The Haliburton contract was put in place by Clinton's administration, long before 9/11 -- you're pointing the finger at the wrong administration with that claim.
9/11 was not a person murdering another person, you frame this as if it were a typical crime we should pursue through our court system. When in reality it was an extremely large entity attacking our country, not just singular persons.
So terrorism is just a crime we pursue in criminal courts? When the movement is as large and concerted as it is, you cannot simply frame it as a typical crime. You may lie to yourself, but the rest of us see (save for the sheep).
The other benefit to IPv6 will be the greatly enhanced routing. As it stands today, so many NSPs and ISPs have routing fubar'd like you wouldn't beleive. Handing off to lame peers, or letting noc monkies who just got their CCNP modify BGP broadcasts, without a clue.
Packet headers may be 40 bytes, but when you're no longer passing through ~15 routers to hit your destination, +24 bytes will be trivial.
Example of XO routing: (I now have a low opinion of them, after ~10 years of using them, beginning with a dial-up UNIX shell in 1994) : Houston, to Dallas, to Atlanta, to Washington DC, to Colorado, to California(LA->SJO no doubt even. hah.). Someone has their routing royally screwed. (Thank god not all transit providers aren't equally stupid)
It's not a matter of "taking off", as it already has in most respects. Sure, your average Joe doesn't an IPv6 connection... yet.
bgp02# sh ipv bgp sum
BGP router identifier x.x.x.x, local AS number xxxxx
385 BGP AS-PATH entries
If you'd look at the IPv6 routing table, you'd see just how many networks are being advertised already. IPv6 won't become widely used until NSP's and ISP's begin selling native links to residential users(target: 2005-2006). Which at this point in time, they have no real reason to do so. If you'd notice, most NSP's are peering up native IPv6 connections, or have some type of native transit. Most people don't purchase transit from them, because there's practically no residential end-users with native connections, thus they'd pay money, yet never use the link. (I personally beleive they should be giving transit away to current customers, in order to try and push IPv6, since in all honesty if you're paying thousands a month for ipv4 transit, why not give ipv6 transit away for free until it's a medium that actually generates traffic!)
We still haven't quite exhausted IPv4 address space, but we will, which was the major reason for IPv6. 128 Bit address space > 32 Bit address space.
You know, I think it'd be wonderful if we all ran around with tiny cloths around our crotches, slaughtered buffalo near extinction, and lived in teepee's today.
Gee, we should give it back, since they're obviously far superior to us. That and they need the land to build more Casinos. Americans are so awful, they just don't care about anyone but themselves. *cough*Bullshit*cough*
Get a clue.
Re:I'll take, FreeBSD release notes for 800, pleas
on
FreeBSD 5.2 Released
·
· Score: 1
FTP imposes a lower threshold connection limit than HTTP.
This is solely based upon individual's configurations.
The excess "overhead" of the FTP protocol introduces trivial amounts of added traffic, especially in today's high bandwidth internet. This is a moot point, and one that should be discussed elsewhere, however I'll atleast respond.
If you don't live under a rock, you know mark-up language adds tags obviously. In turn which adds to the size of the file. The following are the i386 Release notes for 5.2-RELEASE:
58K RELNOTES.HTM
30K RELNOTES.TXT
As you can see, the HTML file is much larger - nearly DOUBLE, which will surely result in MORE traffic. I'm sure there's not 28k of extra over the wire traffic due grabbing this file via FTP as opposed to HTTP. Even if there were say ~30k overhead due to the ftp session, (Which I *know* grabbing this 1 file, would not generate that kind of traffic) then they balance out as equal in the end. (Obviously interpreting hypertext will cause "more" cpu load than opening a plain text file)
I agree with you on a whole that serving content is usually better left to the tender loving care of an HTTP server. That's for many reasons, which aren't all specifically protocol based.
End result, I don't know what you were smoking that lead you to beleive you could only get FreeBSD's release notes via FTP, nor how that would tie into FreeBSD's viability as an Operating System.
Actually, their new scheduler (SCHED_ULE) for example was copied almost verbatim from Linux's O(1) scheduler. Sure, change words around and claim its only vaguely based on Linux, but believe me as someone who has read both sources - its a blatant rip off.
Wrong. Linux ripped the ULE from FreeBSD.Proof HereNOT the other way around.
I'm QUITE sure the code you read on your Linux box is very nearly identical to that of the code in FreeBSD, BECAUSE IT WAS RIPPED FROM FREEBSD BY Francesco Sportolari.
Jeffrey Roberson wrote FreeBSD's ULE, and I'M QUITE sure he'd be rather bothered by your accusations. ULE was designed as an improvement for SMP based systems, yet FreeBSD's scheduler is, and has been 0(1). Perhaps read This sometime. Also ULE is *NOT* the default scheduler, it's still SCHED_4BSD, atleast up until 5.1.(I don't run any 5.2 boxes, my only 5.x boxes are non-production)
Be sure you get your facts straight before you go off making false accusations against FreeBSD, and it's extremely talented and hard-working developers.
I love Linux for what it's accomplished, and I love *BSD for the same reason. My personal preferance is FreeBSD, trying to change that is like trying to tell me that Chocolate Ice Cream is better than Vanilla.
Obviously no one does. I draw my conclusion from seeing their posts, and their logic.
This thread is about arrogance. I was merely saying, arrogance is one obstacle in any possibility of merging code from Linux. (Theoretically, if it could be done.)
There's exceptions to every rule. I think your misread what I'm saying. I'm in no way implying Linux code is evil. I've got a debian box with an uptime that's only been affected by a severly long power outage, it's perfect for what it does.
Well, I see that you took a nice cheapshot at Linux.
No cheapshot, you misconstrue what I'm saying. Again, I'm not saying with any specificity that Linux code is unclean, plenty of parts of Linux code are great. I only say that because of many developer's mindsets, and The BSD Vision and because of the Architectural Guidelines although it's open to interpretation I suppose.
I highly doubt any of the legally copied *BSD code in Linux is just copied and pasted before it goes in
I agree with you on the zealot issue, there's zealots in both camps. The difference I see is, many BSD users have been admins for years, and have ran Linux, or admin'd a machine running it, at some point in time. (Such as myself, I've been a Linux advocate and admin/user since kernel 0.99, albeit that first machine did very little on one of my networks.:) I too have been an Internet junkie for long over a decade, before that, I was a BBS sysop/junkie:p) Whereas a lot of the Linux users, this is their first POSIX, UNIX-like operating system, and they might never have even installed a *BSD, or played with them for more than a few days. (Hardly long enough to derive an educated opinion)
Linus Torvalds: "Oh yeah? Well people compare Linux to BSD constantly. But there can be no comparison. Any time BSD does something better than Linux, we incorporate their code. But our performance wins can't be ported to BSD because the GPL prevents it. The code just flows one-way baby, so there's no real hope that BSD can remain competitive. It's embrace and encumber cats, embrace and encumber."
Reference: SlashNot
Linus blatently stating they rip BSD code at times, which is just fine by me. Yet if *BSD sucks so horridly as the Linux zealots have stated, then why ever rip or merge code from said codebase? Some Linux users certainly need to get their facts straight before shooting their mouths off.
Even if core members could rip Linux code, I highly doubt they'd ever want to. Infact, I'd bet on it. If not for their own arrogance (Not all of them are.), but then for the sanctity, and purity of the codebase.
I say this because:
The BSD Vision seems to reinforce my belief. Just my $0.02.
-mpf
I'll take, FreeBSD release notes for 800, please?
on
FreeBSD 5.2 Released
·
· Score: 0
Release Notes are obtainable via many methods, not just FTP. Neither FTP nor HTTP are inheriently superior to the other for the transfer of files, so your premise is rather skewed. For you to say HTTP is better fitted to serve that specific file is a moot point.
The main FreeBSD repository server is very busy, especially during release time. That's why there's more mirrors for FreeBSD than Michael Jackson has in "neverneverland ranch". So you can download from a mirror that's geographically closest to you, thus you don't go through fubar'd routing and such.
BSD addicts, flame me if you'd like, but I'd rather hear some intelligent reasons why anyone would choose to use a BSD over other faster, cheaper, more stable systems.
I think you need to check your facts again. Did you just actually say "Cheaper"? Oh you make me laugh. (The "faster" and "stable" comments are even funnier.)
Just because you're unable to comprehend how to utilize BSD, doesn't in any way lesser it's value, to anyone but you. All the same, if some Windows user tried to use Linux, and was completely lost, You'd just flame him and tell him to "get a clue" or "rtfm". The person who's too uneducated to utilize it, is the only one who's missing out.
There's more reasons than I can list, I've been running *BSD variants since 1993, and Linux (before any distributions existed) since 0.99. BSD is cleaner, the -core teams usually have brighter developers on them, FreeBSD ports are still number 1, the BSD layout is just far superior IMHO, (I usually tend to run slackware, which is the closest to BSD you can get in a Linux basically) my memory footprint is slightly over 3 MB on most of my servers, with a minimal set of services running, (obviously my mysql, apache, and such servers eat more due to the daemons tieing it up) and the list continues on.
Review my posts, and then tell me I'm trolling.
Refering to BSD users, as "lusers" - absolutely classic. What is this, the third grade? That's why all my Linux buddies say the BSD's are "too hard" for them, right?
Right.
(Which, in itself that statement is laden with humor, to me)
Look at your own trolls first, before you point the finger on us. There's a multitude of Cut and Paste trolling going on, if you'd only pay attention. Am I crying to/.? No, I'm methodically disproving their posts. Don't try and nix free speech, people are free to say what they wish, no matter how oblivious they are to the facts.
I've been running FreeBSD and Linux for over a decade, and I don't bash Linux, especially not to this depth. I've made comments on how it could improve, that's not bashing. I run it. I prefer FreeBSD as a server and desktop, but that's ME. To each his own. Something about running xf86, and having such a tiny OS footprint, is a very beautiful thing.
Trolls are posting absolutely hilarious things, which can't go unchecked. I think the funniest thing I've seen was: "It has no GUI!" There have been so many funny things said by them however, I can't keep track. Some of these Linux zealots have really gone off the deep end, to the point where it's just funny. (Which is the only reason I started participating in these threads, or hopped on/.:) )
It really proves how clueless so many of them are, as most of the time they're talking out their ass, and it shows.
Current Uptimes
I see lots of trolls posting NetCraft stats, so here's yet some more to entertain their lonely lives.
Dead you say? It must be so, look how long these machines have been up, it certainly time for something new. Gander at which OS's dominate the list. This certainly doesn't mean anything, now does it?
For a dead guy, BSD sure has a lot of kick left in them. For those of you that're behind, plans are already being made for the 6.x branch of FreeBSD. That sure sounds like a dying OS, now doesn't it?
I've got no beef with Linux. I run Linux, and multiple BSDs, and have for over a decade. I've got a beef with trolls making false statements and any moderators that bump their post to informative.
*chuckle*
We're far from through.
-mpf
Re:Where did this BSD is dead / dying thing start?
on
FreeBSD 5.2 Released
·
· Score: 1
Honestly, I wish there was some other reason, but it's the most clear and concise explaination.
I don't watch TechTV, (infact, I utterly despise it) yet I remember the offmyserver.com postings about it, and if memory serves, all this seems to have started up around that time? Perhaps I'm mistaken, as I don't frequent/.;)
So to your comment I say, Amen.
This I am aware of, as I've seen this identical posting atleast 4 times in other threads. I just had to go on record and disprove his mindless dribble, if not merely for my own satisfaction.
Experienced users generally favor compiling from ports for the added flexibility and control it provides. Many of us need to tweak numberous compile-time options, which of course a binary package is absolutely useless in this respect.
If you're new to FreeBSD, or you're running off that p90 still, binaries are great, otherwise you'll find building from ports to be rather pleasing. ( Unless, you're building kde or something along those lines, then binary can be advantageous:p )
You must have nothing better to do then troll slashdot, which is fine. It's still more a waste of time then listening to a Captain and Tennille album.
Your link to the SysAdmin story is wonderfully entertaining, but you failed to realize those guys couldn't tune a FreeBSD, or any other *BSD box to save their life. FreeBSD ships with extremely small settings, and requires quite a bit of kernel tweaking via sysctl variables, and loader defaults (or recompile) for proper useage of memory, disk, and network performance on a high load, or high traffic system. (Such as SysAdmin is benchmarking in said test)
Quote from SysAdmin Magazine:
On FreeBSD: Append to/etc/sysctl (or you can use sysctl -w to add these):
kern.maxfiles=65536
kern.maxfilesperproc=32768
The above are their "kernel tweaks", which I must say, is short of absolute hilarity. (All the fbsd admins are probably laughing with me too at these meager attempts to "tweak", by SysAdmin Magazine. This is just short of attempting to put a band-aid over a gash the size of your fist, and expecting the bleeding to stop.)
If you publish technical benchmarks, and state that your company is knowledgable and competent (Which I seriously question here.) atleast have the common courtesy to have someone experienced enough to tweak the server (as any large institution/corp/etc would, or like me, have a default sysctl.conf/loader.rc specifically for large load servers)
Anyone who's ever admin'd a server with 15,000+ concurrent connections knows, a default FreeBSD box will "fall over and die" without tweaking. (More specifically, you'll fill up your mbufs, or exhaust your sockets, long before smashing the kernel to death.) After tweaking, it will easily do it with minimal load (dependant upon which services, obviously) as my servers are living proof of, along with Yahoo, and many, many others. FreeBSD's default kernel values are at very low, "safe" values, which in many areas will lead to much slower, degraded performance from what it is capable. (TCP Window Scaling is extremely low out of the box for just one small, yet very important example, and you'll see further ahead, this will easily lead to deathly slow socket connections, and such.)
The theory for this is, it should lead to a much more stable overall machine, albeit I've never had any of my tweaked machines reboot. Quote from SysAdmin article: FreeBSD performance decreased somewhat when more than 1500 connections were added.
I don't doubt this, as default FreeBSD install will have very minimal settings for max connections, max allocated sockets, max files per process, low fd's, again, low tcp window scaling, and more which for brevity I'll exclude. To say FreeBSD performs worse than Linux, and Solaris, AND, Win... Absolutely laughable. I'm sorry, I've been in the UNIX/Networking field (traversing more medias than I wish to list) for over 12 years, I've run the entire gambit, and still do, so I find that statement highly offensive, because it's plain wrong.
To Sys Admin Magazine: Learn how to tweak a FreeBSD kernel, then run the test. We as the readers have an obligation to expect absolute technical perfection from those publishing these types of articles. Apples to Oranges.
More NetCraft fodder: Seven of the top nine sites run on FreeBSD. The exceptions are Datapipe, which is doing a fine job of promoting the reliability of Windows 2003, and German hosting company komplex.net which runs on Linux.
Posted by mhp at 08:54 AM UTC on Jan 11, 2004
Had you ever contemplated the fact that FreeBSD, nor any other *BSD's success does NOT rely on how many people use it? We really couldn't care less. We'd love more userbase, and more contributors, who wouldn't? The fact that we have less users than Windows, or Linux, or OS X, is totally irrelevant to us.
I wasn't talking about "building" via that method now was I?:) I was just trying to clear up what he had said, since the FreeBSD ports do not install binary packages unless explicitly told to, he was stating to the contrary.
No, You're refering to installing a PACKAGE. FreeBSD PACKAGES are binary forms of the application. I.E.: # pkg_add -r lynx.
This will download and install the binary package of lynx, thus no compiling necessary.
FreeBSD ports, however are built from source.
If you: # cd/usr/ports/www/lynx && make install clean
You're downloading lynx's source tarball and it's dependancies, then compiling it from said source(s) via ports.
Again, onlypackages are binary distributions of a program.
Losing JKH from -core was to say the least, unsavory, yet you think FreeBSD is dependant on 1 or 2 members of the -core team? I mean... come on, seriously? If FreeBSD is "dead" because of a 'decline in market share' or due to losing a member or two of the core team, you're just sadly mistaken. To play devil's advocate: Let's assume you're correct, and we lose half our user base. What a pity!/*sarcasm*/ Regardless, we'll continue to hack away, we really don't care - and our success is NOT measured by how many use our "Product." If that were the case around the board, Linux might as well be refered to as "dead", since Microsoft reigns supreme, still. I quote:
Seven of the top nine sites run on FreeBSD. The exceptions are Datapipe, which is doing a fine job of promoting the reliability of Windows 2003, and German hosting company komplex.net which runs on Linux.
Posted by mhp at 08:54 AM UTC on Jan 11, 2004 in Hosting, Performance.
Reference: news.netcraft.com
FreeBSD seems to have been whooping ass in the hosting field for a while now according to NetCraft surveys. (Long before this article's 6 month window as well.) .. I think I might know why *BSD trolling has seen a sharp increase in the last 6+ months or so, Hmm, gee..
This NetCraft article proves nothing about say, Linux vs. FreeBSD or FreeBSD vs. Win2k3, to me it says, FreeBSD isn't dying off, and if it is, it sure is a pain riddled, horrifcally slow death. (if less people are running FreeBSD, good, less morons to have to help , I'm sick of being asked how to update the locate database after all these years..)
Before you put words in JKH's mouth: See JKH's resignation
Why did Apple hire JKH? Not because he looks good in a push-up bra, I can promise you that.
If JKH thinks FreeBSD is no longer viable code, why does Apple continue to develop a branch off FreeBSD, and why is Panther based from FreeBSD 5.x? OS X Panther(features column)
FreeBSD isn't a "clique" of elites, we didn't all install FreeBSD because it's "Cool". So many Linux zealots have called FreeBSD users out, saying we're trying to "fit in." Which is ironic, since many of them installed it because their buddies said, "Oh l00k, itz 31337 d00d."
I've run Linux since kernel 0.99, there was a time I prefered Linux(0.99 - 2.0.34) over any BSD, those days are long passed. However, I DO NOT look down my nose at Linux users, and I certainly don't troll any Linux "forums", yet why is this occuring vice versa? It's unfortunate, the Linux community is also very plagued by the '14 year old zit-faced kids who just installed RedHat, and want to "hax teh g1bs0n."' (I've taken enough abuse from the Linux crowd, talking about "Fat, unwashed, Star Trek watching *BSD nerds" to justify that last comment.) Which is one reason I don't ever want them crossing over to the FreeBSD crowd, thus if we have less 'market share', in that respect it's a VERY GOOD THING(tm).
Many of my sysadmin friends run purely Linux based networks(server-side), and when I ask why they don't even try a *BSD, the common answer I've received is, "It's too hard." Which in turn makes me chuckle to myself because they're really missing out on something great. I don't make fun of them and/or their networks, nor try to say they're running an antiquated or conviluted system. (To say FreeBSD is either of these, is to deny the facts.) To each his own, and if they choose to live in ignorance, MORE POWER TO THEM, it's their prerogative.
We were around in 1992, and we'll continue to be, regardless what the slashdot trolls attentively (or not so attentively..) splurge in digital form.
-mpf
P.S. I wonder how long Yahoo!(tm) will run a dead operating system (FreeBSD) across it's server pools?
( Next time don't post out of date,
Slashdot Mirror
The Haliburton contract was put in place by Clinton's administration, long before 9/11 -- you're pointing the finger at the wrong administration with that claim.
Conspiracy nuts create FUD. As for Watergate, even a stopped clock is right twice a day.
9/11 was not a person murdering another person, you frame this as if it were a typical crime we should pursue through our court system. When in reality it was an extremely large entity attacking our country, not just singular persons. So terrorism is just a crime we pursue in criminal courts? When the movement is as large and concerted as it is, you cannot simply frame it as a typical crime. You may lie to yourself, but the rest of us see (save for the sheep).
The conspiracy nuts are worse than right-wing christian zealots and left-wing socialist morons put together.
The other benefit to IPv6 will be the greatly enhanced routing. As it stands today, so many NSPs and ISPs have routing fubar'd like you wouldn't beleive. Handing off to lame peers, or letting noc monkies who just got their CCNP modify BGP broadcasts, without a clue.
Packet headers may be 40 bytes, but when you're no longer passing through ~15 routers to hit your destination, +24 bytes will be trivial.
Example of XO routing: (I now have a low opinion of them, after ~10 years of using them, beginning with a dial-up UNIX shell in 1994) : Houston, to Dallas, to Atlanta, to Washington DC, to Colorado, to California(LA->SJO no doubt even. hah.). Someone has their routing royally screwed. (Thank god not all transit providers aren't equally stupid)
-mpf
If there's one thing I cannot stand, it's hearing "Americans stole america from the Indians." Stop the absurdity, please.
Think with your brain, not the organ in your chest.
-mpf
It's not a matter of "taking off", as it already has in most respects. Sure, your average Joe doesn't an IPv6 connection... yet.
bgp02# sh ipv bgp sum
BGP router identifier x.x.x.x, local AS number xxxxx
385 BGP AS-PATH entries
If you'd look at the IPv6 routing table, you'd see just how many networks are being advertised already. IPv6 won't become widely used until NSP's and ISP's begin selling native links to residential users(target: 2005-2006). Which at this point in time, they have no real reason to do so. If you'd notice, most NSP's are peering up native IPv6 connections, or have some type of native transit. Most people don't purchase transit from them, because there's practically no residential end-users with native connections, thus they'd pay money, yet never use the link. (I personally beleive they should be giving transit away to current customers, in order to try and push IPv6, since in all honesty if you're paying thousands a month for ipv4 transit, why not give ipv6 transit away for free until it's a medium that actually generates traffic!)
We still haven't quite exhausted IPv4 address space, but we will, which was the major reason for IPv6. 128 Bit address space > 32 Bit address space.
-mpf
Whos to say it was ever theirs?
You know, I think it'd be wonderful if we all ran around with tiny cloths around our crotches, slaughtered buffalo near extinction, and lived in teepee's today.
Gee, we should give it back, since they're obviously far superior to us. That and they need the land to build more Casinos.
Americans are so awful, they just don't care about anyone but themselves. *cough*Bullshit*cough*
Get a clue.
FTP imposes a lower threshold connection limit than HTTP.
This is solely based upon individual's configurations.
The excess "overhead" of the FTP protocol introduces trivial amounts of added traffic, especially in today's high bandwidth internet. This is a moot point, and one that should be discussed elsewhere, however I'll atleast respond.
If you don't live under a rock, you know mark-up language adds tags obviously. In turn which adds to the size of the file. The following are the i386 Release notes for 5.2-RELEASE:
58K RELNOTES.HTM
30K RELNOTES.TXT
As you can see, the HTML file is much larger - nearly DOUBLE, which will surely result in MORE traffic. I'm sure there's not 28k of extra over the wire traffic due grabbing this file via FTP as opposed to HTTP. Even if there were say ~30k overhead due to the ftp session, (Which I *know* grabbing this 1 file, would not generate that kind of traffic) then they balance out as equal in the end. (Obviously interpreting hypertext will cause "more" cpu load than opening a plain text file)
I agree with you on a whole that serving content is usually better left to the tender loving care of an HTTP server. That's for many reasons, which aren't all specifically protocol based.
End result, I don't know what you were smoking that lead you to beleive you could only get FreeBSD's release notes via FTP, nor how that would tie into FreeBSD's viability as an Operating System.
-mpf
Actually, their new scheduler (SCHED_ULE) for example was copied almost verbatim from Linux's O(1) scheduler. Sure, change words around and claim its only vaguely based on Linux, but believe me as someone who has read both sources - its a blatant rip off.
:)
Wrong. Linux ripped the ULE from FreeBSD. Proof Here NOT the other way around.
I'm QUITE sure the code you read on your Linux box is very nearly identical to that of the code in FreeBSD, BECAUSE IT WAS RIPPED FROM FREEBSD BY Francesco Sportolari.
Jeffrey Roberson wrote FreeBSD's ULE, and I'M QUITE sure he'd be rather bothered by your accusations. ULE was designed as an improvement for SMP based systems, yet FreeBSD's scheduler is, and has been 0(1). Perhaps read This sometime. Also ULE is *NOT* the default scheduler, it's still SCHED_4BSD, atleast up until 5.1.(I don't run any 5.2 boxes, my only 5.x boxes are non-production)
Be sure you get your facts straight before you go off making false accusations against FreeBSD, and it's extremely talented and hard-working developers.
I love Linux for what it's accomplished, and I love *BSD for the same reason.
My personal preferance is FreeBSD, trying to change that is like trying to tell me that Chocolate Ice Cream is better than Vanilla.
-mpf
and quit marking me off-topic, I'm not!
You speak for all of *BSD developers?
;)
Obviously no one does. I draw my conclusion from seeing their posts, and their logic.
This thread is about arrogance.
I was merely saying, arrogance is one obstacle in any possibility of merging code from Linux. (Theoretically, if it could be done.) There's exceptions to every rule. I think your misread what I'm saying.
I'm in no way implying Linux code is evil. I've got a debian box with an uptime that's only been affected by a severly long power outage, it's perfect for what it does.
Well, I see that you took a nice cheapshot at Linux.
No cheapshot, you misconstrue what I'm saying. Again, I'm not saying with any specificity that Linux code is unclean, plenty of parts of Linux code are great. I only say that because of many developer's mindsets, and The BSD Vision
and because of the Architectural Guidelines although it's open to interpretation I suppose.
I highly doubt any of the legally copied *BSD code in Linux is just copied and pasted before it goes in
I don't seeing anyone say that
-mpf
*BSD doesn't wish to use Linux code. :)
:) I too have been an Internet junkie for long over a decade, before that, I was a BBS sysop/junkie :p) Whereas a lot of the Linux users, this is their first POSIX, UNIX-like operating system, and they might never have even installed a *BSD, or played with them for more than a few days. (Hardly long enough to derive an educated opinion)
I agree with you on the zealot issue, there's zealots in both camps. The difference I see is, many BSD users have been admins for years, and have ran Linux, or admin'd a machine running it, at some point in time. (Such as myself, I've been a Linux advocate and admin/user since kernel 0.99, albeit that first machine did very little on one of my networks.
Linus Torvalds: "Oh yeah? Well people compare Linux to BSD constantly. But there can be no comparison. Any time BSD does something better than Linux, we incorporate their code. But our performance wins can't be ported to BSD because the GPL prevents it. The code just flows one-way baby, so there's no real hope that BSD can remain competitive. It's embrace and encumber cats, embrace and encumber."
Reference: SlashNot
Linus blatently stating they rip BSD code at times, which is just fine by me. Yet if *BSD sucks so horridly as the Linux zealots have stated, then why ever rip or merge code from said codebase? Some Linux users certainly need to get their facts straight before shooting their mouths off.
Even if core members could rip Linux code, I highly doubt they'd ever want to. Infact, I'd bet on it. If not for their own arrogance (Not all of them are.), but then for the sanctity, and purity of the codebase.
I say this because: The BSD Vision seems to reinforce my belief. Just my $0.02.
-mpf
What is... FreeBSD 5.2-RELEASE Release Notes Alex?
Release Notes are obtainable via many methods, not just FTP. Neither FTP nor HTTP are inheriently superior to the other for the transfer of files, so your premise is rather skewed. For you to say HTTP is better fitted to serve that specific file is a moot point.
The main FreeBSD repository server is very busy, especially during release time. That's why there's more mirrors for FreeBSD than Michael Jackson has in "neverneverland ranch". So you can download from a mirror that's geographically closest to you, thus you don't go through fubar'd routing and such.
-mpf
BSD addicts, flame me if you'd like, but I'd rather hear some intelligent reasons why anyone would choose to use a BSD over other faster, cheaper, more stable systems.
I think you need to check your facts again. Did you just actually say "Cheaper"? Oh you make me laugh. (The "faster" and "stable" comments are even funnier.)
See NetCraft For proof of what is more stable.
Just because you're unable to comprehend how to utilize BSD, doesn't in any way lesser it's value, to anyone but you. All the same, if some Windows user tried to use Linux, and was completely lost, You'd just flame him and tell him to "get a clue" or "rtfm". The person who's too uneducated to utilize it, is the only one who's missing out.
There's more reasons than I can list, I've been running *BSD variants since 1993, and Linux (before any distributions existed) since 0.99. BSD is cleaner, the -core teams usually have brighter developers on them, FreeBSD ports are still number 1, the BSD layout is just far superior IMHO, (I usually tend to run slackware, which is the closest to BSD you can get in a Linux basically) my memory footprint is slightly over 3 MB on most of my servers, with a minimal set of services running, (obviously my mysql, apache, and such servers eat more due to the daemons tieing it up) and the list continues on.
Don't flame BSD because you can't figure it out.
Sounds like a personal problem.
-mpf
Review my posts, and then tell me I'm trolling.
/.? No, I'm methodically disproving their posts. Don't try and nix free speech, people are free to say what they wish, no matter how oblivious they are to the facts.
Refering to BSD users, as "lusers" - absolutely classic. What is this, the third grade? That's why all my Linux buddies say the BSD's are "too hard" for them, right?
Right.
(Which, in itself that statement is laden with humor, to me)
Look at your own trolls first, before you point the finger on us.
There's a multitude of Cut and Paste trolling going on, if you'd only pay attention. Am I crying to
I've been running FreeBSD and Linux for over a decade, and I don't bash Linux, especially not to this depth. I've made comments on how it could improve, that's not bashing. I run it. I prefer FreeBSD as a server and desktop, but that's ME. To each his own. Something about running xf86, and having such a tiny OS footprint, is a very beautiful thing.
/. :) )
Trolls are posting absolutely hilarious things, which can't go unchecked. I think the funniest thing I've seen was: "It has no GUI!" There have been so many funny things said by them however, I can't keep track. Some of these Linux zealots have really gone off the deep end, to the point where it's just funny.
(Which is the only reason I started participating in these threads, or hopped on
It really proves how clueless so many of them are, as most of the time they're talking out their ass, and it shows.
-mpf
Current Uptimes
I see lots of trolls posting NetCraft stats, so here's yet some more to entertain their lonely lives.
Dead you say? It must be so, look how long these machines have been up, it certainly time for something new.
Gander at which OS's dominate the list. This certainly doesn't mean anything, now does it?
For a dead guy, BSD sure has a lot of kick left in them. For those of you that're behind, plans are already being made for the 6.x branch of FreeBSD. That sure sounds like a dying OS, now doesn't it?
I've got no beef with Linux. I run Linux, and multiple BSDs, and have for over a decade. I've got a beef with trolls making false statements and any moderators that bump their post to informative.
*chuckle*
We're far from through.
-mpf
Honestly, I wish there was some other reason, but it's the most clear and concise explaination.
/. ;)
So to your comment I say, Amen.
I don't watch TechTV, (infact, I utterly despise it) yet I remember the offmyserver.com postings about it, and if memory serves, all this seems to have started up around that time? Perhaps I'm mistaken, as I don't frequent
-mpf
This I am aware of, as I've seen this identical posting atleast 4 times in other threads. I just had to go on record and disprove his mindless dribble, if not merely for my own satisfaction.
-mpf
Experienced users generally favor compiling from ports for the added flexibility and control it provides. Many of us need to tweak numberous compile-time options, which of course a binary package is absolutely useless in this respect.
:p )
If you're new to FreeBSD, or you're running off that p90 still, binaries are great, otherwise you'll find building from ports to be rather pleasing. ( Unless, you're building kde or something along those lines, then binary can be advantageous
There's more than 1 way to skin a cat.
-mpf
You must have nothing better to do then troll slashdot, which is fine. It's still more a waste of time then listening to a Captain and Tennille album.
/etc/sysctl (or you can use sysctl -w to add these):
Your link to the SysAdmin story is wonderfully entertaining, but you failed to realize those guys couldn't tune a FreeBSD, or any other *BSD box to save their life. FreeBSD ships with extremely small settings, and requires quite a bit of kernel tweaking via sysctl variables, and loader defaults (or recompile) for proper useage of memory, disk, and network performance on a high load, or high traffic system. (Such as SysAdmin is benchmarking in said test)
Quote from SysAdmin Magazine:
On FreeBSD: Append to
kern.maxfiles=65536
kern.maxfilesperproc=32768
The above are their "kernel tweaks", which I must say, is short of absolute hilarity. (All the fbsd admins are probably laughing with me too at these meager attempts to "tweak", by SysAdmin Magazine. This is just short of attempting to put a band-aid over a gash the size of your fist, and expecting the bleeding to stop.)
If you publish technical benchmarks, and state that your company is knowledgable and competent (Which I seriously question here.) atleast have the common courtesy to have someone experienced enough to tweak the server (as any large institution/corp/etc would, or like me, have a default sysctl.conf/loader.rc specifically for large load servers) Anyone who's ever admin'd a server with 15,000+ concurrent connections knows, a default FreeBSD box will "fall over and die" without tweaking. (More specifically, you'll fill up your mbufs, or exhaust your sockets, long before smashing the kernel to death.) After tweaking, it will easily do it with minimal load (dependant upon which services, obviously) as my servers are living proof of, along with Yahoo, and many, many others. FreeBSD's default kernel values are at very low, "safe" values, which in many areas will lead to much slower, degraded performance from what it is capable. (TCP Window Scaling is extremely low out of the box for just one small, yet very important example, and you'll see further ahead, this will easily lead to deathly slow socket connections, and such.)
The theory for this is, it should lead to a much more stable overall machine, albeit I've never had any of my tweaked machines reboot.
Quote from SysAdmin article:
FreeBSD performance decreased somewhat when more than 1500 connections were added.
I don't doubt this, as default FreeBSD install will have very minimal settings for max connections, max allocated sockets, max files per process, low fd's, again, low tcp window scaling, and more which for brevity I'll exclude.
To say FreeBSD performs worse than Linux, and Solaris, AND, Win... Absolutely laughable. I'm sorry, I've been in the UNIX/Networking field (traversing more medias than I wish to list) for over 12 years, I've run the entire gambit, and still do, so I find that statement highly offensive, because it's plain wrong.
To Sys Admin Magazine: Learn how to tweak a FreeBSD kernel, then run the test. We as the readers have an obligation to expect absolute technical perfection from those publishing these types of articles. Apples to Oranges.
More NetCraft fodder:
Seven of the top nine sites run on FreeBSD. The exceptions are Datapipe, which is doing a fine job of promoting the reliability of Windows 2003, and German hosting company komplex.net which runs on Linux.
Posted by mhp at 08:54 AM UTC on Jan 11, 2004
Had you ever contemplated the fact that FreeBSD, nor any other *BSD's success does NOT rely on how many people use it? We really couldn't care less. We'd love more userbase, and more contributors, who wouldn't? The fact that we have less users than Windows, or Linux, or OS X, is totally irrelevant to us.
A proj
I wasn't talking about "building" via that method now was I? :) I was just trying to clear up what he had said, since the FreeBSD ports do not install binary packages unless explicitly told to, he was stating to the contrary.
-mpf
No, You're refering to installing a PACKAGE. FreeBSD PACKAGES are binary forms of the application. I.E.: # pkg_add -r lynx.
/usr/ports/www/lynx && make install clean
This will download and install the binary package of lynx, thus no compiling necessary.
FreeBSD ports, however are built from source.
If you: # cd
You're downloading lynx's source tarball and it's dependancies, then compiling it from said source(s) via ports.
Again, only packages are binary distributions of a program.
-mpf
Losing JKH from -core was to say the least, unsavory, yet you think FreeBSD is dependant on 1 or 2 members of the -core team? I mean... come on, seriously? If FreeBSD is "dead" because of a 'decline in market share' or due to losing a member or two of the core team, you're just sadly mistaken. To play devil's advocate: Let's assume you're correct, and we lose half our user base. What a pity! /*sarcasm*/ Regardless, we'll continue to hack away, we really don't care - and our success is NOT measured by how many use our "Product." If that were the case around the board, Linux might as well be refered to as "dead", since Microsoft reigns supreme, still.
.. I think I might know why *BSD trolling has seen a sharp increase in the last 6+ months or so, Hmm, gee..
I quote:
Seven of the top nine sites run on FreeBSD. The exceptions are Datapipe, which is doing a fine job of promoting the reliability of Windows 2003, and German hosting company komplex.net which runs on Linux.
Posted by mhp at 08:54 AM UTC on Jan 11, 2004 in Hosting, Performance.
Reference: news.netcraft.com
FreeBSD seems to have been whooping ass in the hosting field for a while now according to NetCraft surveys. (Long before this article's 6 month window as well.)
This NetCraft article proves nothing about say, Linux vs. FreeBSD or FreeBSD vs. Win2k3, to me it says, FreeBSD isn't dying off, and if it is, it sure is a pain riddled, horrifcally slow death. (if less people are running FreeBSD, good, less morons to have to help , I'm sick of being asked how to update the locate database after all these years..)
Before you put words in JKH's mouth: See JKH's resignation
Why did Apple hire JKH? Not because he looks good in a push-up bra, I can promise you that.
If JKH thinks FreeBSD is no longer viable code, why does Apple continue to develop a branch off FreeBSD, and why is Panther based from FreeBSD 5.x? OS X Panther(features column)
FreeBSD isn't a "clique" of elites, we didn't all install FreeBSD because it's "Cool". So many Linux zealots have called FreeBSD users out, saying we're trying to "fit in." Which is ironic, since many of them installed it because their buddies said, "Oh l00k, itz 31337 d00d."
I've run Linux since kernel 0.99, there was a time I prefered Linux(0.99 - 2.0.34) over any BSD, those days are long passed. However, I DO NOT look down my nose at Linux users, and I certainly don't troll any Linux "forums", yet why is this occuring vice versa? It's unfortunate, the Linux community is also very plagued by the '14 year old zit-faced kids who just installed RedHat, and want to "hax teh g1bs0n."' (I've taken enough abuse from the Linux crowd, talking about "Fat, unwashed, Star Trek watching *BSD nerds" to justify that last comment.) Which is one reason I don't ever want them crossing over to the FreeBSD crowd, thus if we have less 'market share', in that respect it's a VERY GOOD THING(tm).
Many of my sysadmin friends run purely Linux based networks(server-side), and when I ask why they don't even try a *BSD, the common answer I've received is, "It's too hard." Which in turn makes me chuckle to myself because they're really missing out on something great. I don't make fun of them and/or their networks, nor try to say they're running an antiquated or conviluted system. (To say FreeBSD is either of these, is to deny the facts.) To each his own, and if they choose to live in ignorance, MORE POWER TO THEM, it's their prerogative.
We were around in 1992, and we'll continue to be, regardless what the slashdot trolls attentively (or not so attentively..) splurge in digital form.
-mpf
P.S. I wonder how long Yahoo!(tm) will run a dead operating system (FreeBSD) across it's server pools?
( Next time don't post out of date,
My mistake. There's been almost nothing but trolling going on in these threads, so I suppose I've got an itchy trigger finger at times. ;)
-mpf