There's no substiture for good code ( a good coder ).
There's plenty of examples where good coders have done great jobs in C, Qmail, djbdns, BSD lib c, and so on.. so how about we teach programmers better programming? how about we start making and enforcing wrapper libraries like safe string?
I'll be nice so easily switch to so called " safer " languages.. but that won't happen in the real world...
"Yes, BSD has been around a lot longer, but to be honest, it hasn't
really come anywhere in the last few years, whereas Linux distributions
have gone from being as frightening to install and maintain as FreeBSD
(think Slackware 3.0 here) to something as elegant as Conectiva or
Mandrake." (I'm better than you are)
"After reading the FreeBSD Handbook, you attempt the install.
Everything seems to go smoothly, but when you boot the system, it hangs.
You go in search of help, reading the FAQ How-To sections, but can't
figure out where to begin, or what the problem is. So you go on IRC and
get flamed for asking for some help. I guess Microsoft just got another
Windows 2000 user." (Not
everyone is an expert)
"If FreeBSD wants to believe that they are so good, and so elite,
then eventually the system will only be used (and usable) by so few
people that it won't really matter. Instead of admitting their faults,
they could continue to point the finger at everyone else?s, while
ignoring the majority of users. So go ahead, pretend that what you've
got is perfect and that nothing needs to change. Watch what happens."
(New FreeBSD Core
Team)
"In this [Jordan K. Hubbard's] article, he openly admits to some of
the shortcomings of FreeBSD -- particularly its installation process --
and discusses some possible solutions. In my opinion, the future looks
bleak." (The future of
FreeBSD?)
"I wouldn't recommend it [OpenBSD] for a desktop machine, though.
You have to give up some of the "fluff" to get the security." (The uncrackable
OpenBSD)
"New users should steer clear of OpenBSD." (Open BSD 2.6)
"At best, NetBSD is a minimal operating system. It seems stable, but
configuring it is an adventure every time. That, combined with a lack
of applications support, make this an easy candidate to pass over in
lieu of more user-friendly Operating Systems." (A review of NetBSD
1.4.2)
Didn't the new BSDL get rid of the ad clause?
Plus, a great deal of OS still contain in their docs stuff like " Contains code license for the UNiversity of California" etc etc
Links my friend.... Links... Last i've heard, we still mopped the floor with linux on networking...
Have you ever tried to do any bandwith shapping with linux, or routing? anything network intensive? then you would now why most router OS's have based their stacks on the BSD code.
But, since you didn't provide links, I WILL BUDDY.
case 1) NFR's home page.
quote
The performance of NFR on Linux will be poor on any hardware when compared to NFR on BSD-based systems on
the same hardware. Linux does not use the BPF. The libpcap library uses another method to extract packets from
the kernel on Linux. The code for this method does not appear to be written with performance in mind. Programs
such as NFR, which use libpcap to read packets from the interface in promiscuous mode, will experience significant
packet loss on any network that sees traffic of several megabits per second or more.
Linux does not properly handle interfaces in promiscuous mode. It fails to it fails to distinguish packets addressed to
it from packets addressed to other machines. This means that you can subvert the Linux system in various ways:
Other systems on the network can detect Linux based sniffers by looking for responses to requests sent to the
wrong MAC address. The Apostols Web page (http://www.apostols.org/projectz) (in Spanish) describes the
exploit. The source code for the exploit program contains comments and error messages in English.
On an NFR that is multihomed, someone could use the flaws in Linux to route traffic from the promiscuous
interface to other interfaces.
To Have a secure version of linux, or to have a secure version of a Free OS.
Because if it's the latter, linux was the wrong choice.... any of the BSD's would have been ten times a better choice.
ALL the BSD's, OpenBSD being the most adiment about it, audit their code, and the code they put into their OS's, NO linux distribution that i know of does this ( not counting this one)
I would have made much more sense, and produced far better results if they would have started from one of the BSD's...
This is were you people don't get what BSD, X type licenses exist.
when i say you people, i mean GPL advocates.
Let me give you two reasons, of many, for these licenses.
1) They are meant to create standards and procedures that are to be accepted industry wide.
The gpl CAN NOT lend it self to this.
Case in point. TCP/IP, X, Sockets.
2) No company can really "steal code", do you have any idea how hard it is to incorporate foreign code into a different OS or application. They only get ideas on how to implement it in their enviroment. Stealing ideas should not matter much to OSS dev's, since they are not patenting their ideas.
3) A company cannot just take a BSD/X licensed programm, repackage it and resell it bin only. Don;t you think people would notice? and use the free altelnative?.
What they do do is, add missing features and funtionality to these packages and repackage it. They have the RIGHT to do this, just like the original dev's had the right to release it under a free license. They worked hard for these features, and of they feel that they must license them in the way they please, who the hell are you to say other wise, go and code your own.
Not carefull enought, as it was an inferior
VM, SCSI subsystem, and IP stack....
Not to mention general inmature code.
secondly, HE'S NOT STRIPPING THE KERNEL.
that's the ONLY part he's using.
He's basing his os on the linux kernel, but supplying from scratch everything else.
Slashdot Mirror
except the level of expertice of most HP-UX users is far greater than most linux users....
There's no substiture for good code ( a good coder ).
;)
There's plenty of examples where good coders have done great jobs in C, Qmail, djbdns, BSD lib c, and so on.. so how about we teach programmers better programming? how about we start making and enforcing wrapper libraries like safe string?
I'll be nice so easily switch to so called " safer " languages.. but that won't happen in the real world...
just my 2 nickels
Meat is tasty :)
This might seem a little offtopic, but it really is no.
/bin/sh != sh /bin/sh == bash
How do you feel about what some linux distributions do, they replace the Bourne Shell with the Bourne Again Shell.
did you people read their inflamatory and thoughroughly incorrect articles supposedly "pro-BSD"?
Please people, inform yourselfs before u comment...
See here
see their crap??
Yes, it has been supporting these for a lont time, so it is a mor emature solution.
stateful inspection (ipf) and rate limiting (dummy net thru ipfw).
ratelimiting agains't many DoS type attacks is actually built into the default kernel.
and of course, it will very useful for the script kiddie that used the root toolkit of the day to take over your machine....
Oh wait, this is linux, you guys are playing catch up... :-P...
But seriously, any more questions on the maturity/performance of the BSD's networking code? this stuff has been posible for AGES...
Where does it say the price?
Ok, how about newer benchmarks?m g36.htm
/ im g37.htm
http://innominate.org/%7Etgr/slides/performance/i
http://innominate.org/%7Etgr/slides/performance
Didn't the new BSDL get rid of the ad clause?
Plus, a great deal of OS still contain in their docs stuff like " Contains code license for the UNiversity of California" etc etc
Here .... http://www.nfr.net/nfr/SYSTEM_NOTES.html#LinuxGene ral
Have you ever tried to do any bandwith shapping with linux, or routing? anything network intensive? then you would now why most router OS's have based their stacks on the BSD code.
But, since you didn't provide links, I WILL BUDDY.
case 1) NFR's home page. quote
More you ask?
http://neuromancer.rmci.net/linux-vs-freebsd.html
Please stop talking from the wrong end....
sorry, by standard, i really meant "widely used", or "industry" standard...
The BSDL allows for this...
Hence why TCP/IP became a standard... there was a freely available stack for eveyone to use.
Amen Brother ;)
To Have a secure version of linux, or to have a secure version of a Free OS.
Because if it's the latter, linux was the wrong choice.... any of the BSD's would have been ten times a better choice.
ALL the BSD's, OpenBSD being the most adiment about it, audit their code, and the code they put into their OS's, NO linux distribution that i know of does this ( not counting this one)
I would have made much more sense, and produced far better results if they would have started from one of the BSD's...
Just my opinion, flame as you please....
Heh, you just proved another point of mine...
Most GPL and LGPL advocates are either idiots... Or too much of a zealot to realize the truth...
And again, please explain the sockets argument, because that just made you sound like a true MORON.
Am surrounded by idiots....
i was retracting from the insulting tone, NOT from the comments buddy....
the BSD are and will be, a more mature code base,system, and architecture.
sorry hehe
This is were you people don't get what BSD, X type licenses exist.
when i say you people, i mean GPL advocates.
Let me give you two reasons, of many, for these licenses.
1) They are meant to create standards and procedures that are to be accepted industry wide. The gpl CAN NOT lend it self to this. Case in point. TCP/IP, X, Sockets.
2) No company can really "steal code", do you have any idea how hard it is to incorporate foreign code into a different OS or application. They only get ideas on how to implement it in their enviroment. Stealing ideas should not matter much to OSS dev's, since they are not patenting their ideas.
3) A company cannot just take a BSD/X licensed programm, repackage it and resell it bin only. Don;t you think people would notice? and use the free altelnative?.
What they do do is, add missing features and funtionality to these packages and repackage it. They have the RIGHT to do this, just like the original dev's had the right to release it under a free license. They worked hard for these features, and of they feel that they must license them in the way they please, who the hell are you to say other wise, go and code your own.
well, that's my little rant of the day.
Very true.....
;)
This is the reason why linux binaries under API emulation, tend to run faster on FreeBSD
In the old inmortal words.......
You Sir, are a true idiot.
Linux carefull to avoid BSD's mistakes....
Not carefull enought, as it was an inferior VM, SCSI subsystem, and IP stack....
Not to mention general inmature code.
secondly, HE'S NOT STRIPPING THE KERNEL.
that's the ONLY part he's using.
He's basing his os on the linux kernel, but supplying from scratch everything else.
do you happen to know to what extend BSD code is used... and what flavor ( F/O/N BSD or BSDi ) ?
He gives the explanation on why they shouldn't be on disks....
ARRGGHH