Yah, Yah, Yah, but what I really want to know is "What is the largest array of disk drives attached to a single linux system that is operational somwhere in the world today?"
I discussed this topic with some length with a lawyer in Canada. No idea if it applies elsewhere but the thought process probably does
1) Just because some university document say "The students give up all human rights" does not make it so. University policy is more or less irrelevant other than the context of what sort of contracts you may have entered and how it fits into local law. Universities often put stuff in their policies that go way beyond what they actually have domain to control.
2) It is very unlikely that local law will see hundreds of pages of policy you never read when you agreed to go to the university as legally binding on issues as important as copyright.
3) If you in any ways used university equipment or computers to do the work, it may very well be seen as a joint creation with the university having some ownership in it. Once they have some tiny ownership, you need joint agreement to do much with it.
4) If the university was paying for your research in the way of a grant or internship, they probably have partial ownership in much the same way an employer might own software you write.
Universities are doing their best to generate money from licensing technology - you can count on the fact that your university has some industrial liaison office or technology licensing. To make this work they need to make sure they own everything of any use so their goals are 1) own all IP that professors produce 2) own all IP that grad student produce 3) if they think they can sell own all IP that undergrads produce. Undergrads owning the code are a slipper slope to having the grad students own IP which they are very worried about. They very well could be willing to take you to court over a seemingly silly thing because it sets the stage for more complex things.
My personal advice, ask a tenured faculty member if you can GPL the work. If they say yes, ask them to send you that in an email - then go do it.
PS - Obviously legal advice from someone who know nothing about your situation, received over the internet, from someone who is not a lawyer - is truly not worth the 2 bits.
You could just hack your own protocol to do it Why not do it the open standards based way? Use SIP - it is good for voice and there is substantial work to build instant messaging and presence stuff on it. Someone, sipforum I think, showed Quake with SIP.
I have worked on the SIP stuff at www.vovida.org and it is open source.
Productivity is the wrong metric
on
Buried in email?
·
· Score: 1
I work at Cisco and yes I spend lots of time dealing with email but I have no idea how people come up with these productivity gains. The other night I sent an email (via Inmarsat) to my brother who is running a fish boat somewhere in the middle of the Pacific (actually quite near where Mir crashed). Sure email was lots more effective than swimming from San Jose to New Zealand and delivering a message in a bottle but was it a valid comparison for productivity? Most these productivity comparisons are equally absurd. If it were not for email, I would not have sent a message to my brother- it's not always a productivity issue, it about doing things that might have been difficult to do otherwise. Like collaborating on an open source router that runs on Linux...:-)
yah - we laughed a bit about this award at times too. It won some other possibly cheesy awards (Like Linux World Show Favorite) - you can see them at http://www.vovida.com/pressevents.html . The real taste test is in the dog food. Several people here use it to run the telephone they use every day and like it.
Keep in mind RedHat 7.0 is compiled with a non release version of gcc - one that is likely to be incompatible with the next release of gcc. Personally, I think the binary incompatibility hell has just started for RH 7.0 and that RH 7.0 binaries could very well be incompatible with RH 7.1 I think it is fine that RH has released such a bleeding edge product but I think Oracle would be probably doing the right thing is saying "our stuff runs on RH 6.2 but we are not going to support in on 7.0 until it is less bleeding edge. Like actually had the 2.4 kernel and is compiled with a compiler that is at least a beta release of the compiler.
The important point for the average Linux user is, open source does involve getting new stuff out early (thanks RH for doing this) but the flip side of this coin is that just because it is newer does not mean it is better. If it does not do what you want to do, it is is not better and perhaps you should use something else (like RH 6.2)
9 out of 10 times you are prob. right that the code is junk and in the long term needs to be redone but.... 1 in 10, the code is quite good, it is just using some technique or design pattern that you don't understand. Make sure you are not in the second case before you start hacking.
I work on a large open source project (www.vovida.org) It originally had a LGPL licenses but now it has something very close to a BSD license. Since this change we have had far less conversations along the lines of this one.
Slashdot Mirror
Yah, Yah, Yah, but what I really want to know is "What is the largest array of disk drives attached to a single linux system that is operational somwhere in the world today?"
Uh - I think you are confused on what a dB is but your point is still sort of ok.
I doubt Cisco is the largest number but there is a lot of Linux desktops at Cisco - no idea the actual numbers.
And if you want to learn more about what passport is before you get one - you have to go to a web site that requires a pssport. Wonderful.
I discussed this topic with some length with a lawyer in Canada. No idea if it applies elsewhere but the thought process probably does 1) Just because some university document say "The students give up all human rights" does not make it so. University policy is more or less irrelevant other than the context of what sort of contracts you may have entered and how it fits into local law. Universities often put stuff in their policies that go way beyond what they actually have domain to control. 2) It is very unlikely that local law will see hundreds of pages of policy you never read when you agreed to go to the university as legally binding on issues as important as copyright. 3) If you in any ways used university equipment or computers to do the work, it may very well be seen as a joint creation with the university having some ownership in it. Once they have some tiny ownership, you need joint agreement to do much with it. 4) If the university was paying for your research in the way of a grant or internship, they probably have partial ownership in much the same way an employer might own software you write. Universities are doing their best to generate money from licensing technology - you can count on the fact that your university has some industrial liaison office or technology licensing. To make this work they need to make sure they own everything of any use so their goals are 1) own all IP that professors produce 2) own all IP that grad student produce 3) if they think they can sell own all IP that undergrads produce. Undergrads owning the code are a slipper slope to having the grad students own IP which they are very worried about. They very well could be willing to take you to court over a seemingly silly thing because it sets the stage for more complex things. My personal advice, ask a tenured faculty member if you can GPL the work. If they say yes, ask them to send you that in an email - then go do it. PS - Obviously legal advice from someone who know nothing about your situation, received over the internet, from someone who is not a lawyer - is truly not worth the 2 bits.
Hmm, many of the books I own, I have written notes on the pages and ahggh even highlighted things. Should that be illegal too?
You could just hack your own protocol to do it Why not do it the open standards based way? Use SIP - it is good for voice and there is substantial work to build instant messaging and presence stuff on it. Someone, sipforum I think, showed Quake with SIP. I have worked on the SIP stuff at www.vovida.org and it is open source.
I work at Cisco and yes I spend lots of time dealing with email but I have no idea how people come up with these productivity gains. The other night I sent an email (via Inmarsat) to my brother who is running a fish boat somewhere in the middle of the Pacific (actually quite near where Mir crashed). Sure email was lots more effective than swimming from San Jose to New Zealand and delivering a message in a bottle but was it a valid comparison for productivity? Most these productivity comparisons are equally absurd. If it were not for email, I would not have sent a message to my brother- it's not always a productivity issue, it about doing things that might have been difficult to do otherwise. Like collaborating on an open source router that runs on Linux ... :-)
yah - we laughed a bit about this award at times too. It won some other possibly cheesy awards (Like Linux World Show Favorite) - you can see them at http://www.vovida.com/pressevents.html . The real taste test is in the dog food. Several people here use it to run the telephone they use every day and like it.
Freeworld dial up at http://pulver.com/fwd/ is using a bunch of this software to provide IP phone services.
Keep in mind RedHat 7.0 is compiled with a non release version of gcc - one that is likely to be incompatible with the next release of gcc. Personally, I think the binary incompatibility hell has just started for RH 7.0 and that RH 7.0 binaries could very well be incompatible with RH 7.1 I think it is fine that RH has released such a bleeding edge product but I think Oracle would be probably doing the right thing is saying "our stuff runs on RH 6.2 but we are not going to support in on 7.0 until it is less bleeding edge. Like actually had the 2.4 kernel and is compiled with a compiler that is at least a beta release of the compiler. The important point for the average Linux user is, open source does involve getting new stuff out early (thanks RH for doing this) but the flip side of this coin is that just because it is newer does not mean it is better. If it does not do what you want to do, it is is not better and perhaps you should use something else (like RH 6.2)
9 out of 10 times you are prob. right that the code is junk and in the long term needs to be redone but .... 1 in 10, the code is quite good, it is just using some technique or design pattern that you don't understand. Make sure you are not in the second case before you start hacking.
I work on a large open source project (www.vovida.org) It originally had a LGPL licenses but now it has something very close to a BSD license. Since this change we have had far less conversations along the lines of this one.
I often send email from the internet through a gateway to my bother on his boat. He has an imarsat system. Works really well.
Also check out www.vovida.com (I work there) who have put out several open source stacks for VoIP and a software phone that uses SIP and runs on Linux with a Quicknet phonejack card. Vovida has also sponsored a project to create an open source client like you want. It is at http://sourceforge.net/project/?group_id=5560 and the RFP is at http://www.sourcexchange.com/ProjectDetail?project ID=20
Just the improvement in G++ templates is enough to get my vote (yep you can use STL now).