The less restrictive BSD license may have looked nicer to apple execs than the GPL, but i don't think that's what made up their mind. Apple bought NeXT which used a Mach 2.5 kernel that has a BSD API and BSD-like drivers. That's why they're using it. And it's not BSD, it's Mach with a BSD API (2.5) or a BSD server process (3.0).
8megs/sec isn't that fast. and you can't get 10,000 rpm 30meg/sec ata drives. it is true that for most users, the price difference between scsi and ata isn't worth it (unless they really want it).
Also, since my addresses are subnetted (/29), I can have a static IP block, and no one can sniff it, since they aren't on my subnet!
IP addresses have nothing to do with that. The only way they wouldn't be able to sniff is if you're on a different segment or their router/modem/whatever-you-want-to-call-it filters by MAC or IP address.
that's right, models! gyrating barely clad chicks between decked out alpha and ppc boxes, a dream come true. or put them in a refridgerated area with some penguins; the pengiuns will be nice and comfy and you can image why i think the girls should be kept in a cold enviroment.:P
man was i pissed when the linux expo (it was the linux expo, wasn't it?) was moved from DC to new york. i could of gone with a bunch of people i know. it sucked! DC needs the 'guin!
some many people here seem to want to take everything personally and misinturpret(sp) things so that they seem more offensive towards themselves. eg, all these claims of "he's saying all nerds/geeks/tech people are autistic", when that is CLEARLY not what the article said or implied.
seems like there's a lot of people on slashdot who like to fly off the handle and seem to have poor reading comprehension skills.
The obvious question is: why characterize the jock extreme as genial and average, but depict the geek extreme as the early onset of a disease!?
*sigh* go re-read the article. it does not imply that everyone technically inclined is diseased or deformed. it suggests a possible explaination for extreme examples of behavior that most people would call nerdish or geeky. there is no reason for everyone to say things like "this isn't true, i know geeks who aren't socially inept". the article never refuted that.
on a side note (since i'm already typing), i saw a press conference with bill gates once and he was sitting on a table swinging his legs like a 6 year old. at the time it really struct me as strange; not even something i'd consider to be covered by the geek or nerd stereotype.
Um, it's the unknown bugs that are the problem. Making source available does not always result in fewer bugs. At all. There are plenty of rock solid closed applications, and plenty of flakey open source applications*. People who want stability go with stable software, wherever it is from.
with open source bugs usually get fixed orders of magnitude faster than closed source software. you're right though, open source software can suck, too. however, given equal stability and security, i'd choose free, open source over expensive closed programs.
no, macs are not the mac os. they come with it installed, but that doesn't mean they are actually it.
mac os (and mac firmware) do have good plug and play support (much better than windows and x86 bios). the only time i ever had to worry about irqs and other stuff in linux was on an x86 machine which had a pnp 3com nic that i couldn't seem to get working until i disabled pnp on it.
as for "a cold heartless box", is that opposed to a warm and caring mac os box?
one thing mac os definetly doesn't have, is configurability. you can't override irqs, you can't rewrite things (closed source), and most apps hide advanced options from the user (that's if they even allow you to change them at all). my biggest gripe about mac os is that apps always seem dumbed down to the lowest common denominator, and hidden options to protect users from themselves.
(btw, i used mac os for years on a 68k; then i got a 7100 and stuck mklinux on it. i've been using linux ever since then.)
just write code to get a list of the mounted filesystems.
as for crack-a-mac, that doesn't prove anything (neither does the linuxppc crack challenge or any other crack-a-box contest). how do you know that competent people tried to crack it? how do you know they didn't miss something? a source code audit is a much better way to find holes. letting a bunch of script kiddies go at a box doesn't prove anything.
Those who remember these kind of things will note that *the* definitive, original WinNuke was a bug in the TCP handling of an "Out Of Band" packet sent to port 139 on a Windows box. Open door. Boom.
apple's tcp/ip stacks have been vulnerable to the same bugs as MS, linux, and *bsd stacks have been; eg, ping o' death (oversize ping packet).
just cuz mac os doesn't support remote logins doesn't mean they're secure. ever hear of buffer overflows? find one, smash the stack and take over. make a few toolbox calls and erease all the files on the server.
open source software, on the other hand, allows you to check the source yourself (eg, grep strcpy *.c) and quickly fix known bugs.
using mac os stuff and saying it's more secure cuz there's no logins is false. saying it's more secure cuz less crackers try to crack mac os apps or cuz there's less spoilt scripts out there for mac os apps is security though obscurity.
that IS the way it is many places, when compared to freebsd. as for the "technical superiority", of course he thinks freebsd is superior, why else would use it? yes, he is at least somewhat biased, but so are many linux users. and freebsd IS superior at certain things. their cam scsi system is better than linux's current scsi system. what can be done about it? improve linux's scsi support. believe it or not, linux is not perfect, there is still plenty of room for improvement, no matter how good it is now.
he was talking about the process in which code is committed to the freebsd code base, so their code review process is quite relevant. did he SAY that the linux method is poorer? no. and so what if he did? god forbid anyone should point out a shortcomming of the linux community. the fact is some linux code DOES need to be better, and would do well to adopt the freebsd method. redhat is one example i can think of; almost every version of their initscripts i've ever seen have been buggy. personally, though, i use debian now and find it to be a very professional dist (let's not let this degenerate further into a linux dist war, though.) constructive critisism is a good thing.
no, you took that totally out of context. please explain to me how that implies that linux crashes; it doesn't. iirc, he was saying that freebsd's code review helps prevent buggy code from being committed. he didn't say the methods that linux developers use lead to more bugs. learn to read before you go proclaiming something to be anti-linux. i personally found it to be a good article, and in no way offensive. note that i have written code under both the GPL license and a BSD type license, though the vast majority of the code i've written is under the GPL and i consider myself a linux developer.
/me can't wait to get a new g4 and run ppclinux, mklinux, and netbsd on it.
um, how exactly would shadowed passwds stop you from booting from a floppy and editing/etc/shadow? (hint, they won't). from what i can gather, what he did was boot from a floppy, mount the root fs and remove the encrypted root password from/etc/shadow.
most RISC cpus have both big and little endian modes. usually one or the other mode is better supported by the rest of the hardware on the mobo and is the default. for example, alpha defaults to little endian and ppc defaults to big endian.
also, nt was ported to mips. i'm not sure if nt ran it in little endian or big endian (default).
Slashdot Mirror
The less restrictive BSD license may have looked nicer to apple execs than the GPL, but i don't think that's what made up their mind. Apple bought NeXT which used a Mach 2.5 kernel that has a BSD API and BSD-like drivers. That's why they're using it. And it's not BSD, it's Mach with a BSD API (2.5) or a BSD server process (3.0).
8megs/sec isn't that fast. and you can't get 10,000 rpm 30meg/sec ata drives. it is true that for most users, the price difference between scsi and ata isn't worth it (unless they really want it).
IP addresses have nothing to do with that. The only way they wouldn't be able to sniff is if you're on a different segment or their router/modem/whatever-you-want-to-call-it filters by MAC or IP address.
/me -pedantic "you've got mail"
"you've" is a contraction of "you have", so this statement evaluates to "you have got mail", which is poor english. the correct phrases are:
you've gotten mail (you have recieved mail in the past)
or
you've mail (you have mail)
QNX runs on PPC. It probably doesn't support pmacs, though, but i'm guessing you'll probably be able to find embedded altivec g4 based SBCs soon.
has anyone actually looked at what 3com coded? i mean, what did they add? afaik, the 3com drivers worked just fine and weren't lacking any features.
that's right, models! gyrating barely clad chicks between decked out alpha and ppc boxes, a dream come true. or put them in a refridgerated area with some penguins; the pengiuns will be nice and comfy and you can image why i think the girls should be kept in a cold enviroment. :P
man was i pissed when the linux expo (it was the linux expo, wasn't it?) was moved from DC to new york. i could of gone with a bunch of people i know. it sucked! DC needs the 'guin!
finally, someone with some sense!
some many people here seem to want to take everything personally and misinturpret(sp) things so that they seem more offensive towards themselves. eg, all these claims of "he's saying all nerds/geeks/tech people are autistic", when that is CLEARLY not what the article said or implied.
The obvious question is: why characterize the jock extreme as genial and average, but depict the geek extreme as the early onset of a disease!?
*sigh* go re-read the article. it does not imply that everyone technically inclined is diseased or deformed. it suggests a possible explaination for extreme examples of behavior that most people would call nerdish or geeky.
there is no reason for everyone to say things like "this isn't true, i know geeks who aren't socially inept". the article never refuted that.
on a side note (since i'm already typing), i saw a press conference with bill gates once and he was sitting on a table swinging his legs like a 6 year old. at the time it really struct me as strange; not even something i'd consider to be covered by the geek or nerd stereotype.
Um, it's the unknown bugs that are the problem. Making source available does not always result in fewer bugs. At all. There are plenty of rock solid closed applications, and plenty of flakey open source applications*. People who want stability go with stable software, wherever it is from.
with open source bugs usually get fixed orders of magnitude faster than closed source software. you're right though, open source software can suck, too. however, given equal stability and security, i'd choose free, open source over expensive closed programs.
no, macs are not the mac os. they come with it installed, but that doesn't mean they are actually it.
mac os (and mac firmware) do have good plug and play support (much better than windows and x86 bios). the only time i ever had to worry about irqs and other stuff in linux was on an x86 machine which had a pnp 3com nic that i couldn't seem to get working until i disabled pnp on it.
as for "a cold heartless box", is that opposed to a warm and caring mac os box?
one thing mac os definetly doesn't have, is configurability. you can't override irqs, you can't rewrite things (closed source), and most apps hide advanced options from the user (that's if they even allow you to change them at all). my biggest gripe about mac os is that apps always seem dumbed down to the lowest common denominator, and hidden options to protect users from themselves.
(btw, i used mac os for years on a 68k; then i got a 7100 and stuck mklinux on it. i've been using linux ever since then.)
just write code to get a list of the mounted filesystems.
as for crack-a-mac, that doesn't prove anything (neither does the linuxppc crack challenge or any other crack-a-box contest). how do you know that competent people tried to crack it? how do you know they didn't miss something? a source code audit is a much better way to find holes. letting a bunch of script kiddies go at a box doesn't prove anything.
Those who remember these kind of things will note that *the* definitive, original WinNuke was a bug in the TCP handling of an "Out Of Band" packet sent to port 139 on a Windows box. Open door. Boom.
apple's tcp/ip stacks have been vulnerable to the same bugs as MS, linux, and *bsd stacks have been; eg, ping o' death (oversize ping packet).
just cuz mac os doesn't support remote logins doesn't mean they're secure. ever hear of buffer overflows? find one, smash the stack and take over. make a few toolbox calls and erease all the files on the server.
open source software, on the other hand, allows you to check the source yourself (eg, grep strcpy *.c) and quickly fix known bugs.
using mac os stuff and saying it's more secure cuz there's no logins is false. saying it's more secure cuz less crackers try to crack mac os apps or cuz there's less spoilt scripts out there for mac os apps is security though obscurity.
everyone keeps saying "use linux, not a mac". macs are hardware, linux is software; so that statement makes no sense. as for me, i run linux on a mac.
the real question is, does the card reader run linux? :P
s/maclinux/mklinux/
8)
what exactly is the windows CSP used for? general crypto or auth'ing certaing programs as coming from trusted sources or what?
that IS the way it is many places, when compared to freebsd. as for the "technical superiority", of course he thinks freebsd is superior, why else would use it? yes, he is at least somewhat biased, but so are many linux users. and freebsd IS superior at certain things. their cam scsi system is better than linux's current scsi system. what can be done about it? improve linux's scsi support. believe it or not, linux is not perfect, there is still plenty of room for improvement, no matter how good it is now.
he was talking about the process in which code is committed to the freebsd code base, so their code review process is quite relevant. did he SAY that the linux method is poorer? no. and so what if he did? god forbid anyone should point out a shortcomming of the linux community. the fact is some linux code DOES need to be better, and would do well to adopt the freebsd method. redhat is one example i can think of; almost every version of their initscripts i've ever seen have been buggy. personally, though, i use debian now and find it to be a very professional dist (let's not let this degenerate further into a linux dist war, though.) constructive critisism is a good thing.
no, you took that totally out of context. please explain to me how that implies that linux crashes; it doesn't. iirc, he was saying that freebsd's code review helps prevent buggy code from being committed. he didn't say the methods that linux developers use lead to more bugs. learn to read before you go proclaiming something to be anti-linux. i personally found it to be a good article, and in no way offensive. note that i have written code under both the GPL license and a BSD type license, though the vast majority of the code i've written is under the GPL and i consider myself a linux developer.
/me can't wait to get a new g4 and run ppclinux, mklinux, and netbsd on it.
um, how exactly would shadowed passwds stop you from booting from a floppy and editing /etc/shadow? (hint, they won't). from what i can gather, what he did was boot from a floppy, mount the root fs and remove the encrypted root password from /etc/shadow.
most RISC cpus have both big and little endian modes. usually one or the other mode is better supported by the rest of the hardware on the mobo and is the default. for example, alpha defaults to little endian and ppc defaults to big endian.
also, nt was ported to mips. i'm not sure if nt ran it in little endian or big endian (default).