I have done this too. I think IRC is great for this. I worked with a guy who was about 700 miles away. We both just grabed a private chanel on EfNet.
What I don't get is how to introduce this to a small team of 5 developers that all sit in cubes in the same room.
Monopolies are generally bad things. We should think long and hard before we create new ones, particularly ones with explicit legislative
sanction.
Yes, monopolies are bad. But what can we do? It seems like somebody has to have a monopoly on broadband. The only alternative would be to give right of way to every "Tom, Dick, or Harry" who wanted to bury cable to your house.
Currently the only people in the US who own right of way are the municipalities, the electric company, the telephone company, the cable company, and the railroads. Other than the municipalities, all of them are government sanctioned monopolies.
Given those choices, I would vote for the municipalities to control broadband. I think the analogy to the interstate system is a good one. The Feds subsidize, but the state and cities pay for it and oversee it. Few places have perfect roads, but almost all places are drivable.
BTW, there is motivation for cities to keep their internet infrastructure up to date. Think about how much competition there is to get large businesses to move to a particular city.
``We have learned we have to explain what we mean in plain
language because legal terms are sometimes not clear.'' -- Peter Csonka, deputy head of the Council of Europe's economic
crime division which oversees the drafting
By including "GCC 2.96" in Red Hat Linux 7.0 they are creating a situation
of vendor lock in for the users of any proprietary software that is released "for Red Hat Linux" and compiled on
the latest version.
As an employee at a company that will soon be shipping a binary-only version of our software for linux, I'm tempted to be offended by this. I see two implications of your statement. One: you think I'm too stupid to know that I can release more than one binary. I'm so wowed by RedHat's "for RedHat Linux" program to see anything else. Two: you misunderstand vendor lockin. I'm going to assume the second.
Vendor lockin means that I can't switch vendors. I am logistically prevented from doing it. The gcc 2.96 problem is not that case. My source code works with gcc 2.95.2 and 2.96. Since I own the code, I am free to compile it twice. There's no vendor lockin here.
Another point to vendor lockin is that gcc 3.0 eventually will ship. When that happens, any of my 2.96 specific code will work with any distribution that ships 3.0.
I admit I'm not on the inside of the GCC steering commitee stuff. I don't know if anyone had any gag orders or not. However, even I new RedHat 7.0 was going to be shipping a "bogus" gcc 2.96. Even before the 7.0 beta release, I had upgraded my gcc from rawhide. This was back in early July. How much lead time did you want? Is three months not enough? BTW, I upgraded because directed me to in responce to bug #13547 (a problem with g++ and iostream).
I don't see how anyone can criticize RedHat for not disclosing their plans. As soon as they start on the next release, they put the whole thing in rawhide. I don't see MS putting NT64 pre-releases up on the net. I don't see them putting their working copies of anything up on the net.
I agree with Bob Y. You can't make legitimate comparisons between MS and RH. The only exception might be that RH is successful in the Linux world and MS is successful in their nitch.
But remember, Becker doesn't "own" his code anymore - he GPL'ed it.
Licensing code under the GPL doesn't forfeit ownership. It merely describes the terms under which it can be used. If you are using the code under those terms, then you don't need permission of the code's owner. However, if you want to use the code in some fashion not covered under the terms of the GPL, then you need explicit permision from the owner of the code.
To say that Becker doesn't "own" his code is a misunderstanding. Public domain code is not owned by anyone. GPLed code is still owned by either the person who wrote it or whomever the coder transfers ownership. To actually get your code as part of the GNU system, you must transfer the ownership to FSF. Even if he did this (which I can't tell if he did on the drivers in question). It would still behoove him to make sure his code is being used under the terms of the GPL.
The problem is not converting the source/object of Linux code to Solaris code. Heck, gcc itself can make Solarix executables. The problem is what oppenents of the GPL call its "viral" nature. You're not supposed to link GPLed code with non-GPLed code.
Here's the relevant part from paragraph two of the GPL Terms and Conditions:
If identifiable sections of that work are not derived from the Program, and can be reasonably considered
independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when
you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose
permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.
The question is whether the GPLed kernel modules linked into Sun's kernel are considered "independant and seperate works". Sun and Parens says, "Yes, they are seperate and independant". Becker says, "No, the drivers are spinning off threads into the kernel."
I, personally, think Sun's kit is OK. It is like Napster, gcc, or many other examples of conduits. I think Becker's problem is not with the kit, but Sun's choise of shipping his GPL code with their kit for an example conversion. This would be like Napster having a tutorial where they walk you through downloading a Metalica song.
The linking issue has already been addressed. If the code shares the same address space, then it is a derivative work. So, linking GPLed drivers with a proprietary kernel is verboten. I think a good analogy here is glibc. This is "open source" code that it is OK to link against. The reason it is OK is because it is not licensed under the GPL. The LGPL explicitly allows linking against it.
I think what Sun needs to do is write there own LGPL linux kernel driver and then use their kit to convert to a solaris driver. Merely removing Becker's code as their examples would clear this whole thing up.
I assume from your disrespectful tone that you're an atheist, as such people seem to delight in the baseness of
human nature that they believe in. The body is the vehicle of your immortal soul, and a testament to the Will of
the Lord; as such it is worthy of respect even after your soul has depart
I consider myself a devout Christian. I could reveal a long laundry list of qualifications; but in light of Philipians 3 I will not. Life is certainly sacred. Our soul is what makes us different from the plants and other animals around us. Our soul is really what makes us "us". On the other hand, I don't think the Bible says anything about what to do with dead bodies. I think it is mainly custom. As far as the Resurection is concerned, I don't think the state of our bodies when we die (or shortly thereafter) is any impediment to the power of an Almighty God.
I think "respect for the dead" has more to do with what you say and think about them than what you do with their remains. If your father dies, do you respect him more by giving him a $100,000 monument or by living by the ideals your father believed in?
Is this site legit? It seems so, but is difficult to believe. The page appears to be a fragment of a larger site, but there is only one link (and to a bogus domain). Who can vouch for this site's authenticity?
I really like the neural network that netflix.com uses. Basically, you go through and rate some minimum number of movies on a scale of 1 to 5 or "not interested". Their neural net then uses these ratings with criteria about movies to make recommendations. At first, it gets the recommendations wrong. However, eventually it can get really accurate (like 95%). The problem is that it can never get "bang on" accurate due to the limitations of first tear neural nets. The other problem is, of course, the fact that netflix now has a thorough database about me. Not only do they know what movies I watch, but also how well I liked them. I don't really see any way around this if I want the service of getting recommendations.
Free Software development overseas
on
Copyrant
·
· Score: 1
If USA people get trouble because of UCITA whereas the rest of the world enjoy active and prolific Free Software computing, well, maybe your congressmen's neurons will jump to the conclusion that UCITA is bad and should be dropped. I hope so for you. But for now, even if the FSF is based in USA, Free Software is sufficiently well spread to survive without the benevolence of US laws...
I know a significant percentage of free software is not developed in the US. I think in the early days, almost all software (free or otherwise) was developed in the US. It seems like ever since Linus released his OS kernel, Europe and Scandinavia have been developing more and more visible software. While I am glad to see this happen (along with software coming out of other places like Mexico and Brazil) it would make me sad to see the free software developed in the US dwindle down to nothing.
Click-through license at download site
on
Copyrant
·
· Score: 1
And we can always use click-wrap license to protect ourselves: "By accessing the download service, you hereby agree not to sue anyone invloved in any Software you're about to download here. There is no exception nor special clause to this agreement. The source for the code is supplied so you can verify before using if this software is suitable for you, and as you can download it for no cost, you don't place yourself in a customer situation. No reclamations will ever be accepted. If this don't please you, logoff, go elsewhere and don't use our software." or something like that. This seems like a pretty good idea. What do some of you who are more familiar with UCITA think?
This is another reason to support work on free (as in speach) partitioning software. I really like the idea of what Mandrake is doing with their installtion. I wish more distros would do this. Right now for newbies, I have to recommend them buying PartionMagic. None of the newbies I talk with are willing to go whole-sale Linux. Nore are they able to allocate an entire disk. Almost all of them have to re-partition. I certainly like the PartionMagic software, but their license is egregious. I is basically licensed to one person on one machine. That means if I died, my wife would be unable to legally use the software. I talked with them about this and they confirmed how restrictive it is. They sell additional licenses (like site licenses, unlimited users for one machine, and unlimited machines for one user). I told them that this license was begging poeple to violate it. The persons responce was as close to telling me to illegially use the software as it could be without actually condoning it. I then ranted about why they included a license they didn't think their users could abide by.
Re:This it intended to stop dual booting Linux/BSD
on
Copyrant
·
· Score: 1
I have been asked more than once about running Linux off a CD or Zip/LS120 drive. I certainly understand the preformance problems; but, it is something that needs pursuing.
I remember way back talking about The Street Performer Protocol There's a lot of talk around here now that you must either give it away or sell it per copy. I really like the idea of SPP. Basically it goes like this: serialize your work. Give out the first few installments totaly free. Use your favorite format: html, pdf, latex. Then put up some shopping cart thing that says, "contribute as much money as you like, but I'm not releasing the next installment until I get $XX". Then take pledges. When you get enough pledges, charge the credit cards and release the next installment. It's true I wouldn't read Battlefield Earth online. It's way to long to read enjoyably on-line. However, a reasonable novel broken up into reasonable sections could easily be read in this fasion. Hey, you could even try to make a community out of it. Put up discussion boards for readers to comment on the sections as they read them. I really enjoy being part of an "as it happens" project. The question then arises about how much to charge for each section and how large those sections should be.
Re:Reason to be civil
on
DeCSS Update
·
· Score: 2
Not only was I was quite offended by several of the email messages MPAA included, I was amazed at the stupidity of some who originally sent the mail. Come on folks, "I'm going to kill you if you continue with your litigation"? What is that going to gain? All that does is encourage the perception of our community as crack-pots. OK, everyone take a deep breath and go read the Advocacy-mini-HOWTO Actually, reading the sample of mail the MPAA chose to post is an excellent case study in Advocacy. The first couple (at least) are perfect examples of what types of messages not to send. On the other hand, there are several good examples of reasonable requests and statements of possition.
Re:It's not the knowledge, it's the hours ...
on
Too Old To Code?
·
· Score: 1
This is right on. Code about six hours a day; study the others. Somebody moderate this up.
I also have revision 0.4 of the Epox 7KXA (via kx-133). I have not been able to get the onboard sound working with either the kernel via686xxx driver or the alsa driver. Everything compiles and the modules load, but no sound. Also, I have noticed only fair performance from the udma-66 driver. I have a Maxtor Diamond Max 40 30G udma-66 7200rpm drive. I have tweaked it with hdparm and only gotten about 15 MB/sec performace. I have noticed that some of the bios updates have listed IDE performance. I think this is related to Epox's bios stuff.
The same goes for novel etc... publishing. While it's tempting to remove publishers out of the game, so that even unknown authors can get published, the common behaviour on the Net seems to be only looking for famour stuff. This can hardly be helped: for one interesting novel, how many semiliterate ramblings, boring short stories and ersatz sci-fi novels?
I disagree. When it comes to art, I am always looking for something new. Maybe when I was in high school or even college I was stuck on a few particular bands or authors. However, as I get older I am always looking for something new. It's also hard to experiment at $16 a pop for music or $7 a pop for a paperback! Besides that, I have read more junk published by a corporation than I have published by an individual on the net. The hard part in the virtual world and the real world is how to find quality work. That's mainly because quality artists are fairly rare (compared to the population as a whole). What I think something like Napster can provide (note that I have never used it) is some way for people to recommend other artists. I really liked that about Amazon.com. When you found a book you liked, you could see what other books were bought at the same time by other patrons. Basically, it takes the idea of "if you like this, then you may like these also" to the logical extreme of the internet. -- "It's too late for paradise"
The boot up messages are for debugging and status information. OK, fine; add a copyright line too (although I don't really like that). The main problem I have with littering the bootup messages is the limited buffer for dmesg. For those of you who say to increase it, there are kernel considerations not to. Also, not everyone has week long uptimes. Some of us have to dual-boot to share with others. The correct place for credits, sponsors, special thanks and copyrights is in the man pages. OK, fine, you don't like man pages. Put it in the README file or whatever you use for documentation. It's not like I don't "support sponsors". I am all for giving credit where credit is due and also giving thanks to whom you are thankful. As a developer, I really like the idea of someone giving me money to write free software. The point is, everything has its place. I know others have also put spurious information in the boot up messages. I don't necessarily think we are picking on reiserfs here. I personally belive they should all move the copyrights to the man pages.
Slashdot Mirror
I have done this too. I think IRC is great for this. I worked with a guy who was about 700 miles away. We both just grabed a private chanel on EfNet. What I don't get is how to introduce this to a small team of 5 developers that all sit in cubes in the same room.
I would only add one caveat. Everybody hates to see messages with subject lines of "Please help a newbie!".
Yes, monopolies are bad. But what can we do? It seems like somebody has to have a monopoly on broadband. The only alternative would be to give right of way to every "Tom, Dick, or Harry" who wanted to bury cable to your house.
Currently the only people in the US who own right of way are the municipalities, the electric company, the telephone company, the cable company, and the railroads. Other than the municipalities, all of them are government sanctioned monopolies.
Given those choices, I would vote for the municipalities to control broadband. I think the analogy to the interstate system is a good one. The Feds subsidize, but the state and cities pay for it and oversee it. Few places have perfect roads, but almost all places are drivable.
BTW, there is motivation for cities to keep their internet infrastructure up to date. Think about how much competition there is to get large businesses to move to a particular city.
I admit I'm not on the inside of the GCC steering commitee stuff. I don't know if anyone had any gag orders or not. However, even I new RedHat 7.0 was going to be shipping a "bogus" gcc 2.96. Even before the 7.0 beta release, I had upgraded my gcc from rawhide. This was back in early July. How much lead time did you want? Is three months not enough? BTW, I upgraded because directed me to in responce to bug #13547 (a problem with g++ and iostream). I don't see how anyone can criticize RedHat for not disclosing their plans. As soon as they start on the next release, they put the whole thing in rawhide. I don't see MS putting NT64 pre-releases up on the net. I don't see them putting their working copies of anything up on the net. I agree with Bob Y. You can't make legitimate comparisons between MS and RH. The only exception might be that RH is successful in the Linux world and MS is successful in their nitch.
Is this site legit? It seems so, but is difficult to believe. The page appears to be a fragment of a larger site, but there is only one link (and to a bogus domain). Who can vouch for this site's authenticity?
I really like the neural network that netflix.com uses. Basically, you go through and rate some minimum number of movies on a scale of 1 to 5 or "not interested". Their neural net then uses these ratings with criteria about movies to make recommendations. At first, it gets the recommendations wrong. However, eventually it can get really accurate (like 95%). The problem is that it can never get "bang on" accurate due to the limitations of first tear neural nets. The other problem is, of course, the fact that netflix now has a thorough database about me. Not only do they know what movies I watch, but also how well I liked them. I don't really see any way around this if I want the service of getting recommendations.
And we can always use click-wrap license to protect ourselves: "By accessing the download service, you hereby agree not to sue anyone invloved in any Software you're about to download here. There is no exception nor special clause to this agreement. The source for the code is supplied so you can verify before using if this software is suitable for you, and as you can download it for no cost, you don't place yourself in a customer situation. No reclamations will ever be accepted. If this don't please you, logoff, go elsewhere and don't use our software." or something like that. This seems like a pretty good idea. What do some of you who are more familiar with UCITA think?
This is another reason to support work on free (as in speach) partitioning software. I really like the idea of what Mandrake is doing with their installtion. I wish more distros would do this. Right now for newbies, I have to recommend them buying PartionMagic. None of the newbies I talk with are willing to go whole-sale Linux. Nore are they able to allocate an entire disk. Almost all of them have to re-partition. I certainly like the PartionMagic software, but their license is egregious. I is basically licensed to one person on one machine. That means if I died, my wife would be unable to legally use the software. I talked with them about this and they confirmed how restrictive it is. They sell additional licenses (like site licenses, unlimited users for one machine, and unlimited machines for one user). I told them that this license was begging poeple to violate it. The persons responce was as close to telling me to illegially use the software as it could be without actually condoning it. I then ranted about why they included a license they didn't think their users could abide by.
I have been asked more than once about running Linux off a CD or Zip/LS120 drive. I certainly understand the preformance problems; but, it is something that needs pursuing.
I remember way back talking about The Street Performer Protocol There's a lot of talk around here now that you must either give it away or sell it per copy. I really like the idea of SPP. Basically it goes like this: serialize your work. Give out the first few installments totaly free. Use your favorite format: html, pdf, latex. Then put up some shopping cart thing that says, "contribute as much money as you like, but I'm not releasing the next installment until I get $XX". Then take pledges. When you get enough pledges, charge the credit cards and release the next installment. It's true I wouldn't read Battlefield Earth online. It's way to long to read enjoyably on-line. However, a reasonable novel broken up into reasonable sections could easily be read in this fasion. Hey, you could even try to make a community out of it. Put up discussion boards for readers to comment on the sections as they read them. I really enjoy being part of an "as it happens" project. The question then arises about how much to charge for each section and how large those sections should be.
Not only was I was quite offended by several of the email messages MPAA included, I was amazed at the stupidity of some who originally sent the mail. Come on folks, "I'm going to kill you if you continue with your litigation"? What is that going to gain? All that does is encourage the perception of our community as crack-pots. OK, everyone take a deep breath and go read the Advocacy-mini-HOWTO Actually, reading the sample of mail the MPAA chose to post is an excellent case study in Advocacy. The first couple (at least) are perfect examples of what types of messages not to send. On the other hand, there are several good examples of reasonable requests and statements of possition.
This is right on. Code about six hours a day; study the others. Somebody moderate this up.
Email me with what you've tried and I will let you know what I've tried...
I also have revision 0.4 of the Epox 7KXA (via kx-133). I have not been able to get the onboard sound working with either the kernel via686xxx driver or the alsa driver. Everything compiles and the modules load, but no sound. Also, I have noticed only fair performance from the udma-66 driver. I have a Maxtor Diamond Max 40 30G udma-66 7200rpm drive. I have tweaked it with hdparm and only gotten about 15 MB/sec performace. I have noticed that some of the bios updates have listed IDE performance. I think this is related to Epox's bios stuff.
The boot up messages are for debugging and status information. OK, fine; add a copyright line too (although I don't really like that). The main problem I have with littering the bootup messages is the limited buffer for dmesg. For those of you who say to increase it, there are kernel considerations not to. Also, not everyone has week long uptimes. Some of us have to dual-boot to share with others. The correct place for credits, sponsors, special thanks and copyrights is in the man pages. OK, fine, you don't like man pages. Put it in the README file or whatever you use for documentation. It's not like I don't "support sponsors". I am all for giving credit where credit is due and also giving thanks to whom you are thankful. As a developer, I really like the idea of someone giving me money to write free software. The point is, everything has its place. I know others have also put spurious information in the boot up messages. I don't necessarily think we are picking on reiserfs here. I personally belive they should all move the copyrights to the man pages.