If I understand correctly, the DMZ feature won't be so useful until multiple IPs are allowed on RED.
Currently you may only use one "official" IP address (that is the IP address of the RED interface) to "pinhole" the DMZ. That means you may have just one web server on port 80, or just one mail server on port 25 and so on.
Of course you still may be able to serve multiple domains with name-based virtual hosts and such, but I think that multiple IPs on RED is a very desirable feature indeed (planned for 0.2 - yuck!). This is a strong limitation for anything a little bigger than a SOHO.
It shouldn't be hard to implement either, just allow interface aliases for the RED interface. Astaro does that very nicely. And that may also overcome the three interfaces limit...
What I REALLY would like to see in the future is some "security level" setting a-la Cisco PIX. Each interface is assigned a security level, with 100 being the internal LAN (GREEN in SmoothWall/IPCop speak) and 0 the external link to the Internet (RED). Each additional interface is given a security level inbetween. Each interface is allowed by default to talk to an interface having a lesser security level. Interfaces having the same security level may NEVER talk to each other. All of this, of course, unless otherwise stated. I think this is quite smart and simplifies policy design, it may be good to have at least as an option.
Micro$oft isn't missing anything. They have started watching OSS for real. They admitted Linux is top threat.
The signals are clear. They're after us. Like it or not, this is open war, and Microsoft is going to play _dirty_. It's the only game they can play, and it's just started. They have proved before to be masters in the Art of FUD. Go figure what they'll come up with...!
Of course M$ likes BSD. Of course M$ complains about GPL. They will do whatever they can to sh1t on it, you can count on that!
Yesterday we were just buzzing around M$. Now M$ is after us. High time for the FS community to realize this and switch strategies. Cleaner code may not be enough this time.
I believe the Free Software community should seriously look into the way OS X behaves.
From what (not so much indeed) I was able to figure out, it has the most advanced library/package management system to date (I see much was borrowed from NeXT).
From a very fast scan, looks to me that aide lacks the 'networking' feature, which I think is basic in a product of this kind (even if the authors plan on adding it). Couldn't an attacker just rebuild the database after mangling your system files? How are you supposed to protect the hash database if not storing it elsewhere?
In this sense, this stuff seems better than aide. I don't think that using a custom port/protocol was the right choice anyway. I'd better stick with ssh/scp for obvious reasons.
When NeXTStep came out, I thought its name was really proper. THAT was the OS of my life. Crafted to rock. Designed to grow.
When Apple bought NeXT, we all were eager to see the Next Big Thing. But (almost) nothing developed.
Now it looks like something's REALLY coming out of the hat. But the question is: seems to me that NeXT didn't make it at first but it's getting its way now. Don't you feel this way? And if you do, what are the reasons in your opinion? I always believed NeXTStep was just too smart to cope with 80's hardware, for a start. But I see there's other issues, like the REAL NEED for a gentler, stabler OS.
Another thing I'd like people to focus on with more precision is that OSX IS NOT Free/OSS software. Darwin is (and in questionable terms AFAIK).
Still I'm wetting my pants. And whether Apple succeeds or not, one thing is clear IMO: Steve Jobs IS The Man!
A friend of mine borrowed me some Metallica tapes (hey, he owned the LP records!:-) looong time ago, I believe it was the first two albums (Kill'em all and Master of Puppets? Is that right?). I don't like your genre at all, but being an avid listener and (sort of) musician myself, I grew respect of your work from listening those tapes. This is just to point out that I think my opinions are not biased (and may be easily edited off the question:-).
I feel very debated between the (right IMO) point of view of those feeling they are paying too much for a CD, and the fact that artists have to be payed for their hard work. As some advise, electronic distribution may be an answer, but it happens to be not just around the corner for various reasons. And stealing (yeah, stealing; be honest, that is more or less what mp3 is about - I'm not talking about napster) may or may not be a solution.
The question is: don't you think that adding value to your product may be an answer to the problem?
I can say I am a fan of Steely Dan. I knew Donald Fagen from having his solo records, but I never knew SD. Then I downloaded mp3s, found that I liked SD as much, and actually bought the CDs. I recently downloaded high quality mp3s of the new record. I could have mastered those to a very close resemblance of the original recording, but I haven't and I won't. I will buy the record as soon as I have occasion, and you can bet on it.
Want to know why? I couldn't settle for less than the higher CD quality, cover art, and (why not?) joyful feeling that Fagen's usual frown will eventually be displaced from a nice smile when he sees sales reports, even if for a little while, I suppose:-). And not to mention the feeling of having done the right thing, and that this added to the probability there will be a shining new release soon.
I admit I'm biased on the quality/art topics, but what if the CD included videos? Yeah, you can pass that on the net as well, but what the heck. And what about promotions? Ticket discounts? What else?
Your fans are demanding more than the mere music. That's why they buy gadgets (upon which you also profit). Want to sell more CDs? Try and make them more desirable than mp3 files. You'll get only the good part of the mp3 (or tape, or whatever else) trading, and you can't deny there is a good part. It worked for Steely Dan, it will work for you.
Don't get me wrong. You HAVE to get your bucks for what you do, and that is a fact, and you ought not to be forced to do anything but release your music, that's what you're there for. You have the right to defend your property. But this may be a viable "workaround", until some decent solution comes along, stopping your listeners and fans from being ripped as much as (and probably more than) you are.
I also think that a lot of major issues are not being covered when talking about FS/OSS. The one you address (the developers skill or sense of design) is one of the most important IMO.
I believe that, while we have some major achievements in the "code" ground, we still lack a whole lot in the "logistic" area. Given the very nature of free software, I believe this happens to be far more important than the code itself. And here may be where we lack compared to "suits", one thing that I think we do not want and can not afford.
I see that something has been done in that respect in the last times, but I believe it's too little compared to the importance of the issue. Which is no easy one, though. I'm currently working on a piece about this that I hope I'll be able to submit to the web soon.
Slashdot Mirror
"The name of the game is Microsoft wins and you lose,"
They'd deserve to be bashed just for this, even if nothing else from them had ever been annoying.
History teaches to the sage they'd better not be bully.
If I understand correctly, the DMZ feature won't be so useful until multiple IPs are allowed on RED.
Currently you may only use one "official" IP address (that is the IP address of the RED interface) to "pinhole" the DMZ. That means you may have just one web server on port 80, or just one mail server on port 25 and so on.
Of course you still may be able to serve multiple domains with name-based virtual hosts and such, but I think that multiple IPs on RED is a very desirable feature indeed (planned for 0.2 - yuck!). This is a strong limitation for anything a little bigger than a SOHO.
It shouldn't be hard to implement either, just allow interface aliases for the RED interface. Astaro does that very nicely. And that may also overcome the three interfaces limit...
What I REALLY would like to see in the future is some "security level" setting a-la Cisco PIX. Each interface is assigned a security level, with 100 being the internal LAN (GREEN in SmoothWall/IPCop speak) and 0 the external link to the Internet (RED). Each additional interface is given a security level inbetween. Each interface is allowed by default to talk to an interface having a lesser security level. Interfaces having the same security level may NEVER talk to each other. All of this, of course, unless otherwise stated. I think this is quite smart and simplifies policy design, it may be good to have at least as an option.
I didn't see anybody at +3 making the analogy to the GPL vs. the BSD-like licenses.
Welcome to the fantastic world of /. moderation (see my previous post at 0)
don't want to raise old flames, but...
doesn't this sound as a good argument in favour of the GPL?
Yeah.
Cure your cancer with MS DeCancer 2004.
You'll just need the costly upgrades and to reboot yourself once in a while.
"cure this"
Micro$oft isn't missing anything. They have started watching OSS for real. They admitted Linux is top threat.
The signals are clear. They're after us. Like it or not, this is open war, and Microsoft is going to play _dirty_. It's the only game they can play, and it's just started. They have proved before to be masters in the Art of FUD. Go figure what they'll come up with...!
Of course M$ likes BSD. Of course M$ complains about GPL. They will do whatever they can to sh1t on it, you can count on that!
Yesterday we were just buzzing around M$. Now M$ is after us. High time for the FS community to realize this and switch strategies. Cleaner code may not be enough this time.
Am I oversensitive? My $.02 BTW...
why?
This is what I mean, it already passed on Slashdot when Mac OS X DR4 was released and ArsTechnica made a nice article on that.
http://developer.apple.com/techpubs/corefoundation /BundleServices/CFBundleServices/index.h tml
I believe the Free Software community should seriously look into the way OS X behaves.
From what (not so much indeed) I was able to figure out, it has the most advanced library/package management system to date (I see much was borrowed from NeXT).
Seems to me this is the way to go, full stop.
Didn't know about aide, so I checked.
From a very fast scan, looks to me that aide lacks the 'networking' feature, which I think is basic in a product of this kind (even if the authors plan on adding it). Couldn't an attacker just rebuild the database after mangling your system files? How are you supposed to protect the hash database if not storing it elsewhere?
In this sense, this stuff seems better than aide. I don't think that using a custom port/protocol was the right choice anyway. I'd better stick with ssh/scp for obvious reasons.
Let alone the licensing...
Let me ask you one thing.
When NeXTStep came out, I thought its name was really proper. THAT was the OS of my life. Crafted to rock. Designed to grow.
When Apple bought NeXT, we all were eager to see the Next Big Thing. But (almost) nothing developed.
Now it looks like something's REALLY coming out of the hat. But the question is: seems to me that NeXT didn't make it at first but it's getting its way now. Don't you feel this way? And if you do, what are the reasons in your opinion? I always believed NeXTStep was just too smart to cope with 80's hardware, for a start. But I see there's other issues, like the REAL NEED for a gentler, stabler OS.
Another thing I'd like people to focus on with more precision is that OSX IS NOT Free/OSS software. Darwin is (and in questionable terms AFAIK).
Still I'm wetting my pants. And whether Apple succeeds or not, one thing is clear IMO: Steve Jobs IS The Man!
(I adore that guy)
A friend of mine borrowed me some Metallica tapes (hey, he owned the LP records! :-) looong time ago, I believe it was the first two albums (Kill'em all and Master of Puppets? Is that right?). I don't like your genre at all, but being an avid listener and (sort of) musician myself, I grew respect of your work from listening those tapes. This is just to point out that I think my opinions are not biased (and may be easily edited off the question :-).
:-). And not to mention the feeling of having done the right thing, and that this added to the probability there will be a shining new release soon.
I feel very debated between the (right IMO) point of view of those feeling they are paying too much for a CD, and the fact that artists have to be payed for their hard work. As some advise, electronic distribution may be an answer, but it happens to be not just around the corner for various reasons. And stealing (yeah, stealing; be honest, that is more or less what mp3 is about - I'm not talking about napster) may or may not be a solution.
The question is: don't you think that adding value to your product may be an answer to the problem?
I can say I am a fan of Steely Dan. I knew Donald Fagen from having his solo records, but I never knew SD. Then I downloaded mp3s, found that I liked SD as much, and actually bought the CDs. I recently downloaded high quality mp3s of the new record. I could have mastered those to a very close resemblance of the original recording, but I haven't and I won't. I will buy the record as soon as I have occasion, and you can bet on it.
Want to know why? I couldn't settle for less than the higher CD quality, cover art, and (why not?) joyful feeling that Fagen's usual frown will eventually be displaced from a nice smile when he sees sales reports, even if for a little while, I suppose
I admit I'm biased on the quality/art topics, but what if the CD included videos? Yeah, you can pass that on the net as well, but what the heck. And what about promotions? Ticket discounts? What else?
Your fans are demanding more than the mere music. That's why they buy gadgets (upon which you also profit). Want to sell more CDs? Try and make them more desirable than mp3 files. You'll get only the good part of the mp3 (or tape, or whatever else) trading, and you can't deny there is a good part. It worked for Steely Dan, it will work for you.
Don't get me wrong. You HAVE to get your bucks for what you do, and that is a fact, and you ought not to be forced to do anything but release your music, that's what you're there for. You have the right to defend your property. But this may be a viable "workaround", until some decent solution comes along, stopping your listeners and fans from being ripped as much as (and probably more than) you are.
I also think that a lot of major issues are not being covered when talking about FS/OSS. The one you address (the developers skill or sense of design) is one of the most important IMO.
I believe that, while we have some major achievements in the "code" ground, we still lack a whole lot in the "logistic" area. Given the very nature of free software, I believe this happens to be far more important than the code itself. And here may be where we lack compared to "suits", one thing that I think we do not want and can not afford.
I see that something has been done in that respect in the last times, but I believe it's too little compared to the importance of the issue. Which is no easy one, though. I'm currently working on a piece about this that I hope I'll be able to submit to the web soon.