Um, by the same logic, Red Hat shouldn't be distributing 90% of the software that they ship, or should be paying thousands of open source programmers for their work.
And remember: this is open source. If RH stops updaing their software, the community could take over.
Open source is a tit-for-tat business. Red Hat's product is built on the backs of others, and Centos is built on Red Hat's. That's the way it should be. Red Hat is selling a brand, support, and piece of mind for the corporate market. Centos does not compete.
That's modded funny, but the unfortunate truth is that it's a farily accurate portrayal of the Dilbert-esq corporate environment that so many of us deal with.
Purchasing ANYTHING that requires ongoing license fees is a TOTAL PITA in any company.
Um, those are "respin" updates. Very much like the concept of Windows install media having service packs and hotfixes pre-applied. RHEL 3.0 original + current errata = RHEL 3.0 Update 4.
Oh come now. Monster cables are quite good. They just are not significantly better than lamp cord (as in 99.9999% of the people on this planet couldn't hear the difference in a double-blind test. And yes, I made that number up.:-)
IMHO, you are spot on. I worked for a 3D company during that time frame. We even had a deal with Yahoo for VRML. It worked, but it sucked.
There were a number of technical issues, one of which was trying to get stable drivers for (expensive) 3D cards. Drivers sucked big time. Massive bugs in the rendering routines. Coding around these bugs was a total pain. The other problem was performance. Besides needing a REALLY fast machine (RAM was wicked expensive at the time, around $500 for 16M) you needed a fast internet connection. Broadband (DSL / cable modem) didn't really take off until a few years later so most people were on dial-up. 3D input controllers were (are) not common, and still fairly expensive, and a pain to use.
Then there are usability issues. VRML is just not an efficient way to "surf" the net. When most of your info is text, what's the point? Google is fast and easy. It's also expensive to create these VRML worlds. It took 2 very good 3D artists weeks to create all the objects and spaces for yahoo's very limited pilot.
Furthermore, there wasn't an infrastructure to interact with other users. For that you need a central server. So VRML was BORING. VRML worlds were static - things didn't move around. Think 3D viewer for 3D home cad programs...
That's where MMORPG's really shine. The 3D world is well defined, the tools to create it are MUCH MUCH better, most of the info is locally stored (on the game CD's), they use efficient proprietary protocols, and a central server so people can interact. Using a mouse for 3D still sucks though:-)
Why modify the system calls?... Write this functionality in userland... write wrappers to the C calls to do this (etc.)
So, are you saying that the parent should modify every single binary on the system???? Including binaries that he may not have source to? Sounds pretty much unworkable. While I wouldn't propose that the parent poster actually implement such a system, the only reasonable place to do this IS at the system call level where it can be applied to everything.
Personally, I think you are better off with a filesystem snapshot facility to handle the problem the parent is trying to solve.
Understand. But I think what the parent was implying is since you have ALREADY paid for the music (due to the MP3 player tax) then you should be able to (morally) download anything at will, and not feel like you are pirating anything.
VoIP stands for Voice over IP. However you want to get it there, it's still Voice over IP. Skype even refers to it's own service as VoIP. You can pretend all you want that it's not, but that doesn't change the fact that it is. What makes the skype service / protocol different than FWD (other than skype using a proprietary protocl?) They do the same thing. Exactly. The protocol Skype uses is easier with firewalls, but that is a minor issue that you can solve (on FWD) by using IAX, or a SIP proxy. Does that make the service totally different?
Um, do you know what VoIP stands for? VoIP is an application and protocol independant term.
Any new regulations would only apply to VoIP providers that route your calls to the PSTN. BTW, skype has a service for this too.
What you are seeing is corporations using VoIP to link branch offices, and other business partners directly without going through the PSTN at all. Research ENUM for more info.
is causing them a loss of noteriety and thus damages their business.
But don't they have to show some reasonable proof that this is the case before a judge would grant an injunction? (I don't know, I'm just asking. IANAL.)
The search engine results offer an excellent way of doing so
And from posts on here indicate, Tiger Direct's web site comes up BEFORE Apple's on all major search engines but Yahoo. Do you think that would damage their ability to get an injunction?
Um, put it back in context without deleting the other qualifications. That would be "most secure, stable and satisfying consumer operating system." Name another "consumer operating system" that is more "secure, stable, and satisfying."
Of course "satisfying" leaves a lot open to personal interpretation, but I would be that most consumers would include things like ease of use, consistency of the interface, documentation, availablity of common third-party apps, and so on. While I own a couple macs, I still use Linux as my primary desktop (it's faster than my aging G4) and Windows for some work stuff.
Let me put it this way: what computer / OS would you be most comfortable setting up for a non-technical parent / grandparent? While I personally like Linux, I certainly wouldn't set it up for my father-in-law... And his existing Windows box is a constant battle with viruses, spyware, etc. (he gets really frustrated with it.) I'm seriously considering setting him up with a Mac Mini.
The reserved mailbox name may be used in a RCPT command "postmaster" without domain qualification see section 4.1.1.3) and MUST be accepted if so used.
So. What this means is that someone can do "RCPT TO:<postmaster>" instead of "RCPT TO:<postmaster@example.com>" and you must accept the SYNTAX of "postmaster" without the domain qualification. It does not mean that you have to accept the entire transaction, or even the command.
Furthermore, even if you DO misread that statement into saying that I MUST accept RCPT commands addressed to postmaster, I can STILL reject the HELO, MAIL From: and DATA parts "for policy reasons". Reading section 3.3 in the paragraph after MAIL FROM:, you will also read that I can optionally delay rejecting the senders address until after I look at where he wants to send it TO. This means that I can accept Mail From, then reject it at RCPT time if I so choose based on local policy.
Let's go one further. Let's say I use a local blacklist to populate a firewall rule. You can't even TALK to my mail server. Does THIS still violate the (misinterpreted version of the) RFC?
You can't read each sentance of the RFC and interpret them individually. You need to read the whole thing to understand what they say. Context is VERY important. Basically, local policy decisions can override just about everything. Nobody is "forced" by the RFC's to do anything.
Read up on the rfc-ignorant.org postmaster policy and maybe you will realize that even the RFC anal people understand that it's OK to reject mail to postmaster for policy reasons.
... And it's a matter of load. Why send things to spamassassin of the site is already blacklisted? THe chances of being legit are near zero. Better to just reject the message with a pointer to info on how to get whitelisted.
For those of you on dynamic addresses trying to run your own mail server (sending direct and not through a legit relay), your days are numbered. More and more ISP's are just blocking outbound port 25 outright, and more and more mail administrators are blocking connections from dynamic IP space. Thanks to the fact that Windows is so insecure, dynamic IP space has become a giant cesspool of compromised machines. So little legit mail comes out of dynablocks that there is no point in NOT rejecting it outright. It's too bad too, because so many ISP's run horrible mail relays (which is probably why you are running your own in the first place...) This generally means that you need to shell out even more to get a third-party hosting service that you can relay through.
Yeah, I dropped one RBL I was using because it listed AOL. That just doesn't work for our users. It's a matter of trust, and I couldn't trust that that particular RBL to NOT pull some other stupid shit in the future.
I find that my own local blacklist is pretty damn effective - rejecting most mail before any external RBL is checked. That keeps the performance up. It's not realistic to send all mail through spamassassin for example - our servers wouldn't handle the load. We DO have to be pretty binary (and aggressive) about it, and getting whitelisted on my servers is really quick and easy.
Show me the RFC that says that I can't filter postmaster... Oh that's right, it doesn't exist. I just need to HAVE a postmaster addresss.
Due to spammer activity, I filter postmaster too, but my rejection message give a pointer to a web page that has an ALTERNATIVE address that is not filtered. Spammers hit (attempt to hit anyway) postmaster, admin, webmaster, info, marketing, sales, and even abuse on a regular basis. They haven't hit my special address yet in the 4 years that it's been active. If they hit it, I'll just change it.
Back to the topic, I found that RBL's just are not enough, and neither is spamassassin. My local blacklist has grown quite large over the last 4 years, and also includes regex's of hundreds of ISP's naming schemes for dynamic DSL / cablemodem space. It's been quite effective with about 80% hit rate BEFORE it checks for RBL's, and spamassassin. Add in a test for IP addresses embedded in the PTR results for SA with a mid score, and most spam is gone. I would need a couple extra servers to handle the spamassassin load if it weren't for my local blacklist. This also takes care of most of the malware (email virus) before it loads down my virus scanner. Spam is so bad that only 15% of mail at our site is legit.
Of course since I have become so agressive in rejecting anything that isn't kosher, I've had to maintain a whitelist too since some sites are just clueless and don't seem to have the capacity to fix their DNS / mail servers. The whitelist is tiny in comparison, and the effort to maintain it minimal.
OK, that doesn't make my comment about "pirate" not being the best term any less valid, as recording to time-shift IS legit. Why should you be forced down the analog hole in order to time-shift?
Anyway, we in the US may lose our "fair-use rights" with the FCC mandated broadcast flag, and the DMCA / FCC rules prohibit us from doing anything to circumvent it.
I guess I'm going to end up being a "criminal pirate" then, because I will NOT allow the government to dictate how I watch TV I already pay for, how I watch / listen to movies and music I own, etc. If there is no feasable technical way to time-shift any more, I just won't be tuning in. They will lose my viewership and hence market share.
The converter boxes will have to output an analogue signal which can be pirated in the usual ways.
"pirate" is the wrong term here. "recorded under our fair use rights" may be better.
There is not an affordable way to digitally capture an analog HDTV signal at the moment. The equipment to do it costs thousands of dollars. This is why everyone wants to do it BEFORE it get's converted to analog...
My guess is that Nikon want's a license fee - say $1 for every copy of photoshop out there in addition to the SDK fee (which is cheap.) The only reason Nikon can possible have to encrypt data, making it MORE difficult for their own customers to access their own data, is to get more money. Digital cameras are a cut-throat business - LOTS of competition. Let's all hope that Nikon loses market share over this to discourage any other manufacturer from doing the same thing. It's a disgusting practice that needs to stop.
Yep. I've done this in X. But the issue is moot as the OP screwed up and assumed that PCI Express is the same as PCI-X, which it's not.
While I like the advances in bus technology, there have gotten to be too many incompatible buses now. It's horrible. Looking at the ASUS web site, we now have motherboards with 5 slots, but all of them different! That's just insane. Why can't they come up with ONE bus that works at high speeds (64 bit), and works for both video and other devices? Is it totally impossible to have it backwards compatible with normal 64-bit PCI? While there may be challenges, would it really be impossible?
Slashdot Mirror
The ISP could turn them off instantly.
Um, right. You make the assumption that the ISP's give a shit. They don't.
Um, by the same logic, Red Hat shouldn't be distributing 90% of the software that they ship, or should be paying thousands of open source programmers for their work.
And remember: this is open source. If RH stops updaing their software, the community could take over.
Open source is a tit-for-tat business. Red Hat's product is built on the backs of others, and Centos is built on Red Hat's. That's the way it should be. Red Hat is selling a brand, support, and piece of mind for the corporate market. Centos does not compete.
That's modded funny, but the unfortunate truth is that it's a farily accurate portrayal of the Dilbert-esq corporate environment that so many of us deal with.
Purchasing ANYTHING that requires ongoing license fees is a TOTAL PITA in any company.
Well, there is a LITTLE more to it than that, but google finds some nice instructions for "upgrading rhel to centos with yum".
Um, those are "respin" updates. Very much like the concept of Windows install media having service packs and hotfixes pre-applied. RHEL 3.0 original + current errata = RHEL 3.0 Update 4.
Oh come now. Monster cables are quite good. They just are not significantly better than lamp cord (as in 99.9999% of the people on this planet couldn't hear the difference in a double-blind test. And yes, I made that number up. :-)
IMHO, you are spot on. I worked for a 3D company during that time frame. We even had a deal with Yahoo for VRML. It worked, but it sucked.
:-)
There were a number of technical issues, one of which was trying to get stable drivers for (expensive) 3D cards. Drivers sucked big time. Massive bugs in the rendering routines. Coding around these bugs was a total pain. The other problem was performance. Besides needing a REALLY fast machine (RAM was wicked expensive at the time, around $500 for 16M) you needed a fast internet connection. Broadband (DSL / cable modem) didn't really take off until a few years later so most people were on dial-up. 3D input controllers were (are) not common, and still fairly expensive, and a pain to use.
Then there are usability issues. VRML is just not an efficient way to "surf" the net. When most of your info is text, what's the point? Google is fast and easy. It's also expensive to create these VRML worlds. It took 2 very good 3D artists weeks to create all the objects and spaces for yahoo's very limited pilot.
Furthermore, there wasn't an infrastructure to interact with other users. For that you need a central server. So VRML was BORING. VRML worlds were static - things didn't move around. Think 3D viewer for 3D home cad programs...
That's where MMORPG's really shine. The 3D world is well defined, the tools to create it are MUCH MUCH better, most of the info is locally stored (on the game CD's), they use efficient proprietary protocols, and a central server so people can interact. Using a mouse for 3D still sucks though
Why modify the system calls? ... Write this functionality in userland ... write wrappers to the C calls to do this (etc.)
So, are you saying that the parent should modify every single binary on the system???? Including binaries that he may not have source to? Sounds pretty much unworkable. While I wouldn't propose that the parent poster actually implement such a system, the only reasonable place to do this IS at the system call level where it can be applied to everything.
Personally, I think you are better off with a filesystem snapshot facility to handle the problem the parent is trying to solve.
Understand. But I think what the parent was implying is since you have ALREADY paid for the music (due to the MP3 player tax) then you should be able to (morally) download anything at will, and not feel like you are pirating anything.
VoIP stands for Voice over IP. However you want to get it there, it's still Voice over IP. Skype even refers to it's own service as VoIP. You can pretend all you want that it's not, but that doesn't change the fact that it is. What makes the skype service / protocol different than FWD (other than skype using a proprietary protocl?) They do the same thing. Exactly. The protocol Skype uses is easier with firewalls, but that is a minor issue that you can solve (on FWD) by using IAX, or a SIP proxy. Does that make the service totally different?
Um, do you know what VoIP stands for? VoIP is an application and protocol independant term.
Any new regulations would only apply to VoIP providers that route your calls to the PSTN. BTW, skype has a service for this too.
What you are seeing is corporations using VoIP to link branch offices, and other business partners directly without going through the PSTN at all. Research ENUM for more info.
is causing them a loss of noteriety and thus damages their business.
But don't they have to show some reasonable proof that this is the case before a judge would grant an injunction? (I don't know, I'm just asking. IANAL.)
The search engine results offer an excellent way of doing so
And from posts on here indicate, Tiger Direct's web site comes up BEFORE Apple's on all major search engines but Yahoo. Do you think that would damage their ability to get an injunction?
I know I'm weary of waiting for OSX Tiger... :-)
Um, put it back in context without deleting the other qualifications. That would be "most secure, stable and satisfying consumer operating system." Name another "consumer operating system" that is more "secure, stable, and satisfying."
Of course "satisfying" leaves a lot open to personal interpretation, but I would be that most consumers would include things like ease of use, consistency of the interface, documentation, availablity of common third-party apps, and so on. While I own a couple macs, I still use Linux as my primary desktop (it's faster than my aging G4) and Windows for some work stuff.
Let me put it this way: what computer / OS would you be most comfortable setting up for a non-technical parent / grandparent? While I personally like Linux, I certainly wouldn't set it up for my father-in-law... And his existing Windows box is a constant battle with viruses, spyware, etc. (he gets really frustrated with it.) I'm seriously considering setting him up with a Mac Mini.
Um, no.
The reserved mailbox name may be used in a RCPT command "postmaster" without domain qualification see section 4.1.1.3) and MUST be accepted if so used.
So. What this means is that someone can do "RCPT TO:<postmaster>" instead of "RCPT TO:<postmaster@example.com>" and you must accept the SYNTAX of "postmaster" without the domain qualification. It does not mean that you have to accept the entire transaction, or even the command.
Furthermore, even if you DO misread that statement into saying that I MUST accept RCPT commands addressed to postmaster, I can STILL reject the HELO, MAIL From: and DATA parts "for policy reasons". Reading section 3.3 in the paragraph after MAIL FROM:, you will also read that I can optionally delay rejecting the senders address until after I look at where he wants to send it TO. This means that I can accept Mail From, then reject it at RCPT time if I so choose based on local policy.
Let's go one further. Let's say I use a local blacklist to populate a firewall rule. You can't even TALK to my mail server. Does THIS still violate the (misinterpreted version of the) RFC?
You can't read each sentance of the RFC and interpret them individually. You need to read the whole thing to understand what they say. Context is VERY important. Basically, local policy decisions can override just about everything. Nobody is "forced" by the RFC's to do anything.
Read up on the rfc-ignorant.org postmaster policy and maybe you will realize that even the RFC anal people understand that it's OK to reject mail to postmaster for policy reasons.
... And it's a matter of load. Why send things to spamassassin of the site is already blacklisted? THe chances of being legit are near zero. Better to just reject the message with a pointer to info on how to get whitelisted.
For those of you on dynamic addresses trying to run your own mail server (sending direct and not through a legit relay), your days are numbered. More and more ISP's are just blocking outbound port 25 outright, and more and more mail administrators are blocking connections from dynamic IP space. Thanks to the fact that Windows is so insecure, dynamic IP space has become a giant cesspool of compromised machines. So little legit mail comes out of dynablocks that there is no point in NOT rejecting it outright. It's too bad too, because so many ISP's run horrible mail relays (which is probably why you are running your own in the first place...) This generally means that you need to shell out even more to get a third-party hosting service that you can relay through.
Yeah, I dropped one RBL I was using because it listed AOL. That just doesn't work for our users. It's a matter of trust, and I couldn't trust that that particular RBL to NOT pull some other stupid shit in the future.
I find that my own local blacklist is pretty damn effective - rejecting most mail before any external RBL is checked. That keeps the performance up. It's not realistic to send all mail through spamassassin for example - our servers wouldn't handle the load. We DO have to be pretty binary (and aggressive) about it, and getting whitelisted on my servers is really quick and easy.
Doing better than that is likely to be full-time job 9or more0.
Nah. Takes about 5-10 minutes a day. Maybe 30 minutes if you are a large site.
Show me the RFC that says that I can't filter postmaster... Oh that's right, it doesn't exist. I just need to HAVE a postmaster addresss.
Due to spammer activity, I filter postmaster too, but my rejection message give a pointer to a web page that has an ALTERNATIVE address that is not filtered. Spammers hit (attempt to hit anyway) postmaster, admin, webmaster, info, marketing, sales, and even abuse on a regular basis. They haven't hit my special address yet in the 4 years that it's been active. If they hit it, I'll just change it.
Back to the topic, I found that RBL's just are not enough, and neither is spamassassin. My local blacklist has grown quite large over the last 4 years, and also includes regex's of hundreds of ISP's naming schemes for dynamic DSL / cablemodem space. It's been quite effective with about 80% hit rate BEFORE it checks for RBL's, and spamassassin. Add in a test for IP addresses embedded in the PTR results for SA with a mid score, and most spam is gone. I would need a couple extra servers to handle the spamassassin load if it weren't for my local blacklist. This also takes care of most of the malware (email virus) before it loads down my virus scanner. Spam is so bad that only 15% of mail at our site is legit.
Of course since I have become so agressive in rejecting anything that isn't kosher, I've had to maintain a whitelist too since some sites are just clueless and don't seem to have the capacity to fix their DNS / mail servers. The whitelist is tiny in comparison, and the effort to maintain it minimal.
OK, that doesn't make my comment about "pirate" not being the best term any less valid, as recording to time-shift IS legit. Why should you be forced down the analog hole in order to time-shift?
Anyway, we in the US may lose our "fair-use rights" with the FCC mandated broadcast flag, and the DMCA / FCC rules prohibit us from doing anything to circumvent it.
I guess I'm going to end up being a "criminal pirate" then, because I will NOT allow the government to dictate how I watch TV I already pay for, how I watch / listen to movies and music I own, etc. If there is no feasable technical way to time-shift any more, I just won't be tuning in. They will lose my viewership and hence market share.
That's why you re-rip them. Fuck the MPAA.
The converter boxes will have to output an analogue signal which can be pirated in the usual ways.
"pirate" is the wrong term here. "recorded under our fair use rights" may be better.
There is not an affordable way to digitally capture an analog HDTV signal at the moment. The equipment to do it costs thousands of dollars. This is why everyone wants to do it BEFORE it get's converted to analog...
My guess is that Nikon want's a license fee - say $1 for every copy of photoshop out there in addition to the SDK fee (which is cheap.) The only reason Nikon can possible have to encrypt data, making it MORE difficult for their own customers to access their own data, is to get more money. Digital cameras are a cut-throat business - LOTS of competition. Let's all hope that Nikon loses market share over this to discourage any other manufacturer from doing the same thing. It's a disgusting practice that needs to stop.
Yep. I've done this in X. But the issue is moot as the OP screwed up and assumed that PCI Express is the same as PCI-X, which it's not.
While I like the advances in bus technology, there have gotten to be too many incompatible buses now. It's horrible. Looking at the ASUS web site, we now have motherboards with 5 slots, but all of them different! That's just insane. Why can't they come up with ONE bus that works at high speeds (64 bit), and works for both video and other devices? Is it totally impossible to have it backwards compatible with normal 64-bit PCI? While there may be challenges, would it really be impossible?
From "Real Programmers Don't Eat Quiche":
Real Programmers don't comment their code. If it was hard to write, it should be hard to understand and even harder to modify.