In FreeBSD, you get the filesystem, the kernel, a shell... all developed by the same group of SW engineers.
...and libc. It always seemed strange to me that the Linux C library (glibc) was not developed together with the kernel, since the C library is how most programs interface with the kernel.
Hell yeah. I was going to post something similar, but you put it quite nicely. Computers allow us to do things that are not possible in the physical world -- trying to emulate that world and all of its limitations is a mistake.
Another one's ficl, which is basically an embedable Forth interpreter.
Random bit of trivia: The FreeBSD boot loader has a slightly cut-down version of ficl embedded (and all the boot/kernel loading logic is written in Forth).
But hey, who even said that a locked down 'user' account (not even Power User) would have been any less dangerous? Spyware could still install itself to any folder within that users own Documents and Settings folder. Registry keys under HKEY_USER could still be modified.
Same issue with any OS. Malware could easily run under a user account -- in Windows most doesn't (fails horribly if Program Files or HKLM isn't writable). Of course the added bonus here is that if one did, when your friend logs off and you log back on under your user account, that crap they installed under their account isn't running anymore and can't touch your files.
Plus it's a lot harder for spyware to hide out inside a user profile directory -- no mess of DLLs in System32 to camoflauge itself with...
Well, "beind sendmail" is something that I would definitely put in quotes. IMHO, anything useful that sendmail can do, postfix can do also. The only area where sendmail is more flexibile is for doing things like REALLY complicated and arbitrary routing where you basically have to write a program to make the decisions in real-time. Anything sane postfix should be able to handle with regex rules.
That and things like playing Towers of Hanoi in the sendmail.cf file...
... Makes this software something from depths of hades. Can you enlighten me?
I never said that it was. I don't have a problem with that method of distribution. It's his software, he can require users of it to dance a jig before installing for all I care. I was just pointing out that by most definitions, it's not "Open Source" (see section 3). It's more of a look-but-don't-touch license.
How's postfix's security record? i.e. Can I set up a postfix server, then go on an 18-month holiday and be confident that my box will still be working when I get back (like I can with qmail)?
You can be very confident that it will be. Postfix uses privilege separation, runs as its own user account (not root), and is designed with a chroot environment in mind. It's also very componentized and designed so that a breach in one component can be isolated without a risk to the others. To the best of my knowledge, there has never been a remote code execution vulnerability in Postfix.
The last major security problem was a year ago and was just a DoS possibility. Even qmail has DoS problems. Before the DoS, in 2002 there was a problem that might allow someone to use Postfix to portscan another system (no risk to the system running Postfix). Both of these were in the older 1.1 version. The 2.x series, released in 2002, has never had a security problem bad enough to warrant an advisory for.
The only other thing I could find is djb ranting about a Postfix problem that has been fixed for over 6 years.
qmail was recently forked into something called 'netqmail' that integrates the most popular, bug-fix packages that are out there.
...which can only be distributed as a set of patches against the original code. This means no binary packages, either. djb's license forbids the distribution of modified versions. qmail is not open source. It's actually a lot closer to Microsoft's shared-source license.
Agreed. Postfix has rapidly gone from somebody's pet project for a more secure mail server to one of the best MTAs out there. It's extremely stable and robust, and very flexible and feature-rich too. What you can do with it is second only to sendmail -- and I don't really consider writing code in a turing-complete configuration script to be configuration anyway (and Postfix can already do everything that sendmail's canned m4 macros can).
While it might sound expensive, so is the scan tool. It's about 2k and requires additional updated proms to keep it updated. And yes, like everyting else knowledge and experience costs:-)
What I really want is just the specs to the protocol, so I can hook it to an in-car computer system. A button press to switch from the MP3 player to showing the diagnostic status of the car...
All because Exchange can't be accessed via HTTP internally (here, at any rate), and won't "finger" me.
Sorry to be the bearer of bad news, but the Evolution connector uses WebDAV (i.e. HTTP) to connect through Outlook Web Access. It doesn't speak the MAPI protocol that Outlook itself uses to connect to Exchange. If you can't use the webmail internally, the connector won't work either...
No, I don't remember the name though. Actually, it wasn't even really a book. It wasn't finished yet and we just got printouts of the chapters as they arrived...
I'm talking about the other 999. You do harm to these people's professional computer programming careers by teaching them Java before you teach them C++.
Hear, hear!
I honestly think the first language anybody should learn is assembly. Architecture doesn't really matter.
I'm not joking.
Knowing HOW and WHY higher-level languages work as they do is an invaluable asset that can make anybody's code a lot cleaner and more efficient.
I see code all the time that was obviously written by paper CS-majors who have no clue what the hardware is doing. It makes me really really glad that I learned assembly before C/C++/Java/etc. Had to unlearn some Basic first, but that wasn't too terribly difficult as it is (was) mostly procedural anyway.
I'm not saying to do stupid stuff like trying to out-guess the optimizer (use for(;;) instead of while(1)). I mean things like knowing that insane levels of recursion are likely to cause problems for the stack and some things can be done more efficiently procedurally. It can be invaluable asset to producing good code.
Every time I see his name, I remember Bewilderment + Exposure = Obvious. That was one of the questions on the final (which is composed of questions Schram wrote for the book) for my class. Bastard cost me a few points with his bullshit. I don't even remember what point he was trying to make with that.
The motto of our APCS (C++) class was "Bewilderment + Obvious = Schram".
Though "Schram + Knowledge = Bewilderment" was a popular one too...
LOL, the very first thing we did after getting out of the AP Calculus Test of Death was ask our TEACHER about one of the free response questions that nobody had been able to figure out (she wasn't able to either, BTW, and she's one of the smartest people I know).
Kind of surprised that several of us still got 5s on it -- I guess that nobody else had any clue either. On mine I simply drew a smiley face on the graph paper:)
It doesn't use STL, but rather a set of AP classes which are really a subset of the STL classes. They strip the classes down to the minimum necessary to get the point of the class, and have you use that. You only need a minimal knowledge of C++ to handle it.
Ahhhh, gaypstring and gaypvector. About midway through the year I finally got fed up with them and replaced those classes with my own code. It had the same interface but was about a hundred times more efficient... I don't know if everybody was using the same implementation of those or not, but the one that we were given even had a bug in apstring that cropped up a couple times. Thought I was going crazy until I let the debugger trace into the library itself...
I looked at the prep materials they sent, and realized the test was a complete joke and I wouldn't have to study much at all for it.
Yeah, don't worry. You didn't miss much.
In the class itself we had to use an ancient version of BC++ for DOS for the graphics stuff -- the supplied libraries accessed the VGA framebuffer directly. For most of the stuff though, it was Borland C++ for Windows, the *shudder* Win16 version. I remember having to explain to our study group what a memory model was and why it sometimes mattered for our class even though it was antiquated technology to the rest of the world.
The course material itself was SOOOOOO boring -- and badly written -- it was an obvious quick translation from Pascal. They didn't even introduce classes as something that could be used until over halfway through it (though we all used them anyway). Between assignments I did stuff like re-writing the graphics library so it would work in the Windows version too, and write a multithreading library since DOS didn't have one and trying to thunk the Win32 thread functions didn't work quite right... Though it wasn't as bad as it could have been since my friends and I got together and wrote a NetBEUI (that's what they were running in the lab, ugh) based chat program.
Still, all around it was a good senior-year blowoff class.:)
IE starts up and renders faster than mozilla. Do a benchmark. Hell, my computer starts up faster than mozilla.
IE cheats -- it loads with the OS so when you run it, it just opens a window rather than actually loading. For a fair comparison, use the mozilla "quick launch" feature, which does about the same thing.
Unfortunately, there's no way to disable the IE cheating to make Windoze start up faster or hog less memory. SARCASMThank you Microsoft!/SARCASM
BTW, MS Office tries to use the same trick -- that "Microsoft Office" shortcut it adds to the startup group (without asking) to preload all the DLLs and make it take forever to log in. First thing I kill on all the desktops at work...
What's even worse is that Intel's chips are getting more inefficient with every generation. If they made P3s and P4s at the same clock speed, the P3 would blow the P4 away. Some of the higher-end P3s were faster then the first P4s which had a higher clock speed. Itanium is even worse. It has an insanely long pipeline and a mis-predicted branch can cause it to waste hundreds of clock cycles on a single instruction...
Why do you think a 1.4Ghz Opteron beats a 2.8Ghz P4 in many benchmarks?
Slashdot Mirror
In FreeBSD, you get the filesystem, the kernel, a shell... all developed by the same group of SW engineers.
...and libc. It always seemed strange to me that the Linux C library (glibc) was not developed together with the kernel, since the C library is how most programs interface with the kernel.
You forgot the "Netcraft confirms"...
Hell yeah. I was going to post something similar, but you put it quite nicely. Computers allow us to do things that are not possible in the physical world -- trying to emulate that world and all of its limitations is a mistake.
Another one's ficl, which is basically an embedable Forth interpreter.
Random bit of trivia: The FreeBSD boot loader has a slightly cut-down version of ficl embedded (and all the boot/kernel loading logic is written in Forth).
But hey, who even said that a locked down 'user' account (not even Power User) would have been any less dangerous? Spyware could still install itself to any folder within that users own Documents and Settings folder. Registry keys under HKEY_USER could still be modified.
Same issue with any OS. Malware could easily run under a user account -- in Windows most doesn't (fails horribly if Program Files or HKLM isn't writable). Of course the added bonus here is that if one did, when your friend logs off and you log back on under your user account, that crap they installed under their account isn't running anymore and can't touch your files.
Plus it's a lot harder for spyware to hide out inside a user profile directory -- no mess of DLLs in System32 to camoflauge itself with...
Well, "beind sendmail" is something that I would definitely put in quotes. IMHO, anything useful that sendmail can do, postfix can do also. The only area where sendmail is more flexibile is for doing things like REALLY complicated and arbitrary routing where you basically have to write a program to make the decisions in real-time. Anything sane postfix should be able to handle with regex rules.
.cf file...
That and things like playing Towers of Hanoi in the sendmail
... Makes this software something from depths of hades. Can you enlighten me?
I never said that it was. I don't have a problem with that method of distribution. It's his software, he can require users of it to dance a jig before installing for all I care. I was just pointing out that by most definitions, it's not "Open Source" (see section 3). It's more of a look-but-don't-touch license.
How's postfix's security record? i.e. Can I set up a postfix server, then go on an 18-month holiday and be confident that my box will still be working when I get back (like I can with qmail)?
You can be very confident that it will be. Postfix uses privilege separation, runs as its own user account (not root), and is designed with a chroot environment in mind. It's also very componentized and designed so that a breach in one component can be isolated without a risk to the others. To the best of my knowledge, there has never been a remote code execution vulnerability in Postfix.
The last major security problem was a year ago and was just a DoS possibility. Even qmail has DoS problems. Before the DoS, in 2002 there was a problem that might allow someone to use Postfix to portscan another system (no risk to the system running Postfix). Both of these were in the older 1.1 version. The 2.x series, released in 2002, has never had a security problem bad enough to warrant an advisory for.
The only other thing I could find is djb ranting about a Postfix problem that has been fixed for over 6 years.
qmail was recently forked into something called 'netqmail' that integrates the most popular, bug-fix packages that are out there.
...which can only be distributed as a set of patches against the original code. This means no binary packages, either. djb's license forbids the distribution of modified versions. qmail is not open source. It's actually a lot closer to Microsoft's shared-source license.
Postfix and Exim are my current favorite MTA's.
Agreed. Postfix has rapidly gone from somebody's pet project for a more secure mail server to one of the best MTAs out there. It's extremely stable and robust, and very flexible and feature-rich too. What you can do with it is second only to sendmail -- and I don't really consider writing code in a turing-complete configuration script to be configuration anyway (and Postfix can already do everything that sendmail's canned m4 macros can).
While it might sound expensive, so is the scan tool. It's about 2k and requires additional updated proms to keep it updated. And yes, like everyting else knowledge and experience costs:-)
What I really want is just the specs to the protocol, so I can hook it to an in-car computer system. A button press to switch from the MP3 player to showing the diagnostic status of the car...
Looks like FreeBSD's ports were updated about 5 and a half hours ago.
portupgrade -R subversion
CVS is part of the base system -- it was fixed in all the security branches early this morning.
"Posted speed limit" is the phrase, aka "ignored arbitrary number."
In Texas, "artibrary" is not a word.
All because Exchange can't be accessed via HTTP internally (here, at any rate), and won't "finger" me.
Sorry to be the bearer of bad news, but the Evolution connector uses WebDAV (i.e. HTTP) to connect through Outlook Web Access. It doesn't speak the MAPI protocol that Outlook itself uses to connect to Exchange. If you can't use the webmail internally, the connector won't work either...
Was that C++ for You++?
No, I don't remember the name though. Actually, it wasn't even really a book. It wasn't finished yet and we just got printouts of the chapters as they arrived...
I'm talking about the other 999. You do harm to these people's professional computer programming careers by teaching them Java before you teach them C++.
Hear, hear!
I honestly think the first language anybody should learn is assembly. Architecture doesn't really matter.
I'm not joking.
Knowing HOW and WHY higher-level languages work as they do is an invaluable asset that can make anybody's code a lot cleaner and more efficient.
I see code all the time that was obviously written by paper CS-majors who have no clue what the hardware is doing. It makes me really really glad that I learned assembly before C/C++/Java/etc. Had to unlearn some Basic first, but that wasn't too terribly difficult as it is (was) mostly procedural anyway.
I'm not saying to do stupid stuff like trying to out-guess the optimizer (use for(;;) instead of while(1)). I mean things like knowing that insane levels of recursion are likely to cause problems for the stack and some things can be done more efficiently procedurally. It can be invaluable asset to producing good code.
Every time I see his name, I remember Bewilderment + Exposure = Obvious. That was one of the questions on the final (which is composed of questions Schram wrote for the book) for my class. Bastard cost me a few points with his bullshit. I don't even remember what point he was trying to make with that.
The motto of our APCS (C++) class was "Bewilderment + Obvious = Schram".
Though "Schram + Knowledge = Bewilderment" was a popular one too...
Leon Schram
Schram?! Holy crap, don't tell me the college board is still using his "books".
He wrote the C++ book back in '99 too and it SUCKED. Full of bugs and bad design. Some of the example code didn't even compile.
LOL, the very first thing we did after getting out of the AP Calculus Test of Death was ask our TEACHER about one of the free response questions that nobody had been able to figure out (she wasn't able to either, BTW, and she's one of the smartest people I know).
:)
Kind of surprised that several of us still got 5s on it -- I guess that nobody else had any clue either. On mine I simply drew a smiley face on the graph paper
Yeah, don't worry. You didn't miss much.
In the class itself we had to use an ancient version of BC++ for DOS for the graphics stuff -- the supplied libraries accessed the VGA framebuffer directly. For most of the stuff though, it was Borland C++ for Windows, the *shudder* Win16 version. I remember having to explain to our study group what a memory model was and why it sometimes mattered for our class even though it was antiquated technology to the rest of the world.
The course material itself was SOOOOOO boring -- and badly written -- it was an obvious quick translation from Pascal. They didn't even introduce classes as something that could be used until over halfway through it (though we all used them anyway). Between assignments I did stuff like re-writing the graphics library so it would work in the Windows version too, and write a multithreading library since DOS didn't have one and trying to thunk the Win32 thread functions didn't work quite right... Though it wasn't as bad as it could have been since my friends and I got together and wrote a NetBEUI (that's what they were running in the lab, ugh) based chat program.
Still, all around it was a good senior-year blowoff class.
A longhorn. As in cattle.
IE starts up and renders faster than mozilla. Do a benchmark. Hell, my computer starts up faster than mozilla.
IE cheats -- it loads with the OS so when you run it, it just opens a window rather than actually loading. For a fair comparison, use the mozilla "quick launch" feature, which does about the same thing.
Unfortunately, there's no way to disable the IE cheating to make Windoze start up faster or hog less memory. SARCASMThank you Microsoft!/SARCASM
BTW, MS Office tries to use the same trick -- that "Microsoft Office" shortcut it adds to the startup group (without asking) to preload all the DLLs and make it take forever to log in. First thing I kill on all the desktops at work...
What's even worse is that Intel's chips are getting more inefficient with every generation. If they made P3s and P4s at the same clock speed, the P3 would blow the P4 away. Some of the higher-end P3s were faster then the first P4s which had a higher clock speed. Itanium is even worse. It has an insanely long pipeline and a mis-predicted branch can cause it to waste hundreds of clock cycles on a single instruction ...
Why do you think a 1.4Ghz Opteron beats a 2.8Ghz P4 in many benchmarks?
As time approaches infinity, the probability of a cataclysmic event destroying the entire human race at once approaches 1.
Life if self-defeating because eventually you die anyway. So why bother trying?