnet.
Example: 255.255.255.128 means that 192.168.0.1 through 192.168.1.126 will be valid, and fine. But.127 will be the broadcast for that subnet, and it will not see the packets in the 128 - 254 subnet (unless there is a n explicity connection). It has to do with logical hiearchies, etc.
If this man is saying he can use Netmasks as extra address bytes, he has clearly pointed his ass at the computer and spewed forth bullshit.
Class C subnets are Of the form net.net.net.node (and have a netmask of 255.255.255.0).. There are more defined in the applicable RFCs. Like class A, and B.
The problem with IPv4 was that class C was 255 addresses, class B was 65,025 addresses, and class A was 16,581,375 addresess. If your corporate network had more than 65,000 PCs (possible if you had many servers, and happened to be a huge accounting firm), you basically had to take 16 MILLION addresses away from the global pool because that's how IANNA assigns IPv4 numbers.
Ludicrous! But logical, and in fact proper. This is why IPv6 is good. We *can* piss away IP addresses easily:-)
The only thing I can think to say (that hasn't already been said), is that I bet Rob and Jeff can figure out whena Katz posting has been sucessful by the sudden flood of packets alone:-)
It's an interesting thesis statement to say that a minority of some kind is getting the short-shirft in a technological country, because it helps to expose some of the racist tendancies that we, even in our wonderful towers of morality, still have not eliminated. The reason that most people of a certain skin colour seem to be poor can be traced back to the rampant racism and stereo-tying that still existed after Licoln's emancipation proclimation. The recent SouthPark movie illustrated this (enjoyably so, including the awsome Win98 bit). Chef: "Haven't you ever heard of the emancipation proclimation?" General: "We're not communists here!"
African-americans were gradually allowed to have the freedoms guaranteed them by the US consitution, something I completely support. If you say on paper everyone is equal, implement it that way. However, the until the 1980s, when it became finally a bit better for them, it was horrible. Even the 1970s were awful, with may living in squalid conditions. Do this conditions still exist today? Most certainly, yes. I've seen through TV a few of the public housing areas in the US that are, quite frankly, unfit for habitation, human or otherwise.
Now enter Canada. Here we have a proper support infrastructure for a society. Health care, REAL health care is available freely for all. I certainly don't mind a percent of my paycheque going to perhaps fund an operation that will save someone's life if I myself am fine, well fed, and have a nice house to live in.
Welfare? We have that, too. Anyone can apply for it, even "homeless" people. As long as you're a citizen of Canada, and sometimes if you're not, the Canadian Government will help you along for a bit, and try to get you back into society nicely.
From what I have seen of the US, this structure is sorely underdeveloped. Do you think the great economic imbalances would still exist if people didn't have to pay for live-saving operations? Perhaps if they also could be guaranteed enough food to eat, so they can focus on education (through the many public libraries, perhaps). Where there is a will, there is a way.
This is why I do not understand people who cry, "why should I give MY money to lousey deadbeats who can't even go out to find a job?" Have you ever been so hungry, that the thought of movement is painful? Have you ever had a hard time concentrating, or even thinking in a class situation, because you were so hungry feeling? I have. I lived in househould that was supported through welfare most of my childhood. You learn to live off whatever you can. It's not a life I would wish on my most hated enemies.
Today, I am succesfull. Why? I managed to parley a small job into enough money for a computer. From there, my talents snowballed to the point where I have very markettable and lucrative skills at a time when they are most in demand. Good luck? Perhaps, but certainly it would have been much harder, if it would have happened at all, without the proper social infrastructure a society should guarantee its citizens. I propose that the US consitution have a clause that provides live-saving health care for free, a clause that provides a guarantee of at least minimum food and shelter, in a decent way. And most of all, something to let these people live, not as "deadbeats" or "drifters," but as people who can not worry about their lives long enough to find a job, and give back to the community. To be able to serve and help their fellow humankind with dignity, like the rest of us. We are all humans, much more brothers and sisters to each other than the animals many of us seem ready to die to protect.
Note: I'm Canadian. I'm caucasian, 18, and happy where I am, well fed, nicely housed, and educated.
Slashdot Mirror
Slackware is an awsome distribution, so why do you go and FUD saying it doesn't matter?
IT DOES MATTER TO ME!
You hateful person. FUDing away your life...
I hate Debian's SysV init scripts. Give me Slackware's BSD init scripts *ANY* day.
What's up with you having NO colour LS by default!? HORRIBLE!
How can you use a distro that doesn't let you *do* whatever you want? Horrible!!!!
Slackware is for me. No package-systems (except the nice bare-bones tgz) to get in the way, no other things. Just grab code, compile, install, enjoy.
I've learned a lot more about Unixen from Slackware than I'm sure I would've from Debian (I didn't keep it installed for long). And that's not FUD.
If you are looking for a complete distribution where tight control is kept, you want Slackware. All the applications that come with it are tested, and it does have a great deal of applications. Also, there is active support for it by people who go and package up little tarballs for it (Slackware uses tar.gz format, so no evil RPMs around).
:-) (I've always wanted to say that), and seems to have fairly nice support.
Debian is looser, as it's backed by a consortium
Red Hat, et all, is for people who like buggy Linux.
:-P
Slackware-current on ftp.cdrom.com is the first Slackware to do Glibc (4.0 was 2.2 Kern + Libc5), and it's as stable in its current state as Red Hat ever was. Of course, I'll use my hybrid Slackware 3.6/4.0 system as my server for a long time before doing any ugprades on that, but the glibc Slackware is awsome for workstations. VMware with Win95OSR2 in it completes my working experience.
A string is not an int.
If you have an int overflow, it just resets the int to the lower/higher bytes, depending on endiannes.
A string is different. A string is a HUMAN data object, not a computer one. Having a fixed-length is practically required because, you guessed it, computers can't encode this data in a really good way (Pascal's int then string array idea is fine, until the string array is > an int).
The C language was designed to be extensible, and lightweight; Linux is like this, too. Why should we have dynamic strings in the language? If you want dynamic strings, write a C++ class or a C function library to deal with such. It can be done.
But things like FTP commands, and directorys are naturally of a fixed string length. Why use a dynamic string to work with it?
On top of that, how can you get data, then malloc the data? You'd have to redesign the function to read into a buffer of some kind (an infinite buffer? Oooh, maybe a fixed-length one), and then copy to a buffer of appropriate length. You'd also have to grow (realloc) the buffer as needed. This is a lot of work that is not needed 99% of the time when dealing with strings. I don't need 1%er language functionality, that just adds bloat and slows code.
Besides, why should I have dynamic string when the content is static? Waste of CPU time, et all.
I hate people who come from basic/pascal, and then complain because they think of strings in rather a different light then normal C prorgammers do. Please, if you use a language, think in that language when working with it.
Here's a little problem I've noticed (including relating to the recent ProFTPD root exploit).
:-)
People think they can get away with strcpy, or sprintf, or similar. This is wrong. You should ALWAYS verify the amount of data copied, wether it be to a fixed sized buffer, or a malloced region.
strncpy, and snprintf are very, very good ways to secure your code from the start.
But this is often disregarded! Agh! Pascal and Basic make people soft about how they handle strings, because they encode length in them and use their buffers in a way that seems logical at first, but is very holey when it comes to actually implementing things.
Strings in C != hard, if you can accept the ideas of pointers, string library functions (I like the abstraction), and general good coding techniques
In case you didn't know, Daemon was the word for "Angel" in the old testament (yes, obscure thing for an atheist to know).
:-)
;-)
Daemons, thusly, are beignly cool software that communicates between the supreme kernel, and the unworthy user
Since [Chuck]Beastie] is a Daemon, he needs a halo.
Man, penguins make things simpler. "They eat fish and don't mind cold"
No, that's what they see most of when they use *Office*.
If we're looking for a "most seen" thing, then we should use the program crash dialog, or perhaps a nice BlueScreenOfDeath...
:-) Nice to see another Saskatonian, though.
>Most probably this will generate a thread of (but MY system has been up for XXX days). Well "that don't impress me much" unless it's > 5 years.
:-)
Well, what can I say.
IIRC, 5 years ago Linux was 1.2.0-ishly new. Maybe 1.0 was still the stable tree. Eitherway, you'd be NUTS to run that. Once we have a super-good kernel (2.2 is coolio, 2.4 will be very nice for this. Perahps 3.0 will be The One for this), then we can stick it in a corner running a super important job forever...
The problem is that in 5 years, much much much has happened. While it may be cool to be able to upgrade stuff modularly, you do have to change the kernel. Even if we manage to make the entire networking/IP stack modular to the point where we can upgrade it while it's on, it'd probably be a gross hack. You do have to replace/upgrade the root filesystem disk *sometime*
Besides, I like using redundant clusters. Who needs uptime if you can put 5 systems on the project? You should check out the Shuttle's cluster design.
First, thank you for not writing a blatant flame/troll. Another thank you for letting me exploit my reference books on Software Design to share this post with you :-) (I'll be drawing from "The C++ Programming Language, third Editior" 1997, Bjarne Stroustrup)
;-)
You say things were "done right the first time." That may very well be true, but the thing I've found has always been best in terms of software development is openness. Share ideas, share content, think think think, and then do it all over. The design, implementation, and organization of software is a very critical factor as to how well the program is, or how it will be perceived.
I think Linus has struck a good balance with the way Linux is currently managed. Overall design and structure are the two major ways a program can go bad. If the overall design is lacking, you get corner cutting ("we'll just ship this one, and fix the problem in the next release"). If the structure is overempahised, you get a situations where programs are delayed by constant program reorganizations ("but this new structure is much better than the old one; people will want to wait for it.").
From what I have read here, which is by no means definitive, some BSD users think the progress is too slow, and that to get the benefits, they must run the current tree. That is bad. BSD is a bit too much on structure for my taste.
Microsoft, of course (must mock them), seems to have *both* problems, at the same time. How so? The various subsystems (like input, video output, etc) seem to have been "done enough to ship, we'll fix the problems later," whereas with things like the NT kernel change over, etc, it seems to be a "time to redo this and that and this," but this is mainly because it wasn't done right the first time, and they've been just bolting new things on to an inadaquate base.
Linux, of course, has struck a nice balance (MY OPINION, NOT NECCESARILY YOURS). Each stable tree doesn't change things too too much (as is proper), ie: 2.0.0 kernel and 2.0.38 kernel run same programs (or should), unless overwise stopped by a bug. The development series is where people get to throw around ideas, and hack things up. This is a good way to get development, but it is tempered by the fact that improvements go to maintainers, go to the overall sub-tree maintainer, go to someone like Alan Cox, and then hit Linus, who can just read the diff, see how it works in with his unifiying vision of the kernel, and provide feedback.
Good design of any software involves the flow of ideas back and forth between the design and implementation. Companies like Microsoft seem to use the "waterfall" implementation, where the designers make rigid specs to send to the mere "coders," who must then slot their code into the well defined slots. However, their slots are changed over time, and the programmers can't really give feedback back to the designers without a lot of hassle. Eventually, they just start over and let the old project slam into the ground, stopped by its own uselessness and cruft. They are actively trying to ditch the Win9x kernel because of this (yay), and use the NT kernel (starting to accumulate cruft). They also have WinCE (new stuff is usually OK, but I don't trust MS anymore).
FreeBSD is good software. It runs stable servers, and seems to be doing well for itself. But I like the design process of Linux, and I really like reading the Alan Cox diaries
Dear lord, no!
;-)
:-)
I had Slackwaree 3.6 on my system here.
How did I upgrade?
Simple, I went to kernelnotes.org, read "the wonderful world of 2.2," and checked what software upgrades would be required. Except for some shell utils (which were rather old, but upgraded easily), my Slackware system had software that was either up to date, or past the min requirements. Installed the new kernel in one day (and I was a lot less experienced with Linux than I am now).
I'm assuming a FreeBSD upgrade should be the same. The kernel should be modular enough you can just drop in a new one.
I'd love to use OpenBSD here sometime on an internal machine to learn about BSDes in general, and OpenBSD specifically, but Linux is my love. I develop for it, and enjoy it.
The GNU Hurd, of course, is another thing I'd love to play with and learn about, and I hope that the way some of you BSD people seem to turn into ravening flamers whenever an [opposing BSD camp | Linux] is mentioned, doesn't also affect that line of development.
We're all friends here. I like Unixen systems, you like Unixen systens. BSD is based on Berkely hacker code, Linux is based on Finnish hacker code, and GNU Hurd is based on that crazy Mach thing that Steve Jobs is somehow associated with
A quick rebuttal! :-)
:-)
;-)
>10) FreeBSD is faster
Not on my hardware that it doesn't support.
>9) FreeBSD is more secure
1 problem on Bugtraq for my current Slackware Linux server VS several for FreeBSD (OpenBSD is cool, though. I think the philosophy, and fact it is based in my home town just rule).
>8) FreeBSD has better networking
Better at what networking now? TCP/IP, Apple Talk, IPv6, firewalling, etc, etc? Opinions don't count as objectivity...
>7) FreeBSD has a better written kernel
Which is, of course, an opinion statement.
>6) FreeBSD runs more major Web Sites
And can continue to do so, away from my hardware.
>5) FreeBSD is more efficient
At making zealots who hurt the freesoftware movement. You aren't all bad, but the post this in reply to is.
>4) FreeBSD is free from such useless efforts as WINE and Gnome.
Cool, I wrote an emulator for SNES! Since that one guy said an emulator for Windows functionality was useless, I won't port this there, either. No software is not good software.
As for GNOME, why on earth wouldn't you want a cool GUI to work in? GIMP for console doesn't work too well, budski.
>3) FreeBSD is more stable
Opinion statement #3. My Slackware Linux server is as stable as the power company will let it be. This is why I am buying a UPS
>2) FreeBSD is BSD based
Humans are carbon based. There are a lot of asshole humans. Go figure.
>1) FreeBSD has not been contaimated by the billionaires at the cathedral RedHat.
Er, Slackware Linux is not RedHat. If you're crazy to avoid Linux because of RedHat, I guess I should be crazy and avoid *BSD because of a certain ravening FreeBSD user.
You are REALLY funny, though. Thanks for reminding me that I should be kind, and courtious to all computer users, regardless of their "sacreligous" operating system choices. I might still be a bit curt to Mac users, though
Yes, it's gif2png!
/ gif2png-0.6.zip
:-)
ftp://swrinde.nde.swri.edu/pub/png/applications
Compiles most places. Wonderful little thing takes regexps (gif2png *.gif).
Now, to make a program that scans through all HTML files replacing gif with PNG... Or, you could just add an Apache URL translator rule and not edit the HTML at all
You mean some pervert hacked Lynx to show images?!?
I think I'm going to be sick....
http://www.cdrom.com/pub/mng/
/sounds/ cool.
:-) It
The spec is finalised, which is cool. Now we just have to wait for some cool hacker to put it in Mozilla. Screw IE, it can't even show PNGs semi-properly, except on Win32 platforms (Mac, etc, ports are b0rked).
Ahhh, that would rock!
:-)
Carnegie Mellon is the best university for cool AI projects. MIT is just a school for engineering weenies who like making things with their hands, CMU is where people go to make new versions of Eliza
And Cyrus IMAPD, hehee.. IT RULES!!!
They screwed up the design! The penguin on those cards looks slow and stupid! The head is all wrong!
Not to mention the fact you have to be a US citizen (I'm Canadian, can't they mail the damned card across the border?!)
Sigh. Time to go plot a killing spree again.
Seriously, our wicked beastly connections must have killed this server dead-ish..
;-)..
If those poor souls only had 64mb of ram (horrors), they are doomed.
My little server is a P200 w/ 96mb of ram.. If I knew, or at least had forewarning I'd be Slashdotted, I'd quickly swap cards/HDs with my most powerful machine (K6-2 263 w/ 128mb of ram, although I might get the ram up to 160mb or more
Methinks they'll also need the wake-on kernel patch to make Apache more efficient if we do this again.
I have a Slackware 2.3 CD from 1995. It's a modified version in a 4CD set (it has the RedHat mother's day release, ewww). The cool thing is that it is using Kernel 1.2.8 and the ELF beta libs (also has kernel sources from first to 1.2.13 and 1.3.15). A nice piece of history.
The libc4 to libc5/ELF transition took over a year. Why did the libc5/glib transistion take longer? Seems it's only been a year and a half. I'm glad of that time because it means a) I can use Slackware 4.9 for a stable 2.2.x/libc5 setup (like my current server), and have the 5.0 (possibly unstable, but good) glib on the desktop.
If you haven't noticed, most versions of the glib to the point in time were't quite good enough, at least when we are talking about uptimes of months or years.
Jee, XMMS works fine!
:-)
ftp://rasputin.linuxos.net
Look at the Slackware packages. I love it!
You'll notice that, yes, it's the same author. The first and second post were posted within minutes of each other. Could it be, they do indeed appear to originate from the same person.
:-)
Anyway, my evil-bastard-modded Slackware 3.6 (now on Kernel 2.2.9 w/ Andreas patch 3 + devfs, as well as Cyrus, and some Slackware 4.0 packages) server will remain the same, but I am quite eager to get glibc2 on my desktop machine
"In that sense, the site is no different from the tradition of alternative weekly newspapers published in just about every large American city -- papers that praise a few holy cows and lambaste regular targets. "
Casini was loaded with 75 pounds of plutonium. The only people to raise a ruccuss when NASA lauched it (on a rocket with a 1 in 12 failure rate), were the alternative news media. The "Tradditional" media often either glosses over, for _FAILS_ to cover some very serious subject material. Hell, Rob knew about Casini, and not because of CNN or Wired.
Has anyone here ever heard of Project Censored? It's a wonderful project for journalism majors at a US university in California (can't remember which one). Each year, they pick and review the 10 least published but potentially very important stories. Remember echelon? Reported in alternative news papers. Cancer causing agents in makeup and skins lotions in 10x the amounts they showed up in meats like Bacon int he 1970s (anyone remember that scare?), only alternative news covered that.
Alternative news may seem to have an agenda, but it doesn't. These papers are usually put out by University students, or other people who care about getting the word around about potentially important stories.
Just look up "Project Censored" in AltaVista, it will take you to what you want.
I'm tempted to add cnn.com and wired.com to my Internet Junkbuster filter, or just axe off access with IPchains.
Side note, my math was a bit off (I've been up for 20 hours, sue me).
.0 is a network and .255 is a broadcaster.. Forgot to subtract network.
:-)
It's not 255, it's 254 in a Class C..
254 ^ 2 for class B, and 254 ^ 3 for a class A.
Because
Special bonus game: spot more logic flaws I missed
The netmask is a per-computer think. It basically controls the broadcast address your computer will use for that subnet.
.127 will be the broadcast for that subnet, and it will not see the packets in the 128 - 254 subnet (unless there is a n explicity connection). It has to do with logical hiearchies, etc.
:-)
Example: 255.255.255.128
means that 192.168.0.1 through 192.168.1.126 will be valid, and fine. But
If this man is saying he can use Netmasks as extra address bytes, he has clearly pointed his ass at the computer and spewed forth bullshit.
Class C subnets are Of the form net.net.net.node
(and have a netmask of 255.255.255.0)..
There are more defined in the applicable RFCs. Like class A, and B.
The problem with IPv4 was that class C was 255 addresses, class B was 65,025 addresses, and class A was 16,581,375 addresess. If your corporate network had more than 65,000 PCs (possible if you had many servers, and happened to be a huge accounting firm), you basically had to take 16 MILLION addresses away from the global pool because that's how IANNA assigns IPv4 numbers.
Ludicrous! But logical, and in fact proper. This is why IPv6 is good. We *can* piss away IP addresses easily
God-damnit! I mean, WHAT THE HELL!!!
Grammar! THERE IS NO E!!?!?
Do none of you use ispell?!?!
Do none of you even bother now, having become lazy by the "spellcheck on send" features of every usefull mailer?!
AGJ!!!!!!
What the hell is a "clack" ???
Damnit, man. So much for not being entirely non-racist.
The only thing I can think to say (that hasn't already been said), is that I bet Rob and Jeff can figure out whena Katz posting has been sucessful by the sudden flood of packets alone :-)
It's an interesting thesis statement to say that a minority of some kind is getting the short-shirft in a technological country, because it helps to expose some of the racist tendancies that we, even in our wonderful towers of morality, still have not eliminated. The reason that most people of a certain skin colour seem to be poor can be traced back to the rampant racism and stereo-tying that still existed after Licoln's emancipation proclimation. The recent SouthPark movie illustrated this (enjoyably so, including the awsome Win98 bit). Chef: "Haven't you ever heard of the emancipation proclimation?" General: "We're not communists here!"
African-americans were gradually allowed to have the freedoms guaranteed them by the US consitution, something I completely support. If you say on paper everyone is equal, implement it that way. However, the until the 1980s, when it became finally a bit better for them, it was horrible. Even the 1970s were awful, with may living in squalid conditions. Do this conditions still exist today? Most certainly, yes. I've seen through TV a few of the public housing areas in the US that are, quite frankly, unfit for habitation, human or otherwise.
Now enter Canada. Here we have a proper support infrastructure for a society. Health care, REAL health care is available freely for all. I certainly don't mind a percent of my paycheque going to perhaps fund an operation that will save someone's life if I myself am fine, well fed, and have a nice house to live in.
Welfare? We have that, too. Anyone can apply for it, even "homeless" people. As long as you're a citizen of Canada, and sometimes if you're not, the Canadian Government will help you along for a bit, and try to get you back into society nicely.
From what I have seen of the US, this structure is sorely underdeveloped. Do you think the great economic imbalances would still exist if people didn't have to pay for live-saving operations? Perhaps if they also could be guaranteed enough food to eat, so they can focus on education (through the many public libraries, perhaps). Where there is a will, there is a way.
This is why I do not understand people who cry, "why should I give MY money to lousey deadbeats who can't even go out to find a job?" Have you ever been so hungry, that the thought of movement is painful? Have you ever had a hard time concentrating, or even thinking in a class situation, because you were so hungry feeling? I have. I lived in househould that was supported through welfare most of my childhood. You learn to live off whatever you can. It's not a life I would wish on my most hated enemies.
Today, I am succesfull. Why? I managed to parley a small job into enough money for a computer. From there, my talents snowballed to the point where I have very markettable and lucrative skills at a time when they are most in demand. Good luck? Perhaps, but certainly it would have been much harder, if it would have happened at all, without the proper social infrastructure a society should guarantee its citizens. I propose that the US consitution have a clause that provides live-saving health care for free, a clause that provides a guarantee of at least minimum food and shelter, in a decent way. And most of all, something to let these people live, not as "deadbeats" or "drifters," but as people who can not worry about their lives long enough to find a job, and give back to the community. To be able to serve and help their fellow humankind with dignity, like the rest of us. We are all humans, much more brothers and sisters to each other than the animals many of us seem ready to die to protect.
Note: I'm Canadian. I'm caucasian, 18, and happy where I am, well fed, nicely housed, and educated.