How many times do we have to have this argument on slashdot/Usenet/lists/wherever... we all have the same objective, and we all have different tastes. If people want to use Linux, that's upto them, and likewise with BSD
Both options are better than an NT/closed-source solution... why can't we all just admit we're happy with what we are, and the OS is suited to what we do?
We'd all be far more productive if we stopped arguing... personally, I'm getting tired of the arguments on slashdot between BSD and Linux. If you don't know which is best for you already, use Redhat - it's simpler to install... if you are an admin of a multi-server, high-availability, high-performance site, then you should have the qualifications and resources to DO THE RESEARCH YOURSELF!!!!!
Would you agree that it is impossible for large-traffic sites to ever be able to protect themselves from distributed traffic-generation attacks? I ask this assuming that the type of attack uses traffic that looks no different to 'normal browsing' traffic (which is probably the most difficult to detect). What's more, if there are analysis systems in place to detect such attacks, what measure can be taken to ensure that those systems themselves don't fall under the DoS attack by being flooded with traffic that they have to analyse?
Well, it wouldn't hurt, but yes there is the question as to whether it is really worth the time and effort, just to help people who sound as they would be happier in a windowed environment anyway (I just can't understand what is so difficult about man'ing the command to find out the options)
My main argument is one of backward compatability with scripts, and the fact that those of us who have learnt all the standard command line flags for things like tar, gzip, sendmail and god knows what else will have to re-learn, as well as re-code our scripts. If the newbies want to have a standard set of the longer GNU-like options (--version, --force, etc.) then that's fine, but don't even think about changing the shorter options...
There's also the point that options should be more intuitive to the function of the program, so that -f would mean 'filename' in one program but 'force' in another - this plan of standardisation means that filename suddenly becomes -X or something in later releases... uggghhhh...
I've SMP running on FreeBSD-3.2 (I'll upgrade when I can afford the downtime) on a couple of dual PIII-450's.
Generally, seems fine, but very coarse granularity in general. What I've found which is kind of cute, is that if there is something like crack running on there, it'll basically get an entire CPU to itself and everything else gets shipped onto the other. Also make -j2 type work seems to fly-by in a nice fashion.
BSD in general got a rep for a stable IP stack, and we could go back and argue about those weird Linux/NT tests and how NT uses a basically BSD-derived stack, but we won't, eh?
Although FreeBSD does seem to fly, the real speed-freaks run their webservers with Zeus rather than Apache, and as nmap'ing the top guys at Webperf.net shows, there are further tricks to be played out with Solaris as well.
Not a Solaris nut, but from what I can tell, and from what I've heard, there is an option somewhere to get the first four bytes of the actual data payload to be sent back with the SYN-ACK which makes things appear to scream along in the benchmarks.
This is not neccesarily (sp?) true. When I first got into 'nix properly it was in a large ISP operation that was running a weird mixture of Linux 2.0.33/WinNT 3.51 and 4.0 (it was a long time ago, OK?). The admins were happy with this because it was the best available at the time.
After a year or two I had worked myself into a position whereby I had input on choice of operating system, and having worked with the hell that was and still is Redhat and Slackware (I don't want to get into a religious argument here, but in 100% uptime-needed heavily-loaded ISP they're crap, OK? Have you run a network with 250,000 users? Shut up then:) ), I opted we move to a BSD.
We looked at FreeBSD then, but passed over it without trying it for OpenBSD which was then at 2.3 as it had additional security (supposedly), and seemed cleaner and purer as a Unix. Sure enough the Linux guys were pretty angry at this, but within 3 months we were building a cluster to support 500,000 customers, and we chose OpenBSD.
I've now left there, but they're migrating to FreeBSD because of the SMP support and because it's 'nicer' than OpenBSD to newer admins (which you have to consider when there is such a shortage of admins these days). I've used FreeBSD on the desktop and I find it easier, nicer, smoother, faster and more reliable than any other Unix I've ever used. In fact the ISP I am now Tech. Manager of (at the tender age of 21. woop!), is a FreeBSD house. We'll look at Solaris 8 when we have more servers to play with, but for now we're happy at FreeBSD 3.x, and I definitely want to look at 4.0
Sure, it's not like Redhat where you never have to see a command line if you don't want to, but who the hell wants that? If I wanted that, I'd use Windows. I think it's good there is diversity within the OS 'marketplace', and I think different people need different things for different jobs - but I wouldn't say that FreeBSD is no good for the desktop market, nor would I say it's in decline - I'm impartial and don't work in the FreeBSD community, and as far as I can see, it's going from strength to strength.
Most people who slag off FreeBSD haven't used it since 2.2 or something and don't know what to expect. It's fast, it's sexy and it's more-or-less 100% uptime guaranteed. I wouldn't pee on a linux box these days (unless it was that or NT), but that's cos I don't want to do Linux things - I'm an ISP admin and I need reliability, security and speed - I just don't get quite what I want from Linux.
OK, so we now change all of the flags passed to applications that were in use when you were still in short pants, and everything will break. But that's OK, because at least you know why it's broken.:)
As an example, you specified -f to mean 'force'. OK, so we'll change that, and then every time I want to tar/untar something I'll just have to do it a tape device, as I can't specify the file.:)
To be honest, although I can see why standardisation around the kiddies flags (--version and so on) is useful, I really don't see the pain in typing 'man cmd' to find out what 'cmd' does and what flags it takes. If people aren't prepared to accept that, I suggest that they have chosen the wrong OS, and should think instead about something 'easier' to learn such as MacOS or Windows.
I honestly think that if you want to fight the cause of usability in unix environments, you should be pushing for standardisation of X apps that should all carry some sort of familiarity in them (we've all seen the File/Edit/View standard of Windows). This is less likely to break things from a backwards-compatability front, is going to appeal to those that need it (the sort that will want to work exclusively in a windowed environment), and could still be done before it's too late!:)
Slashdot Mirror
How many times do we have to have this argument on slashdot/Usenet/lists/wherever... we all have the same objective, and we all have different tastes. If people want to use Linux, that's upto them, and likewise with BSD
Both options are better than an NT/closed-source solution... why can't we all just admit we're happy with what we are, and the OS is suited to what we do?
We'd all be far more productive if we stopped arguing... personally, I'm getting tired of the arguments on slashdot between BSD and Linux. If you don't know which is best for you already, use Redhat - it's simpler to install... if you are an admin of a multi-server, high-availability, high-performance site, then you should have the qualifications and resources to DO THE RESEARCH YOURSELF!!!!!
Can we all stop arguing now please?
Would you agree that it is impossible for large-traffic sites to ever be able to protect themselves from distributed traffic-generation attacks? I ask this assuming that the type of attack uses traffic that looks no different to 'normal browsing' traffic (which is probably the most difficult to detect). What's more, if there are analysis systems in place to detect such attacks, what measure can be taken to ensure that those systems themselves don't fall under the DoS attack by being flooded with traffic that they have to analyse?
Well, it wouldn't hurt, but yes there is the question as to whether it is really worth the time and effort, just to help people who sound as they would be happier in a windowed environment anyway (I just can't understand what is so difficult about man'ing the command to find out the options)
My main argument is one of backward compatability with scripts, and the fact that those of us who have learnt all the standard command line flags for things like tar, gzip, sendmail and god knows what else will have to re-learn, as well as re-code our scripts. If the newbies want to have a standard set of the longer GNU-like options (--version, --force, etc.) then that's fine, but don't even think about changing the shorter options...
There's also the point that options should be more intuitive to the function of the program, so that -f would mean 'filename' in one program but 'force' in another - this plan of standardisation means that filename suddenly becomes -X or something in later releases... uggghhhh...
I've SMP running on FreeBSD-3.2 (I'll upgrade when I can afford the downtime) on a couple of dual PIII-450's.
Generally, seems fine, but very coarse granularity in general. What I've found which is kind of cute, is that if there is something like crack running on there, it'll basically get an entire CPU to itself and everything else gets shipped onto the other. Also make -j2 type work seems to fly-by in a nice fashion.
Just my 2p worth
BSD in general got a rep for a stable IP stack, and we could go back and argue about those weird Linux/NT tests and how NT uses a basically BSD-derived stack, but we won't, eh?
Although FreeBSD does seem to fly, the real speed-freaks run their webservers with Zeus rather than Apache, and as nmap'ing the top guys at Webperf.net shows, there are further tricks to be played out with Solaris as well.
Not a Solaris nut, but from what I can tell, and from what I've heard, there is an option somewhere to get the first four bytes of the actual data payload to be sent back with the SYN-ACK which makes things appear to scream along in the benchmarks.
This is not neccesarily (sp?) true. When I first got into 'nix properly it was in a large ISP operation that was running a weird mixture of Linux 2.0.33/WinNT 3.51 and 4.0 (it was a long time ago, OK?). The admins were happy with this because it was the best available at the time.
:) ), I opted we move to a BSD.
After a year or two I had worked myself into a position whereby I had input on choice of operating system, and having worked with the hell that was and still is Redhat and Slackware (I don't want to get into a religious argument here, but in 100% uptime-needed heavily-loaded ISP they're crap, OK? Have you run a network with 250,000 users? Shut up then
We looked at FreeBSD then, but passed over it without trying it for OpenBSD which was then at 2.3 as it had additional security (supposedly), and seemed cleaner and purer as a Unix. Sure enough the Linux guys were pretty angry at this, but within 3 months we were building a cluster to support 500,000 customers, and we chose OpenBSD.
I've now left there, but they're migrating to FreeBSD because of the SMP support and because it's 'nicer' than OpenBSD to newer admins (which you have to consider when there is such a shortage of admins these days). I've used FreeBSD on the desktop and I find it easier, nicer, smoother, faster and more reliable than any other Unix I've ever used. In fact the ISP I am now Tech. Manager of (at the tender age of 21. woop!), is a FreeBSD house. We'll look at Solaris 8 when we have more servers to play with, but for now we're happy at FreeBSD 3.x, and I definitely want to look at 4.0
Sure, it's not like Redhat where you never have to see a command line if you don't want to, but who the hell wants that? If I wanted that, I'd use Windows. I think it's good there is diversity within the OS 'marketplace', and I think different people need different things for different jobs - but I wouldn't say that FreeBSD is no good for the desktop market, nor would I say it's in decline - I'm impartial and don't work in the FreeBSD community, and as far as I can see, it's going from strength to strength.
Most people who slag off FreeBSD haven't used it since 2.2 or something and don't know what to expect. It's fast, it's sexy and it's more-or-less 100% uptime guaranteed. I wouldn't pee on a linux box these days (unless it was that or NT), but that's cos I don't want to do Linux things - I'm an ISP admin and I need reliability, security and speed - I just don't get quite what I want from Linux.
OK, so we now change all of the flags passed to applications that were in use when you were still in short pants, and everything will break. But that's OK, because at least you know why it's broken. :)
:)
:)
As an example, you specified -f to mean 'force'. OK, so we'll change that, and then every time I want to tar/untar something I'll just have to do it a tape device, as I can't specify the file.
To be honest, although I can see why standardisation around the kiddies flags (--version and so on) is useful, I really don't see the pain in typing 'man cmd' to find out what 'cmd' does and what flags it takes. If people aren't prepared to accept that, I suggest that they have chosen the wrong OS, and should think instead about something 'easier' to learn such as MacOS or Windows.
I honestly think that if you want to fight the cause of usability in unix environments, you should be pushing for standardisation of X apps that should all carry some sort of familiarity in them (we've all seen the File/Edit/View standard of Windows). This is less likely to break things from a backwards-compatability front, is going to appeal to those that need it (the sort that will want to work exclusively in a windowed environment), and could still be done before it's too late!