Go to www.activestate.com and download ActivePerl for Windows. I've been using Perl for sysadmin scripts on NT and Unix networks for years. It's pretty amusing really. Unix admins take scripting as pretty much of a given. You show the average MCSE the same kind of automation on NT and their head explodes!!:-)
>Debian has tried to merge the installation process with the kernel compilation.
What?? I just installed Debian last week. The installation process does not involve recompiling the kernel in any way. Your on crack.
Debian *does* ask you what modules you want to load, and does expect you to know what hardware you have on your system. But until recently, all Linux installs did.
I will agree that Debian's install is no as easy as the big commercial distro's. No worse than older versions of the commercial distro's though.
>Red Hat & Gnome are sacrificing their ideals >for ?the sake of a quick buck. You can't get >into bed with the Commercial Devil and not die a >little. The Gnome people are always Whoring >themselves in front of Sun, Red Hat & IBM - >especially over the last year.
>IMO, the true inheritors of the FSF's open >source ideals is KDE & the European distros, now >that Troll Tech has GPL'ed the Qt libraries.
Do you have any evidence for that? Or are you just trolling??
In case you haven't noticed, there are just as many commercial companies on the KDE League as on the Gnome Foundation. I don't think this is a bad thing for either desktop. As long as the source is GPL'd...
BTW. The FSF has definately never had any "open source" ideals...:-) They definately have *free software* ideals though...
I ran into the same problem when I used to have @home. My friend and I (both on @home) couldn't ping each other's PC's. A quick traceroute revealed it to be a routing error in @home's network. I tried calling tech support, but got lots of confused idiots trying to tell me to change my proxy settings, re-install Windows, etc.
It might not be an even @home plot against peer-to-peer sharing, it might just be plain old-fashioned incompetence!!!
BTW, I now use Sprint BroadBand direct (wireless). I'm getting much better speed and reliability than I had with @home for about the same price!
Although it isn't point and drool.. The Gnu CFENGINE (http://www.iu.hioslo.no/cfengine/) allows easy management of 1000's of systems. It is scripting based, but the scripting language is very, very, very simple and easy to learn.
I believe PIKT (http://pikt.uchicago.edu/pikt/) is another similar app..
As for one click managment, even in the NT world I've never seen a decent system for managing large numbers of systems that is totally point and click. Even Microsoft System Managment Server requires some scripting. (as well as requiring you to set up SQL server).
>To scan every message for viruses, you'll need a >third party product, which will likely reduce >performance and stability to the point that you >have to turn it off
Perhaps you picked the wrong antivirus software?? I've run Trend Scanmail (www.trend.com) and really like it. No noticable impact on the performance or stability of the Exchange server..
Of course, I'm definately not saying that Exchange is as stable or robust as a Unix Sendmail server... Just that if your careful, you can make it workable. Not exactly a ringing endorsement.... but there ya go..
>Somebody is likely to suggesting hiring an MCSE >to run it. This person will get paid too much, >not know RFC 821 or 822, or anything remotely >technical, yet they'll drain the company of >$80k/year, which could be better spent on more >beer for the developers
In most areas of the country that's pretty high for an MCSE, unless they have other qualifications (other certs, lots of experience, etc). MCSE's pretty much grow on trees these days.
The MCSE tests are pretty easy, so I don't think it's at all unreasonable to want anyone administering an NT server to have/get an MCSE. It's really a pretty entry level cert.
But more to the point, your argument there is a bit of a red herring. All of the Unix admins I know get paid much better than most MCSE's and NT admins. (One of the reasons I love Unix!:-)
It looks like almost everyone in this message thread keeps bringing up the e-mail virus issue. Unfortunately, none of them know what the hell their talking about. I've run large Sendmail and large Exchange servers. Contrary to what appears to be common belief here on Slashdot, running Exchange *DOES NOT* necessarily increase your risk from e-mail viruses..
There has never been an e-mail virus that effects Exchange server.. Email viruses are entirely a client side problem. Specifically, there are two factors that *may* increase your vulnerability to e-mail viruses:
1) Migrating all of your users to Outlook on Windows. Although Exchange will do vanilla POP/IMAP/SMTP (thereby allowing non-Outlook users to get and send mail) I'm assuming that your management wants everyone to use the shared folders, calendering, etc of Outlook. (Otherwise what is the point of moving to Exchange??). Outlook is *THE* target platform for e-mail virus writers. If a significant portion of your users are currently using something else for an e-mail client, moving them all to Outlook will DEFINATELY increase your vulnerability to e-mail viruses. However, if most of your users are already on Outlook or Outlook Express, you've already got that level of vulnerability. Whatever you run as an e-mail server isn't going to help or hurt.
2) In addition to a normal personal address book, Outlook/Exchange users all have a global address book that lists the e-mail address's etc. of every account on the Exchange site, as well as distribution lists, etc. This is one of the nice usability features managers love about the Outlook/Exchange combo, but is also one of the things that makes it a lightning rod for e-mail worms like ILOVEYOU. ILOVEYOU would distribute itself to everyone in your address book. For Exchange sites, that meant EVERYONE in the damn company. Each of these systems would then AGAIN send it to EVERYONE, etc. Note however, that this again isn't an Exchange specific problem. All Exchange sites will have this problem, but so will any sites that use another method for creating a company wide address book in your e-mail program. This includes using LDAP, or just manually/automatically, etc pushing out a new address book file to all desktops.
As you can see, Exchange *ITSELF* isn't vulnerable to e-mail viruses, and does not *necessarily* increase your vulnerability to e-mail viruses. An Exchange setup does, however, encourage you to create a nearly worst-case possible client-side e-mail setup. This is definately bad, but you might *already* have this level of vulnerability no matter what you're running as a mail server. It's the client site setup that is the problem...
In any case, no matter what your mail server is, you should definately be running an antivirus software on the server to scan all incoming and outgoing e-mails. Trend Micro makes a good one for both Exchange and Sendmail, and I believe Norton does as well. Running a good antivirus software is the ONLY effective, realistic, server side measure to protect your users from e-mail viruses.
You might definately want to find out *WHY* management want you to move to Exchange. Is it just because it's the new buzz word around the CEO clubs? Or do they want Exchange specific features??
If you do have to go with Exchange, like most NT "solutions" it can be made reasonably reliable, but that means throwing quite a bit of hardware at it. Plan on a lot more hardware to run your Exchange server than you currently use for your Sendmail server. For just POP/IMAP/SMTP, Exchange is unbelievably bloated. It's size only really makes sense if you look at all the other stuff it does...
Keep in mind when sizing your server that unlike a Unix Sendmail server, a lack of resources (cpu, ram, etc) won't just make the system run slower, it will dramatically decrease the stability of the system. Also, an Exchange server (unless it's for only a VERY small number of users) should definately *not* be running any other services (it should not even be a domain controller). This seems strange to most Unix admins, who are used to being able to push thier boxes to high levels of utilization, but is unfortunately pretty standard practice in the MS world.
You might also be able to run a hybrid mail setup. Use Exchange in the office for user mailboxes, scheduling, etc. Leave your existing Sendmail servers up as your "mail hubs".. That way your exchange servers can remain behind your firewall, with the more easily securable Unix systems getting the exposure in your DMZ. Also, if your Exchange servers should take a noze dive, Internet mail will just pile up in the mail queue on the Unix servers until you can get things straightened out.
Or you could just ran an antivirus program (like Norton or Trend) on the Exchange server.. they scan all incoming and outgoing e-mails.. works great..
Re:Sealand may be the answer
on
Scour is Dead
·
· Score: 1
Scour would have to re-locate their entire company to Sealand. It doesn't matter if the server is in the US or in Sealand. As long as the company itself or its officers are in the US they can still be procecuted for violating US laws. Also, any company property (ie. money) located in the US can be confiscated by a court.
>I whole-heartedly believe that the people who >think those who make more should pay a higher >percentage of their income are completely >mathematically illiterate
Those who don't understand progressive taxation are historically illiterate.
Progressive taxation, property taxes, and inheritance taxes were all designed to by the U.S. founding fathers to prevent wealth from accumulting in the hands of a few families (as was common in Europe at the time). Children, grandchildren, etc of those wealthy families would live off of the inherited wealth of thier fathers without ever having to do anything to "earn" a living. They lived a life of luxury and privilage, and eventually began to think of themselves as a different "kind of people" from the masses.
Progressive taxation, property taxes, and inheritence taxes were designed to prevent the formation, or at least the long term continuance, of this kind of aristocracy. Large accumulations of wealth and large estates would be difficult to maintain without continuing to work/invest and earn more money.
It may well be social engineering, but whether you like it or not this country was *founded* on social engineering. The United States has always been a giant social experiment.
Seriously, I can't stand Tcl, but I don't go off on huge rants about how much it sucks. Some people like it. Some people like Pepsi (which I also can't stand).
My drinking a soda you don't like would probably never send you into rant mode about how much my soda sucks and how cool yours is. But if, god forbid, I use a programming language you hate......
Geesh people.. It's not the end of the frigging world...
Although the examples thrown out in the article (writing code in Perl that looks like Python, Java, etc) is kind of lame, I think the idea behind the feature is to allow the easy creation of "little" languages. For example, if your project requires more extensibility than just a static config file, but you'd like to make the extensibility simpler than actually modifying the code for the program. You could create a simplistic configuration language.
For example, a future version of MRTG (which is already written in Perl), could have a simple configuration scripting language to automate monitoring.
Or a Perl based game engine (yuck??) could have a little language that defines the rules of the game.
In theory you could do this now with Perl, but it would require you to write an interpreter.
I'm surprised I haven't seen an messages in here about the REALLY funny and ironic part of that story.... It was an *email trojan* that allowed the crackers to get access. The exact weakness that everyone has been bitching at Microsoft for months about (ever since ILOVEYOU and MELISSA).
This *should* (but probably won't) make it clear to everyone why email trojans are really dangerous! (particularly for MS-OS's!!
>I agree to a certain extent. One thing that >pisses me off is watching all of the duplication >of effort in the Open Source Community. For >example, if all of the groups out there working >on accounting projects would combine forces and >agree on technologies (Perl, Apache, PostgreSQL, >RPM, etc.), we could soon have an Open Source, >modular, scalable accounting package comparable >to something like Great Plains Dynamics
No, we would have far fewer people working on ANY kind of open source accounting package. With some exceptions, people working on open source packages aren't doing it because they have to. They aren't doing it to please you. They're doing it because they WANT TO!!!
Aside from that, your argument is still completely erronious. Open source packages benefit from competition with other open source packages. Each team can make different design and implimentation decisions and can learn from the successes and mistakes of the other team. A good example: Much of the push for "hardened" and "secure" linux distributions has come from comparisons to OpenBSD.
There is still a big difference between a profit driven company that has the wisdom to contribute back to the community it depends upon for it's very existence, and a profit driven company that doesn't. The previous poster's point was not the TurboLinux is a profit driven company (not in dispute), but that TurboLinux hasn't contributed much, if any, code back to the Linux community.
Most likely the Universities have realized that banning Napster *because of its copywrite violation potential* is not in thier best interest. Banning Napster would be easy, but it doesn't take a rocket scientist to realize that Napster will not be the last protocol/service/etc that someone will ask them to ban. Eventually, that could lead to a huge headache for the Univerities, trying to ban dozens of protocols, URLs, etc.
Even university administrators who oppose Napster would certainly rather have Napster and the RIAA fight it out between themselves and leave third parties (like the universities) out of it.
So these refusals don't necessarily mean these universities support Napster, or won't ban it for other reasons (bandwidth). I think it mostly means they don't want to be dragged into the battle!
That the above comment was ever moderated up is sad... Until recently, free software was never about "mindshare" and IPO's. It was about programmers creating things they loved and believed in. RMS didn't start the GNU project because he though it would someday have a huge "mindshare". Linus, Alan Cox, and the other early Linux developers didn't start hacking on Linux for "mindshare". They did it because they believe in the project, and because it was something they loved tinkering with.
Now I'm not saying that I'm unhappy about the commercial success that some companies are having with free software. I love the face that a greater percentage of the servers I work with are Linux servers.
But free software isn't only about things like mindshare and marketshare. Sometimes it's just about a project you love and want to hack on. And it doesn't really matter what the "installed base" of your program is. It matters how much you love writing it.
As long a there is a free software movement, as long as hackers write code for the love of learning, and the love of the challenge, there will still be room for more free OS's and other free software projects. The day there isn't, is the day that the "free" software movement has truely died.
You just being paranoid. I've signed plenty of NDA's and then not taken the job. Many high tech companies, and almost all ".com" companies will want you to sign a NDA for an interview. If the interview is going well at all, they will want to talk to you about the company, what it does, where it's going, what kind of project or projects they expect to be initially using you for, etc.
Now days most companies will also tell you quite a bit about their financing situation too. I thought that was a bit strange at first, but I found out that a lot of job applicants ask about that. Everyone seems to what to know when the bit IPO cash out is going to be! (don't get me started on how fucked up that is).
Anyway, don't sweat it. IANAL, but I'm quite certain there is nothing in those NDA's that could keep you from working for a competitor. They just don't want you running out of the interview and telling everyone on Slashdot what their up too!!!!:-)
BTW. If your really worried, I'm certain you *could* ask a lawyer what the impact of signing an NDA in an interview would be.
You don't need a seperate machine to use IPCHAINS. Most people are more familier with using IPCHAINS to construct forwarding rules for a dedicated firewall. But, you can also use IPCHAINS to create input rules to secure your workstation (or server).
It's very similar to creating forwarding rules, but instead of accepting or denying packets forwarded through your box, you accept or deny packets directed to your box.
My favorite is the easter egg in the Google search engine. For your search query, type in "more evil than the devil himself". Then click "I feel lucky".....:-)
If your running an RPM based system, you can use rpm -qf will tell you what package a file came from. rpm -V will verify the package and tell you any files installed by that package have been altered or removed.
I'm certain DPKG has similar capabilities, I just don't remember the commandline offhand.
Obviously this doesn't help with a completely custom compiled system. Most business do not "roll thier own" distro for their servers. They buy a commercial Linux distro (or use Debian). It saves time, and gives other advantages such as the above verify command.
(BTW. If you do a quick search around the net you can find "root kits" for most unixes (commercial and open source) that will replace common system binaries (passwd, ls , ps, etc) will "hacked" version. Similar type programs exist for most non-Unix systems (i.e. Back Oriface 2000).)
Slashdot Mirror
I would disagree with that. I've met a lot of supposed "computer professionals" who know all the jargon but have no idea what it means.
>Only problem is, how do I run Perl in Windows?
:-)
Go to www.activestate.com and download ActivePerl for Windows. I've been using Perl for sysadmin scripts on NT and Unix networks for years. It's pretty amusing really. Unix admins take scripting as pretty much of a given. You show the average MCSE the same kind of automation on NT and their head explodes!!
>Debian has tried to merge the installation process with the kernel compilation.
What?? I just installed Debian last week. The installation process does not involve recompiling the kernel in any way. Your on crack.
Debian *does* ask you what modules you want to load, and does expect you to know what hardware you have on your system. But until recently, all Linux installs did.
I will agree that Debian's install is no as easy as the big commercial distro's. No worse than older versions of the commercial distro's though.
>Red Hat & Gnome are sacrificing their ideals >for ?the sake of a quick buck. You can't get >into bed with the Commercial Devil and not die a >little. The Gnome people are always Whoring >themselves in front of Sun, Red Hat & IBM - >especially over the last year.
:-) They definately have *free software* ideals though...
>IMO, the true inheritors of the FSF's open >source ideals is KDE & the European distros, now >that Troll Tech has GPL'ed the Qt libraries.
Do you have any evidence for that? Or are you just trolling??
In case you haven't noticed, there are just as many commercial companies on the KDE League as on the Gnome Foundation. I don't think this is a bad thing for either desktop. As long as the source is GPL'd...
BTW. The FSF has definately never had any "open source" ideals...
I ran into the same problem when I used to have @home. My friend and I (both on @home) couldn't ping each other's PC's. A quick traceroute revealed it to be a routing error in @home's network. I tried calling tech support, but got lots of confused idiots trying to tell me to change my proxy settings, re-install Windows, etc.
It might not be an even @home plot against peer-to-peer sharing, it might just be plain old-fashioned incompetence!!!
BTW, I now use Sprint BroadBand direct (wireless). I'm getting much better speed and reliability than I had with @home for about the same price!
Although it isn't point and drool.. The Gnu CFENGINE (http://www.iu.hioslo.no/cfengine/) allows easy management of 1000's of systems. It is scripting based, but the scripting language is very, very, very simple and easy to learn.
I believe PIKT (http://pikt.uchicago.edu/pikt/) is another similar app..
As for one click managment, even in the NT world I've never seen a decent system for managing large numbers of systems that is totally point and click. Even Microsoft System Managment Server requires some scripting. (as well as requiring you to set up SQL server).
>To scan every message for viruses, you'll need a >third party product, which will likely reduce >performance and stability to the point that you >have to turn it off
:-)
Perhaps you picked the wrong antivirus software?? I've run Trend Scanmail (www.trend.com) and really like it. No noticable impact on the performance or stability of the Exchange server..
Of course, I'm definately not saying that Exchange is as stable or robust as a Unix Sendmail server... Just that if your careful, you can make it workable. Not exactly a ringing endorsement.... but there ya go..
>Somebody is likely to suggesting hiring an MCSE >to run it. This person will get paid too much, >not know RFC 821 or 822, or anything remotely >technical, yet they'll drain the company of >$80k/year, which could be better spent on more >beer for the developers
In most areas of the country that's pretty high for an MCSE, unless they have other qualifications (other certs, lots of experience, etc). MCSE's pretty much grow on trees these days.
The MCSE tests are pretty easy, so I don't think it's at all unreasonable to want anyone administering an NT server to have/get an MCSE. It's really a pretty entry level cert.
But more to the point, your argument there is a bit of a red herring. All of the Unix admins I know get paid much better than most MCSE's and NT admins. (One of the reasons I love Unix!
It looks like almost everyone in this message thread keeps bringing up the e-mail virus issue. Unfortunately, none of them know what the hell their talking about. I've run large Sendmail and large Exchange servers. Contrary to what appears to be common belief here on Slashdot, running Exchange *DOES NOT* necessarily increase your risk from e-mail viruses..
There has never been an e-mail virus that effects Exchange server.. Email viruses are entirely a client side problem. Specifically, there are two factors that *may* increase your vulnerability to e-mail viruses:
1) Migrating all of your users to Outlook on Windows. Although Exchange will do vanilla POP/IMAP/SMTP (thereby allowing non-Outlook users to get and send mail) I'm assuming that your management wants everyone to use the shared folders, calendering, etc of Outlook. (Otherwise what is the point of moving to Exchange??). Outlook is *THE* target platform for e-mail virus writers. If a significant portion of your users are currently using something else for an e-mail client, moving them all to Outlook will DEFINATELY increase your vulnerability to e-mail viruses. However, if most of your users are already on Outlook or Outlook Express, you've already got that level of vulnerability. Whatever you run as an e-mail server isn't going to help or hurt.
2) In addition to a normal personal address book, Outlook/Exchange users all have a global address book that lists the e-mail address's etc. of every account on the Exchange site, as well as distribution lists, etc. This is one of the nice usability features managers love about the Outlook/Exchange combo, but is also one of the things that makes it a lightning rod for e-mail worms like ILOVEYOU. ILOVEYOU would distribute itself to everyone in your address book. For Exchange sites, that meant EVERYONE in the damn company. Each of these systems would then AGAIN send it to EVERYONE, etc. Note however, that this again isn't an Exchange specific problem. All Exchange sites will have this problem, but so will any sites that use another method for creating a company wide address book in your e-mail program. This includes using LDAP, or just manually/automatically, etc pushing out a new address book file to all desktops.
As you can see, Exchange *ITSELF* isn't vulnerable to e-mail viruses, and does not *necessarily* increase your vulnerability to e-mail viruses. An Exchange setup does, however, encourage you to create a nearly worst-case possible client-side e-mail setup. This is definately bad, but you might *already* have this level of vulnerability no matter what you're running as a mail server. It's the client site setup that is the problem...
In any case, no matter what your mail server is, you should definately be running an antivirus software on the server to scan all incoming and outgoing e-mails. Trend Micro makes a good one for both Exchange and Sendmail, and I believe Norton does as well. Running a good antivirus software is the ONLY effective, realistic, server side measure to protect your users from e-mail viruses.
You might definately want to find out *WHY* management want you to move to Exchange. Is it just because it's the new buzz word around the CEO clubs? Or do they want Exchange specific features??
If you do have to go with Exchange, like most NT "solutions" it can be made reasonably reliable, but that means throwing quite a bit of hardware at it. Plan on a lot more hardware to run your Exchange server than you currently use for your Sendmail server. For just POP/IMAP/SMTP, Exchange is unbelievably bloated. It's size only really makes sense if you look at all the other stuff it does...
Keep in mind when sizing your server that unlike a Unix Sendmail server, a lack of resources (cpu, ram, etc) won't just make the system run slower, it will dramatically decrease the stability of the system. Also, an Exchange server (unless it's for only a VERY small number of users) should definately *not* be running any other services (it should not even be a domain controller). This seems strange to most Unix admins, who are used to being able to push thier boxes to high levels of utilization, but is unfortunately pretty standard practice in the MS world.
You might also be able to run a hybrid mail setup. Use Exchange in the office for user mailboxes, scheduling, etc. Leave your existing Sendmail servers up as your "mail hubs".. That way your exchange servers can remain behind your firewall, with the more easily securable Unix systems getting the exposure in your DMZ. Also, if your Exchange servers should take a noze dive, Internet mail will just pile up in the mail queue on the Unix servers until you can get things straightened out.
Good luck!!!
Or you could just ran an antivirus program (like Norton or Trend) on the Exchange server.. they scan all incoming and outgoing e-mails.. works great..
Scour would have to re-locate their entire company to Sealand. It doesn't matter if the server is in the US or in Sealand. As long as the company itself or its officers are in the US they can still be procecuted for violating US laws. Also, any company property (ie. money) located in the US can be confiscated by a court.
>I whole-heartedly believe that the people who >think those who make more should pay a higher >percentage of their income are completely >mathematically illiterate
Those who don't understand progressive taxation are historically illiterate.
Progressive taxation, property taxes, and inheritance taxes were all designed to by the U.S. founding fathers to prevent wealth from accumulting in the hands of a few families (as was common in Europe at the time). Children, grandchildren, etc of those wealthy families would live off of the inherited wealth of thier fathers without ever having to do anything to "earn" a living. They lived a life of luxury and privilage, and eventually began to think of themselves as a different "kind of people" from the masses.
Progressive taxation, property taxes, and inheritence taxes were designed to prevent the formation, or at least the long term continuance, of this kind of aristocracy. Large accumulations of wealth and large estates would be difficult to maintain without continuing to work/invest and earn more money.
It may well be social engineering, but whether you like it or not this country was *founded* on social engineering. The United States has always been a giant social experiment.
If you hate Perl.. Don't use it. Geesh!!
Seriously, I can't stand Tcl, but I don't go off on huge rants about how much it sucks. Some people like it. Some people like Pepsi (which I also can't stand).
My drinking a soda you don't like would probably never send you into rant mode about how much my soda sucks and how cool yours is. But if, god forbid, I use a programming language you hate......
Geesh people.. It's not the end of the frigging world...
Although the examples thrown out in the article (writing code in Perl that looks like Python, Java, etc) is kind of lame, I think the idea behind the feature is to allow the easy creation of "little" languages. For example, if your project requires more extensibility than just a static config file, but you'd like to make the extensibility simpler than actually modifying the code for the program. You could create a simplistic configuration language.
For example, a future version of MRTG (which is already written in Perl), could have a simple configuration scripting language to automate monitoring.
Or a Perl based game engine (yuck??) could have a little language that defines the rules of the game.
In theory you could do this now with Perl, but it would require you to write an interpreter.
I'm surprised I haven't seen an messages in here about the REALLY funny and ironic part of that story.... It was an *email trojan* that allowed the crackers to get access. The exact weakness that everyone has been bitching at Microsoft for months about (ever since ILOVEYOU and MELISSA).
This *should* (but probably won't) make it clear to everyone why email trojans are really dangerous! (particularly for MS-OS's!!
I can't believe you didn't have a cable-lock on the laptop. You left a laptop with confidential information unsecured *overnight*?????
DNS does have a hierarchial structure. You can hardly tell from the way DNS is currently used.... but it is completely hierarchial.
>I agree to a certain extent. One thing that >pisses me off is watching all of the duplication >of effort in the Open Source Community. For >example, if all of the groups out there working >on accounting projects would combine forces and >agree on technologies (Perl, Apache, PostgreSQL, >RPM, etc.), we could soon have an Open Source, >modular, scalable accounting package comparable >to something like Great Plains Dynamics
No, we would have far fewer people working on ANY kind of open source accounting package. With some exceptions, people working on open source packages aren't doing it because they have to. They aren't doing it to please you. They're doing it because they WANT TO!!!
Aside from that, your argument is still completely erronious. Open source packages benefit from competition with other open source packages. Each team can make different design and implimentation decisions and can learn from the successes and mistakes of the other team. A good example: Much of the push for "hardened" and "secure" linux distributions has come from comparisons to OpenBSD.
Manually assigning IP's for servers is definately the way to go.
However, if your manually assigning IP's to your workstations you've either got a small network, or your a time-wasting moron.
There is still a big difference between a profit driven company that has the wisdom to contribute back to the community it depends upon for it's very existence, and a profit driven company that doesn't. The previous poster's point was not the TurboLinux is a profit driven company (not in dispute), but that TurboLinux hasn't contributed much, if any, code back to the Linux community.
Most likely the Universities have realized that banning Napster *because of its copywrite violation potential* is not in thier best interest. Banning Napster would be easy, but it doesn't take a rocket scientist to realize that Napster will not be the last protocol/service/etc that someone will ask them to ban. Eventually, that could lead to a huge headache for the Univerities, trying to ban dozens of protocols, URLs, etc.
Even university administrators who oppose Napster would certainly rather have Napster and the RIAA fight it out between themselves and leave third parties (like the universities) out of it.
So these refusals don't necessarily mean these universities support Napster, or won't ban it for other reasons (bandwidth). I think it mostly means they don't want to be dragged into the battle!
That the above comment was ever moderated up is sad... Until recently, free software was never about "mindshare" and IPO's. It was about programmers creating things they loved and believed in. RMS didn't start the GNU project because he though it would someday have a huge "mindshare". Linus, Alan Cox, and the other early Linux developers didn't start hacking on Linux for "mindshare". They did it because they believe in the project, and because it was something they loved tinkering with.
Now I'm not saying that I'm unhappy about the commercial success that some companies are having with free software. I love the face that a greater percentage of the servers I work with are Linux servers.
But free software isn't only about things like mindshare and marketshare. Sometimes it's just about a project you love and want to hack on. And it doesn't really matter what the "installed base" of your program is. It matters how much you love writing it.
As long a there is a free software movement, as long as hackers write code for the love of learning, and the love of the challenge, there will still be room for more free OS's and other free software projects. The day there isn't, is the day that the "free" software movement has truely died.
Just my 2 cents worth.
You just being paranoid. I've signed plenty of NDA's and then not taken the job. Many high tech companies, and almost all ".com" companies will want you to sign a NDA for an interview. If the interview is going well at all, they will want to talk to you about the company, what it does, where it's going, what kind of project or projects they expect to be initially using you for, etc.
:-)
Now days most companies will also tell you quite a bit about their financing situation too. I thought that was a bit strange at first, but I found out that a lot of job applicants ask about that. Everyone seems to what to know when the bit IPO cash out is going to be! (don't get me started on how fucked up that is).
Anyway, don't sweat it. IANAL, but I'm quite certain there is nothing in those NDA's that could keep you from working for a competitor. They just don't want you running out of the interview and telling everyone on Slashdot what their up too!!!!
BTW. If your really worried, I'm certain you *could* ask a lawyer what the impact of signing an NDA in an interview would be.
You don't need a seperate machine to use IPCHAINS. Most people are more familier with using IPCHAINS to construct forwarding rules for a dedicated firewall. But, you can also use IPCHAINS to create input rules to secure your workstation (or server).
It's very similar to creating forwarding rules, but instead of accepting or denying packets forwarded through your box, you accept or deny packets directed to your box.
My favorite is the easter egg in the Google search engine. For your search query, type in "more evil than the devil himself". Then click "I feel lucky"..... :-)
If your running an RPM based system, you can use rpm -qf will tell you what package a file came from. rpm -V will verify the package and tell you any files installed by that package have been altered or removed.
I'm certain DPKG has similar capabilities, I just don't remember the commandline offhand.
Obviously this doesn't help with a completely custom compiled system. Most business do not "roll thier own" distro for their servers. They buy a commercial Linux distro (or use Debian). It saves time, and gives other advantages such as the above verify command.
(BTW. If you do a quick search around the net you can find "root kits" for most unixes (commercial and open source) that will replace common system binaries (passwd, ls , ps, etc) will "hacked" version. Similar type programs exist for most non-Unix systems (i.e. Back Oriface 2000).)