"The serials should always be the same on all nodes."... "But the most serious limitation are the buggy DNS servers around the world. This setup assumes that a DNS server or resolver obeys the expire time of a zone record (the 60 seconds used above). Unfortunatly, there are a lot of servers out there which don't do that."
Aren't other DNS servers allowed to look at your SOA serial number, notice it hasn't changed, and not bother doing any other work? Isn't that the point of having serial numbers?
It sounds like you are blaming all those other DNS servers for following the RFC.
But I'm sure you've given this more thought that I have, so tell me what I'm missing.
I wonder if anyone will admit to having a SCO certification....
Their effort doesn't scale well
on
The Virus Squad
·
· Score: 4, Interesting
All that effort and the anti virus companies still haven't figured out a way to share their work with a common signature file. No wonder there is so much drugery.
They can call it a gift all they want, but if you press the issue they will probably almost all acknowledge that there are strings attached. I find it more appropriate to call a gift with conditions an exchange. Granted, it is an exchange in which you can come out of it without paying anything back, but only if you limit your behaviour.
I didn't say majority is always right, I said enough people disagree with you that you should consider that possibility that you are wrong. Way different.
You ignored the bit where I pointed out that the GPL is an exchange and BSD is a gift.
"I think there would be a market for customized software if it wasn't for the GPL."
Great, you get to think that. Welcome to the real world. The GPL exists. Deal with it.
There was a market for the GPL, and the GPL provides sufficient value to sufficient people that it has gained widespread acceptance and use. There is a reason for that. If you think the GPL is a bad thing, figure out what the people who use the GPL get out of it, and if you can find a way to get them what they want/need in a way that doesn't do the harm you percieve, sell them on that idea instead.
Consider the marketplace for license schemes. If you want to outcompete the GPL, you've got some thinking to do. Complaining about the GPL and equating it with an alien invasion isn't likely to help.
Feel free to offer an alternative. Complaining doesn't impress anyone.
If you offer an alternative and it is rejected, it will be because it isn't as good. Again, complaining won't help.
There is nothing natural about government regulation. Government regulation has no place in the natural balance of the marketplace.
GPL is the cutthroat competition you ask talk about.
Either you accept the terms and get all the rewards (source code) or you opt out, avoid the GPL, do everything the hard way, get frustrated with the tedium of working overtime to rebuild countless inferior wheels, and end up on Slashdot ranting like a loon.
Better luck next life.
They charge for their proxy and satellite update products, so every one of your servers has to connect to their servers and use their bandwidth unless you pay them for the priviledge of saving both of you time and bandwidth. You end up downloading the same bits over and over again.
There isn't any good technical reason to connect to redhat.com when you run up2date. That's actually a bad idea, and it is an artifact of their choice to charge for subscriptions and centralize everything.
As I see this they've failed to take advantage of the network effect that would be avaiable if up2date could work off proxies and mirrors.
As another example, I didn't want to run up2date on my 2 RHEL boxes but I needed to be able to download the binary update RPMs. I had to write my own script to use curl to pull the damn things off their https website with a username/password because they wouldn't make them available for FTP, rsync (ssh), bittorrent or any other common tool.
I'd certainly agree that God hasn't been alive any time in the last 90 years. This neatly avoid any questions about any existance or life prior to that time.
No, I've read stuff Stallman has written on this subject, and I've heard him speak in person. Stallman has a very clear understanding of the difference between the kernel and the collection of software that makes up an OS, and his statements maintain this clarity.
It is vastly more likely that the author got it wrong than that Stallman wasn't clear on this issue.
I've used Dia a LOT under Linux, and I've never had that problem. Assuming the features have been ported correctly you should be able to go to page setup, and tell it to fit the image to any number of pages. If you select 1 by 1 your entire drawing will be on 1 page. If you go 4 by 4 it will be on 16 pages you can stictch together. Take a look at the page setup options. Notice that the settings effect where the blue lines are. Those will be the page boundaries when you print.
Contrast this with propriatary systems. I'm finding bugs in things like BEA's WebLogic Server all the time, and I can barely get them to acknowledge the bug, to say nothing of fixing it.
"Based on that, I would say there are probably other Linux boxes out there administered by idiots."
Certainly, nobody should argue with that point, but there is no reason to think that your testimonial example is representative either. Additionally, companies that are being bought aren't usually a great place to look for good examples, they've often done a number of stupid things to get where they are.
There are idiots everywhere. It hardly matters if those idiots run unpatched RedHat 7.0 boxes or unpatched Windows 98 boxes, they will all be compromised, if they haven't already been.
ObStupidRant: I'm sick and tired of all these stories that focus exclusively on Britain. Don't the moronic editors here know that the Internet is not limited to the UK?
The problem is all the people and property being dragged to the south portion of England. They need to move a few large cities to the north end to balance things out better.;-}
"Yes this is all assuming that there isn't a meta-matrix in which the matrix is run. It seems pretty obvious to me that that is what the last scene was trying to portray in a subtle way.
And if there is a meta-matrix, what prevents having a meta-meta-matrix and so on? Its really impossible to speculate the age of the matrix based on this information."
Yeah, so there's a meta matrix, and Neo has woken up in that outer world, which explains the coma.
Right, whatever.
So I expect the third movie will go something like this: Neo realizes he is in the meta matrix, and begins to wonder if there is a meta meta matrix. He tries to wake up from that matrix, by hitting something to see if it also feels 'wrong'. Repeat this a few times, and Neo is just diving out windows each time he 'wakes up', until at the end of the third movie we see some pimple-faced kid take a dive out a high-rise apartment complex, and the movie ends with a news crew interviewing neighbors, who blame teen suicide on these new immersive video games.
"Now people are worried that because SCO also distributed this same code under the GPL, that somehow "legitimizes" its release, which makes the GPL seem "viral". In other words, because SCO owned the original code, by releasing that code under the GPL they in effect dual-licensed it, even though the Linux contributor who put the code in there didn't have a legal right to do so.
Do people think this will really hold up? In order to release something under the GPL in the first place, you have to own the code or have the right to license that code to begin with. If I steal Microsoft's code and release it under the GPL, that doesn't make the code actually GPL'd, because I never had the right to so license the code."
The confusion comes about because SCO had their own Linux distribution (SCO Linux) which probably included the code they've been claiming was wrongfully used. Hard to tell yet because they aren't mitigating damages by releasing details, but it seems quite likely that they released the code under the GPL when they distributed their own version of SCO Linux.
If these things are true, SCO will probably claim they didn't mean to release those bits under the GPL, and this happened inadvertantly because the code had been inserted without their consent.
All that assumes they can actually point to any code that was really copied from
their copyrighted code base in the first place.
" Then your company starts distributing RedHat. Then you discover that you've been distributing your own code, inadvertantly, under the GPL, so there's nothing that you can do except fire and sue your employee.
Some random company isn't going to be distributing RedHat. They might be mirroring a copy of it (which RedHat appears to tolerate because it reduces their bandwith requirements), or they might be distributing their own distribution of Linux, which they might have based on a source distribution of RedHat and modified so it doesn't infringe on any of RedHat's copyrights or trademarks.
If you mean, someone might have created their own distribution of Linux, and distributed a binary version along with the requirements to make source code available for three years and all that, they certainly should have taken the time to verify they have all the appropriate rights to distribute everything in the distribution, and oh yeah, they might also care to run a quick compare against any code they already own that they want to keep propriatary, to avoid inadvertantly granting a license they didn't intend to grant.
This seems to be the situation SCO is in.
SCO didn't just wake up one day and throw together the SCO Linux distribution.
SCO would have made lots of choice about what applications and utilities to include to make SCO Linux useful, and which kernel options to compile with.
SCO has claimed they've been able to find their own code in the kernel, but they haven't given any information on why they found this after they started distributing SCO Linux instead of before distributing SCO Linux.
I don't think they've suggested they were duped, and it only occured to they to run a compare after someone pointed and laughed. It sounds more like one hand not knowing what the other was doing.
"50K per message, are you INSANE??? I hate it when servers limit me to 2 or 5 MB let alone something so insane. Let me guess, you never send anything with attachments, get real."
Why are you passing by value instead of passing by reference?
That is to say, why are you sending an attachement instead of a URL or some other pointer to the file?
Chewie did say "... if possible". That hardly sounds insane to me.
That idea would only penalize those who use email legitimately, because people who spam would avoid paying the tax, or push the tax onto some unwitting victim.
Instead, mandate the already existing "Precedence: bulk" header every time a nearly identical message is sent to more than 50 people.
This would allow recipients to whitelist the mailing lists they subscribe to, and discard all other incoming email with the "Precedence: bulk" header.
The real problem with spam is that spam is sent to vast numbers of people, but it doesn't identify itself as being a bulk mailing.
By requiring that all bulk mailings be identified as such, and providing for penalties when the identification is missing, we can prosecute people who intentionally harm others, which is the basis of all legitimate law in the first place.
Successful prosecution would be possible when 20 different people bounce the same infringing message (unwanted bulk email without "Precedence: bulk" header) to their favorate spam prosecution group (different groups would have different rules about how they prosecute spam, who gets the proceeds of sucessful convictions, and which other spam prosecution groups they will work with). This provides a safety mechanism for people who send unobjectionable mailings and forget the "Precedence: bulk" tag, because it requires that there be victims. It also provides for credible testimony, because many people have to forward the same infringing message.
Changing a few characters here and there wouldn't matter, because a judge could evaluate the "nearly identical" clause.
If spammers would use the "Precedence: bulk" tag, we could easily filter out their unwanted messages.
If spammers had to compose new versions of their pitch for every 49 people they wanted to reach, they wouldn't be sending out nearly as many messages.
"[Sending email] is not a problem: You merely have to relay your outgoing email through your service provider's SMTP server. This is the way that you should your outbound email configured anyway, you're incorrectly configured if you're not passing mail upstream through your ISP."
You are assuming that the domains are the same. If a home user has their own domain they may have legitimate need/desire to seperate their domain's email from their ISP's domain. This does not constitute incorrect configuration.
Slashdot Mirror
"The serials should always be the same on all nodes." ... "But the most serious limitation are the buggy DNS servers around the world. This setup assumes that a DNS server or resolver obeys the expire time of a zone record (the 60 seconds used above). Unfortunatly, there are a lot of servers out there which don't do that."
Aren't other DNS servers allowed to look at your SOA serial number, notice it hasn't changed, and not bother doing any other work? Isn't that the point of having serial numbers?
It sounds like you are blaming all those other DNS servers for following the RFC.
But I'm sure you've given this more thought that I have, so tell me what I'm missing.
I wonder if anyone will admit to having a SCO certification....
All that effort and the anti virus companies still haven't figured out a way to share their work with a common signature file. No wonder there is so much drugery.
Next on the list, that alarm clock on my headboard that I sleep next to for 5-6 hours a day. Not that I'll be sad to see it go.
They can call it a gift all they want, but if you press the issue they will probably almost all acknowledge that there are strings attached. I find it more appropriate to call a gift with conditions an exchange. Granted, it is an exchange in which you can come out of it without paying anything back, but only if you limit your behaviour.
Haha, you funny guy.
I didn't say majority is always right, I said enough people disagree with you that you should consider that possibility that you are wrong. Way different.
You ignored the bit where I pointed out that the GPL is an exchange and BSD is a gift.
"But my point was that BSD vs. proprietrary has the exact same advantages."
I call Bullshit.
First, if that were accurate, you wouldn't find so many people disagreeing with you. Consider the possibility that they know something you don't know.
A big difference from the developers perspective is how much they get back in exchange.
Think of it this way if you like: BSD is a gift, GPL is a trade. Gift != Trade.
Simple enough?
Now, you seem to be pissed that people are offering you a trade you don't want to accept instead of the gift you think you should be intitled to.
Go figure, you aren't happy with the fact that they refuse to convert over and just give you gifts.
The only strange thing here is that your greed gets so much attention.
"I think there would be a market for customized software if it wasn't for the GPL."
Great, you get to think that. Welcome to the real world. The GPL exists. Deal with it.
There was a market for the GPL, and the GPL provides sufficient value to sufficient people that it has gained widespread acceptance and use. There is a reason for that. If you think the GPL is a bad thing, figure out what the people who use the GPL get out of it, and if you can find a way to get them what they want/need in a way that doesn't do the harm you percieve, sell them on that idea instead.
Consider the marketplace for license schemes. If you want to outcompete the GPL, you've got some thinking to do. Complaining about the GPL and equating it with an alien invasion isn't likely to help.
Feel free to offer an alternative. Complaining doesn't impress anyone.
If you offer an alternative and it is rejected, it will be because it isn't as good. Again, complaining won't help.
There is nothing natural about government regulation. Government regulation has no place in the natural balance of the marketplace. GPL is the cutthroat competition you ask talk about. Either you accept the terms and get all the rewards (source code) or you opt out, avoid the GPL, do everything the hard way, get frustrated with the tedium of working overtime to rebuild countless inferior wheels, and end up on Slashdot ranting like a loon. Better luck next life.
They charge for their proxy and satellite update products, so every one of your servers has to connect to their servers and use their bandwidth unless you pay them for the priviledge of saving both of you time and bandwidth. You end up downloading the same bits over and over again. There isn't any good technical reason to connect to redhat.com when you run up2date. That's actually a bad idea, and it is an artifact of their choice to charge for subscriptions and centralize everything. As I see this they've failed to take advantage of the network effect that would be avaiable if up2date could work off proxies and mirrors. As another example, I didn't want to run up2date on my 2 RHEL boxes but I needed to be able to download the binary update RPMs. I had to write my own script to use curl to pull the damn things off their https website with a username/password because they wouldn't make them available for FTP, rsync (ssh), bittorrent or any other common tool.
Can't we just expand the scope a bit and agree that authoritarian governments have been monumental failures?
They should set up signs next to those that read "gun free".
I'd certainly agree that God hasn't been alive any time in the last 90 years. This neatly avoid any questions about any existance or life prior to that time.
No, I've read stuff Stallman has written on this subject, and I've heard him speak in person. Stallman has a very clear understanding of the difference between the kernel and the collection of software that makes up an OS, and his statements maintain this clarity. It is vastly more likely that the author got it wrong than that Stallman wasn't clear on this issue.
I've used Dia a LOT under Linux, and I've never had that problem. Assuming the features have been ported correctly you should be able to go to page setup, and tell it to fit the image to any number of pages. If you select 1 by 1 your entire drawing will be on 1 page. If you go 4 by 4 it will be on 16 pages you can stictch together. Take a look at the page setup options. Notice that the settings effect where the blue lines are. Those will be the page boundaries when you print.
Contrast this with propriatary systems. I'm finding bugs in things like BEA's WebLogic Server all the time, and I can barely get them to acknowledge the bug, to say nothing of fixing it.
Certainly, nobody should argue with that point, but there is no reason to think that your testimonial example is representative either. Additionally, companies that are being bought aren't usually a great place to look for good examples, they've often done a number of stupid things to get where they are.
There are idiots everywhere. It hardly matters if those idiots run unpatched RedHat 7.0 boxes or unpatched Windows 98 boxes, they will all be compromised, if they haven't already been.
ObStupidRant: I'm sick and tired of all these stories that focus exclusively on Britain. Don't the moronic editors here know that the Internet is not limited to the UK?
The problem is all the people and property being dragged to the south portion of England. They need to move a few large cities to the north end to balance things out better. ;-}
And if there is a meta-matrix, what prevents having a meta-meta-matrix and so on? Its really impossible to speculate the age of the matrix based on this information."
Yeah, so there's a meta matrix, and Neo has woken up in that outer world, which explains the coma.
Right, whatever.
So I expect the third movie will go something like this: Neo realizes he is in the meta matrix, and begins to wonder if there is a meta meta matrix. He tries to wake up from that matrix, by hitting something to see if it also feels 'wrong'. Repeat this a few times, and Neo is just diving out windows each time he 'wakes up', until at the end of the third movie we see some pimple-faced kid take a dive out a high-rise apartment complex, and the movie ends with a news crew interviewing neighbors, who blame teen suicide on these new immersive video games.
Do people think this will really hold up? In order to release something under the GPL in the first place, you have to own the code or have the right to license that code to begin with. If I steal Microsoft's code and release it under the GPL, that doesn't make the code actually GPL'd, because I never had the right to so license the code."
The confusion comes about because SCO had their own Linux distribution (SCO Linux) which probably included the code they've been claiming was wrongfully used. Hard to tell yet because they aren't mitigating damages by releasing details, but it seems quite likely that they released the code under the GPL when they distributed their own version of SCO Linux.
If these things are true, SCO will probably claim they didn't mean to release those bits under the GPL, and this happened inadvertantly because the code had been inserted without their consent.
All that assumes they can actually point to any code that was really copied from their copyrighted code base in the first place.
If you mean, someone might have created their own distribution of Linux, and distributed a binary version along with the requirements to make source code available for three years and all that, they certainly should have taken the time to verify they have all the appropriate rights to distribute everything in the distribution, and oh yeah, they might also care to run a quick compare against any code they already own that they want to keep propriatary, to avoid inadvertantly granting a license they didn't intend to grant.
This seems to be the situation SCO is in.
SCO didn't just wake up one day and throw together the SCO Linux distribution.
SCO would have made lots of choice about what applications and utilities to include to make SCO Linux useful, and which kernel options to compile with.
SCO has claimed they've been able to find their own code in the kernel, but they haven't given any information on why they found this after they started distributing SCO Linux instead of before distributing SCO Linux.
I don't think they've suggested they were duped, and it only occured to they to run a compare after someone pointed and laughed. It sounds more like one hand not knowing what the other was doing.
Bummer, sucks to be them.
Why are you passing by value instead of passing by reference?
That is to say, why are you sending an attachement instead of a URL or some other pointer to the file?
Chewie did say "... if possible". That hardly sounds insane to me.
Instead, mandate the already existing "Precedence: bulk" header every time a nearly identical message is sent to more than 50 people.
This would allow recipients to whitelist the mailing lists they subscribe to, and discard all other incoming email with the "Precedence: bulk" header.
The real problem with spam is that spam is sent to vast numbers of people, but it doesn't identify itself as being a bulk mailing.
By requiring that all bulk mailings be identified as such, and providing for penalties when the identification is missing, we can prosecute people who intentionally harm others, which is the basis of all legitimate law in the first place.
Successful prosecution would be possible when 20 different people bounce the same infringing message (unwanted bulk email without "Precedence: bulk" header) to their favorate spam prosecution group (different groups would have different rules about how they prosecute spam, who gets the proceeds of sucessful convictions, and which other spam prosecution groups they will work with). This provides a safety mechanism for people who send unobjectionable mailings and forget the "Precedence: bulk" tag, because it requires that there be victims. It also provides for credible testimony, because many people have to forward the same infringing message.
Changing a few characters here and there wouldn't matter, because a judge could evaluate the "nearly identical" clause.
If spammers would use the "Precedence: bulk" tag, we could easily filter out their unwanted messages.
If spammers had to compose new versions of their pitch for every 49 people they wanted to reach, they wouldn't be sending out nearly as many messages.
You are assuming that the domains are the same. If a home user has their own domain they may have legitimate need/desire to seperate their domain's email from their ISP's domain. This does not constitute incorrect configuration.