I'm guessing the GPLv2 at least ensured him some level of "keep my open source stuff open", whereas the BSD might not have.
I kind of follow his argument, from a business perspective regarding its product's quality reputation ("guaranteeing it will function as advertised").
I personally don't see a problem with GPLv3 including such a restriction. To me, it looks to just mean that coders who want their own code on the GPLv3 license will not likely see it used on a hardware product from a business that wants to include such restrictions... they'll have to look for other code to use in their product (maybe GPLv2 code, maybe Linux). At the end of the day, the GPLv3 coder made his choice, the GPLv2 coder made his choice, and the business made its choice. Looks to me like having both GPLv2 and GPLv3 available in the world just means more choices.
I'm still smirking at the fact that Linus' opinion of each GPLv3 draft seems to be the first litmus test that all the talkers look to, as in "will you buy v3 now, Mr. Linus?". Linux is one project (albeit very big and very important all around), but as its one project, Linus has to consider the versions' different impacts on the lifeblood of his project. Other v2 projects will be looking at v3 with the same scrutiny before choosing whether to take the plunge or not.
I think we need to give Linus a break from these repeated knocks at his door. Are we hounding the Apache and Mozilla guys like this, "will you buy v3 now, Mr. Mozache?".
That cost estimate is pretty accurate. I went through getting a quote on installing solar panels on my $160k house, to just replace 50% of electric usage, and it priced at >$270k. It calculated out as "pays for itself after 42 years".
Contrast that with considering replacing a traditional electric heat pump with a geothermal system. About twice the cost to install, but electric costs associated with using it drop tremendously, as in $225/month summer dropping to $80/month. This calculates out to paying for itself after only about five years.
I recommend reading Ricardo Semler's book "The Seven Day Weekend" for an interesting way to manage the employees of a large company. It really makes the typical authoritarian CEO/Board style rather laughable.
The last fifteen years of listening to CNN and Wall Street makes me think that unless YOU are the one who built the company up from your garage into a $10B business, then YOUR interest in the company as CEO/Board is what YOUR cut will be after the profits, after the layoffs, and after the sellout to bigger company that wants to eat you. Your heart is not in it, and therefore everyone in the company that suffers from your "leadership" is not in your heart either.
I would now like to see the SEC and/or the major SCO stockholders (non-MS obviously) hold the executive team accountable for this major company loss of money, business, and most of all CREDIBILITY in the technology market. Those Linux-using companies that SCO intimidated into buying indemnity licenses should further pursue legal action to get their costs back, with punitive damages to boot.
In my mind, this strategy of theirs fits right in line with the same kind of covert accounting strategies that Enron/Worldcom/etc were investigated for, where the EXECUTIVES themselves were held accountable to the tune of big $$$ and jail time.
I've sent a note to the ACLU via its website to see if they can get involved. I REALLY hate to see all the "librarian was wrong" talk in that article. What people don't seem to understand is that the ends do NOT justify the means. Maybe today you cheer because the police forced their way in to immediately capture a murder suspect "before he could escape"... tomorrow that wrong suspect may be you, because someone that dislikes you called in an anonymous tip on a local murder and fingered you... a judge would never grant a warrant on that alone, but if the police don't think they have to be hassled by getting the warrant...
Things like this always remind me of my favorite line in "JFK" - "A patriot must always be ready to defend his country against his government."
Don't get me wrong... I'm not knocking the production capacity of any of the distributions. I've been living on Gentoo for over 3 years now. No one has to convince me that any distribution is capable.
People who are comfortable with managing OSs on servers based mostly/solely on their own competence will be found in any camp. But for a company, whose business would be injured by a lengthy outage, having the guaranteed, paid-for support available for their production OS is the final safety net that lies under the level of whether or not TuxDisciple can fix his box's current problem... if he can, great... if not, call support. Without that support, what does the business do when TuxDisciple can't figure the problem out?
For studies like these, I really think you have to remove from the equation any variables related to systems without support contracts. When you include systems without support contracts, you're having to include all the downtime involved when the admin person had to do all the research to solve the problem. Put that kind of time-to-repair against the same kind of outage where support was called for help.
Considering that last example further, I'll bet Linux admins are more likely to dig deeper into a problem themselves and stay there longer before calling support that the equivalent Windows admins, probably because Linux allows its admins to actually dig deeper. So, the Linux admins diligence to solve problems themselves may be hurting their "mean-time-to-restoral" measurements when compared to the Windows admins.
I think the only Linux distributions that should be considered in a study like this are the ones that have corporate support contracts available. Bringing in our experiences with other such distributions (like Fedora) isn't relevant because sensible companies should NOT be allowing any OS into their production corporate infrastructure without proper support available.
If you depend on your savvy hacker to set your production boxes up on Fedora, where is your company going to be when Mr. Savvy moves on? Where's your reliable production support now? Support contracts are insurance against your own employee knowledge base as much as insurance against the server's hardware/software.
I have to agree. Obviously it's nice to think we can hide in our hacker darkroom and pretend all the legal crap is only in reruns of "Law & Order". Open source groups that are large/smart enough to prepare themselves against possible legal action help take some of the uncertainty out of contributors' minds, allowing them to focus on true development contributions.
Unfortunately, it smells like some of the occasional hacker anarchy could prevail here, just because of some suddenly rampant "kill all the lawyers, kill 'em tonight" attitudes making the most noise.
I hope Gentoo manages to survive after Debian has imploded itself.
Why not just plomp down an AMD PIC or two per classroom and use the same time sharing process?
Given that I unfortunately doubt there will millions of orders that translate into 2Xmillions of donated laptops, I imagine the donated units will have to be divied out such that there will indeed only be one or two per classroom.
I suppose if we really enjoy debating the pros/cons of any/all methods of helping, we could form several executive committees and pick up several sets of poor corporate executives that are currently being paid below the $1M line. Would that be more productive?
Given that the pledge website appears to have been/.'ed right now, it may end up that our enjoyable debating has indeed put the brakes on the effort.
Server suicide is right... I'm watching the paint dry and the grass grow while just trying to go forward one single web page... just when I thought I finally got a successful "Check Out", I'm stuck in the mud trying to "Create a New Account" !@#$%^
Slashdot Mirror
I'm guessing the GPLv2 at least ensured him some level of "keep my open source stuff open", whereas the BSD might not have.
I kind of follow his argument, from a business perspective regarding its product's quality reputation ("guaranteeing it will function as advertised").
I personally don't see a problem with GPLv3 including such a restriction. To me, it looks to just mean that coders who want their own code on the GPLv3 license will not likely see it used on a hardware product from a business that wants to include such restrictions... they'll have to look for other code to use in their product (maybe GPLv2 code, maybe Linux). At the end of the day, the GPLv3 coder made his choice, the GPLv2 coder made his choice, and the business made its choice. Looks to me like having both GPLv2 and GPLv3 available in the world just means more choices.
I'm still smirking at the fact that Linus' opinion of each GPLv3 draft seems to be the first litmus test that all the talkers look to, as in "will you buy v3 now, Mr. Linus?". Linux is one project (albeit very big and very important all around), but as its one project, Linus has to consider the versions' different impacts on the lifeblood of his project. Other v2 projects will be looking at v3 with the same scrutiny before choosing whether to take the plunge or not.
I think we need to give Linus a break from these repeated knocks at his door. Are we hounding the Apache and Mozilla guys like this, "will you buy v3 now, Mr. Mozache?".
That cost estimate is pretty accurate. I went through getting a quote on installing solar panels on my $160k house, to just replace 50% of electric usage, and it priced at >$270k. It calculated out as "pays for itself after 42 years".
Contrast that with considering replacing a traditional electric heat pump with a geothermal system. About twice the cost to install, but electric costs associated with using it drop tremendously, as in $225/month summer dropping to $80/month. This calculates out to paying for itself after only about five years.
I recommend reading Ricardo Semler's book "The Seven Day Weekend" for an interesting way to manage the employees of a large company. It really makes the typical authoritarian CEO/Board style rather laughable.
The last fifteen years of listening to CNN and Wall Street makes me think that unless YOU are the one who built the company up from your garage into a $10B business, then YOUR interest in the company as CEO/Board is what YOUR cut will be after the profits, after the layoffs, and after the sellout to bigger company that wants to eat you. Your heart is not in it, and therefore everyone in the company that suffers from your "leadership" is not in your heart either.
"Slow but sure, to win the race", says the quietly confident tortoise...
I would now like to see the SEC and/or the major SCO stockholders (non-MS obviously) hold the executive team accountable for this major company loss of money, business, and most of all CREDIBILITY in the technology market. Those Linux-using companies that SCO intimidated into buying indemnity licenses should further pursue legal action to get their costs back, with punitive damages to boot.
In my mind, this strategy of theirs fits right in line with the same kind of covert accounting strategies that Enron/Worldcom/etc were investigated for, where the EXECUTIVES themselves were held accountable to the tune of big $$$ and jail time.
I did, I did! To arms, /.ers... err, to keys... (yeah that's better).
Tell the library they should stand behind their librarian, since she stood up for the proper way to handle this whole headache.
I've sent a note to the ACLU via its website to see if they can get involved. I REALLY hate to see all the "librarian was wrong" talk in that article. What people don't seem to understand is that the ends do NOT justify the means. Maybe today you cheer because the police forced their way in to immediately capture a murder suspect "before he could escape"... tomorrow that wrong suspect may be you, because someone that dislikes you called in an anonymous tip on a local murder and fingered you... a judge would never grant a warrant on that alone, but if the police don't think they have to be hassled by getting the warrant... Things like this always remind me of my favorite line in "JFK" - "A patriot must always be ready to defend his country against his government."
I'm more inclined to think they hope to achieve a monetary settlement out of it all, rather than anything else.
The article indicated it has been moved off the PC body and put onto the actual external power supply.
Don't get me wrong... I'm not knocking the production capacity of any of the distributions. I've been living on Gentoo for over 3 years now. No one has to convince me that any distribution is capable.
People who are comfortable with managing OSs on servers based mostly/solely on their own competence will be found in any camp. But for a company, whose business would be injured by a lengthy outage, having the guaranteed, paid-for support available for their production OS is the final safety net that lies under the level of whether or not TuxDisciple can fix his box's current problem... if he can, great... if not, call support. Without that support, what does the business do when TuxDisciple can't figure the problem out?
For studies like these, I really think you have to remove from the equation any variables related to systems without support contracts. When you include systems without support contracts, you're having to include all the downtime involved when the admin person had to do all the research to solve the problem. Put that kind of time-to-repair against the same kind of outage where support was called for help.
Considering that last example further, I'll bet Linux admins are more likely to dig deeper into a problem themselves and stay there longer before calling support that the equivalent Windows admins, probably because Linux allows its admins to actually dig deeper. So, the Linux admins diligence to solve problems themselves may be hurting their "mean-time-to-restoral" measurements when compared to the Windows admins.
I think the only Linux distributions that should be considered in a study like this are the ones that have corporate support contracts available. Bringing in our experiences with other such distributions (like Fedora) isn't relevant because sensible companies should NOT be allowing any OS into their production corporate infrastructure without proper support available.
If you depend on your savvy hacker to set your production boxes up on Fedora, where is your company going to be when Mr. Savvy moves on? Where's your reliable production support now? Support contracts are insurance against your own employee knowledge base as much as insurance against the server's hardware/software.
I have to agree. Obviously it's nice to think we can hide in our hacker darkroom and pretend all the legal crap is only in reruns of "Law & Order". Open source groups that are large/smart enough to prepare themselves against possible legal action help take some of the uncertainty out of contributors' minds, allowing them to focus on true development contributions.
Unfortunately, it smells like some of the occasional hacker anarchy could prevail here, just because of some suddenly rampant "kill all the lawyers, kill 'em tonight" attitudes making the most noise.
I hope Gentoo manages to survive after Debian has imploded itself.Given that I unfortunately doubt there will millions of orders that translate into 2Xmillions of donated laptops, I imagine the donated units will have to be divied out such that there will indeed only be one or two per classroom.
I suppose if we really enjoy debating the pros/cons of any/all methods of helping, we could form several executive committees and pick up several sets of poor corporate executives that are currently being paid below the $1M line. Would that be more productive?
Given that the pledge website appears to have beenServer suicide is right... I'm watching the paint dry and the grass grow while just trying to go forward one single web page... just when I thought I finally got a successful "Check Out", I'm stuck in the mud trying to "Create a New Account" !@#$%^