A lot has been said here and elsewhere about the idea of taking away the rights of "imaginary people" (aka Corporations) to own patents. I like this idea, but it also doesn't really work out (Those imaginary people are, after all, just slaves to the real people who own them!).
This got me to thinking about the number of real, honest to gosh inventors that get screwed by big Corporations. If you work for a Corporation, you typically sign away all of your rights to invented technologies to that corporation (at least, when they're invented on company time).
So here's an interesting proposal, and y'all can debate it at will:
* As stated, remove the ability for Corporations to hold patents.
* Allow individuals to file patents on things they've invented while working for a Corporation (don't flip out just yet, I'm not trying to screw the Corporations, either!).
* In exchange for the above right to file for the individual, they MUST assign perpetual, free-and-clear USE RIGHTS to the Corporation. These rights, however, are non-exclusive: the actual inventor of said technology may license the technology to anyone else he or she chooses, absent a specific signed contract stating otherwise. By law, the contract must specifically mention the patent number involved.
* Make it illegal for the above mentioned specific contract to be a condition of employment in any way, shape, or form.
The end result that I'm trying for here is simple: individuals invent things. The company can use them. But if the Company wants exclusive rights to the intellectual property that said individual developed, they must PAY FAIRLY FOR IT.
I think this would have the following effects:
* Eliminate corporate patent abuse, as they can no longer hold patents.
* Transform corporate IP litigation into much simpler Contract/License litigation.
* Compensate brilliant employees fairly for their work, thus better distributing wealth where it belongs.
There are (obviously) some potential issues with such a system that would have to be dealt with, but this is just my dinnertime brainstorm presented as text.
Have at it, flame me if you must.:-)
--S
[reading this through, I wonder if maybe we just need to prevent corporations from FILING patents, and still allow them to OWN the patents. Then they can just purchase them from the actual inentor...]
You will note that I never said "stop spam" or "don't stop spam" -- I simply pointed out one of the consequences of making less spam "more difficult" to send.
This is as opposed to stopping ALL spam.
In other words, I was pointing out the flaw in the implied argument that making spam difficult WILL make it stop. It won't. It needs to be damn near impossible, not just difficult.
Either that, or find a way to make it so expensive that the ROI simply sucks for spammers.
But if you do make spamming difficult, then there is less spam, which makes the spam that DOES get through increasingly valuable from the sapmmer's perspective. The ones that get through more often can charge more for their services and get a higher response percentage.
Ohhh, I think I understand where you're going -- basically provide a third-party API that's stable, and the internal API-to-kernel translation layer is extremely thin and can change with every dot release?
That would be a great idea. If I thought I knew enough about the kernel I might take that one on -- but I don't >:)
As to binary drivers being bad, I don't quite put it in those terms. Would I rather have a well-maintained open-source driver? Abso-fraggin'-lutely. But being able to ship binary drivers has a lot of advantages (most notably, providing driver CD's for brand spanking new hardware). The open source drivers can come later if they're needed. But I don't agree that binary drivers are inherently "bad".
...but what rock did this guy crawl out from under?
Not surprisingly, millions of people and corporations are making the switch to Linux or another variant of Unix to save costs or maybe cut their chances of having to resort to piracy to continue to operate but from what we hear daily Microsoft and the news is that Linux doesn't seem to be any less expensive than the Windows platform and for some could limit the very things that they used to take for granted.
That may look like a paragraph, but it's actually just one little sentence......and a painful one to read, at that.
What happened to people being taught to at least write *passable* English? I'm no grammar nazi, but this article is awful.
I've had vastly the opposite experience -- Windows graphics is painfully slow compared to Linux (I have an A64 3600+ with a nVidia 6800GT, too). Unfortunately, if you're not using nVidia or another superbly-manufacturer-supported video card (ATI doesn't qualify, I ditched my 9700Pro in favor the 6800GT for that reason), you may have different results.
IMHO, that's one of the major things that will hold Linux off the desktop: manufacturer hardware supports. That won't happen en masse until either (a) Windows drivers are supported by xorg, the kernel, and everything else (I don't see why it isn't possible given the success of ndiswrapper, but I'm also not a kernel guy), or (b) a stable ABI for drivers is released that doesn't change with every point release of the Linux kernel.
I know Linus has a different view on the latter subject, I just happen to strongly disagree with his position.
Just for completeness, the only the control core in the Cell is PPC-based (and it looks to be a totally new PPC desgin at that, being in-order and all). The other 8 cores are a new design that has nothing to do with the PPC.
Besides, the Cell could be built with the control core running any architecture. It's the headaches with the basic technology that I believe AMD would be interested in -- imagine one of those suckers with an Ath64 control core?:-)
Personally, I want a chip with an Opteron 8xx-based control core and 8 high-speed, in order, 386 cores with mmx and sse added to them...:-) I'm thinking heat would be a problem tho...
PowerPC isn't IBM's only line. How about Cell? If the rumors about Intel's "new direction" prove out, having someone who developed something like the Cell in house could prove to be very fortuitous.
(NOTE: I do realize how hard internet "law" is to enforce, but suspend your disbelief for a moment and humor me).
The only reason you can't block a web site (note: specifically HTTP as opposed to other services) by blocking DNS is because the IP often refers to same target site. There is a solution to this problem.
Require all pornography sites to operative via name-based hosting, where the default server (read: the server that matches the IP address) does NOT contain porn or any information that can be used to reach porn.
Then if you go to http://1.2.3.4/ instead of hugeyams.com, you get some non-offensive page.
Obviously this would be exceedingly difficult to implement on an international scale -- but if there are enough countries interested in requiring it, and the fines/punishment are stiff enough for those caught not conforming to the new law, it would become much harder to "accidentally" run accross porn while surfing.
As an aside, this is not about forcibly blocking a normal adult's ability to access porn (at least here in the USA). It's about giving them the option to choose to eliminate it from the menu, an equally important right given that many people find it highly offensive.
I don't think the Ricer comparison really holds up. In fact, it's pretty much the opposite.
Ricer: Loud (both visually and audibly), screams "I'm fast" (just like its owner), but doesn't actually GO fast in 9 out of 10 cases.
Bleeding-edge Gaming Rig: Virtually silent system that actually does outperform just about anything else on the market.
I am not myself a true hardware addict (I upgrade maybe once every three years on average, and that's only because I actually need the hardware), but I still find your comparison insulting.:p
Sometimes I wish I were my own sibling. My sister keeps getting dibs on my "old" hardware when stuff like this happens and forces the geek in me to go on an upgrade spree.
She will shortly have an Ath64 3500+ and a 6800GT.
Or maybe he was trying to show that all operating systems have associated costs, using the most commonly available target to construct a rather tongue-in-cheek post?
Harping on Linux because there are administrative costs is just plain silly.
Interestingly, one significant cost of adopting any given OS is the ability to hire people that already know the technology. Something tells me qualified Linux people are easier to find than qualified SCO people. Probably cheaper to hire, too.
I might agree with this if we didn't have these things called "Law" and "Government". If you're really so upset about these sorts of things, why aren't you out lobbying your government to do something about it, and getting everyone you know to do the same?
Oh, I forgot, we're all too busy buying cool toys.
Slashdot Mirror
A lot has been said here and elsewhere about the idea of taking away the rights of "imaginary people" (aka Corporations) to own patents. I like this idea, but it also doesn't really work out (Those imaginary people are, after all, just slaves to the real people who own them!).
:-)
This got me to thinking about the number of real, honest to gosh inventors that get screwed by big Corporations. If you work for a Corporation, you typically sign away all of your rights to invented technologies to that corporation (at least, when they're invented on company time).
So here's an interesting proposal, and y'all can debate it at will:
* As stated, remove the ability for Corporations to hold patents.
* Allow individuals to file patents on things they've invented while working for a Corporation (don't flip out just yet, I'm not trying to screw the Corporations, either!).
* In exchange for the above right to file for the individual, they MUST assign perpetual, free-and-clear USE RIGHTS to the Corporation. These rights, however, are non-exclusive: the actual inventor of said technology may license the technology to anyone else he or she chooses, absent a specific signed contract stating otherwise. By law, the contract must specifically mention the patent number involved.
* Make it illegal for the above mentioned specific contract to be a condition of employment in any way, shape, or form.
The end result that I'm trying for here is simple: individuals invent things. The company can use them. But if the Company wants exclusive rights to the intellectual property that said individual developed, they must PAY FAIRLY FOR IT.
I think this would have the following effects:
* Eliminate corporate patent abuse, as they can no longer hold patents.
* Transform corporate IP litigation into much simpler Contract/License litigation.
* Compensate brilliant employees fairly for their work, thus better distributing wealth where it belongs.
There are (obviously) some potential issues with such a system that would have to be dealt with, but this is just my dinnertime brainstorm presented as text.
Have at it, flame me if you must.
--S
[reading this through, I wonder if maybe we just need to prevent corporations from FILING patents, and still allow them to OWN the patents. Then they can just purchase them from the actual inentor...]
You will note that I never said "stop spam" or "don't stop spam" -- I simply pointed out one of the consequences of making less spam "more difficult" to send.
This is as opposed to stopping ALL spam.
In other words, I was pointing out the flaw in the implied argument that making spam difficult WILL make it stop. It won't. It needs to be damn near impossible, not just difficult.
Either that, or find a way to make it so expensive that the ROI simply sucks for spammers.
--S
I've heard of large consicous beings, but one of the Americas actually posts to Slashdot? Daaaamn. :-)
--S
But if you do make spamming difficult, then there is less spam, which makes the spam that DOES get through increasingly valuable from the sapmmer's perspective. The ones that get through more often can charge more for their services and get a higher response percentage.
Nasty, huh.
--S
Actually, she's 33.... and married. >:)
;-)
I can't say if she's hot or not. She's my sister. and all brothers think their sisters are ugly
-S
Ohhh, I think I understand where you're going -- basically provide a third-party API that's stable, and the internal API-to-kernel translation layer is extremely thin and can change with every dot release?
That would be a great idea. If I thought I knew enough about the kernel I might take that one on -- but I don't >:)
As to binary drivers being bad, I don't quite put it in those terms. Would I rather have a well-maintained open-source driver? Abso-fraggin'-lutely. But being able to ship binary drivers has a lot of advantages (most notably, providing driver CD's for brand spanking new hardware). The open source drivers can come later if they're needed. But I don't agree that binary drivers are inherently "bad".
--S
AFAIK, nVidia's drivers aren't wrappers, but rather actual native xorg drivers.
:-)
If I'm wrong, then I agree with your obseration.
--S
Heh. I don't mind it so much in slashdot comments. That was from the article text, where you're supposed to see have things like Editors...
That may look like a paragraph, but it's actually just one little sentence...
What happened to people being taught to at least write *passable* English? I'm no grammar nazi, but this article is awful.
--S
I've had vastly the opposite experience -- Windows graphics is painfully slow compared to Linux (I have an A64 3600+ with a nVidia 6800GT, too). Unfortunately, if you're not using nVidia or another superbly-manufacturer-supported video card (ATI doesn't qualify, I ditched my 9700Pro in favor the 6800GT for that reason), you may have different results.
IMHO, that's one of the major things that will hold Linux off the desktop: manufacturer hardware supports. That won't happen en masse until either (a) Windows drivers are supported by xorg, the kernel, and everything else (I don't see why it isn't possible given the success of ndiswrapper, but I'm also not a kernel guy), or (b) a stable ABI for drivers is released that doesn't change with every point release of the Linux kernel.
I know Linus has a different view on the latter subject, I just happen to strongly disagree with his position.
--S
Just for completeness, the only the control core in the Cell is PPC-based (and it looks to be a totally new PPC desgin at that, being in-order and all). The other 8 cores are a new design that has nothing to do with the PPC.
:-)
:-) I'm thinking heat would be a problem tho...
Besides, the Cell could be built with the control core running any architecture. It's the headaches with the basic technology that I believe AMD would be interested in -- imagine one of those suckers with an Ath64 control core?
Personally, I want a chip with an Opteron 8xx-based control core and 8 high-speed, in order, 386 cores with mmx and sse added to them...
--S
PowerPC isn't IBM's only line. How about Cell? If the rumors about Intel's "new direction" prove out, having someone who developed something like the Cell in house could prove to be very fortuitous.
-S
There are things beyond hardware that require money for growth. Marketing and more employees to handle said growth, to name two. Both are expensive.
--S
Er, not possible on x86 computers? What universe are you living in?
Most x86 motherboards made in recent years support booting from a USB storage device.
--S
(NOTE: I do realize how hard internet "law" is to enforce, but suspend your disbelief for a moment and humor me).
The only reason you can't block a web site (note: specifically HTTP as opposed to other services) by blocking DNS is because the IP often refers to same target site. There is a solution to this problem.
Require all pornography sites to operative via name-based hosting, where the default server (read: the server that matches the IP address) does NOT contain porn or any information that can be used to reach porn.
Then if you go to http://1.2.3.4/ instead of hugeyams.com, you get some non-offensive page.
Obviously this would be exceedingly difficult to implement on an international scale -- but if there are enough countries interested in requiring it, and the fines/punishment are stiff enough for those caught not conforming to the new law, it would become much harder to "accidentally" run accross porn while surfing.
As an aside, this is not about forcibly blocking a normal adult's ability to access porn (at least here in the USA). It's about giving them the option to choose to eliminate it from the menu, an equally important right given that many people find it highly offensive.
--S
And thus, the users get held hostage for the benefit of a single librarian...
--S
His username speaks volumes. 'nuff said. :-)
--S
Talking to the wrong end of the supply chain here, man. Now, you could always ask my sister for HER hand-me-downs... :-)
--S
Dude, you're channeling me. :-)
--S
I don't think the Ricer comparison really holds up. In fact, it's pretty much the opposite.
:p
Ricer: Loud (both visually and audibly), screams "I'm fast" (just like its owner), but doesn't actually GO fast in 9 out of 10 cases.
Bleeding-edge Gaming Rig: Virtually silent system that actually does outperform just about anything else on the market.
I am not myself a true hardware addict (I upgrade maybe once every three years on average, and that's only because I actually need the hardware), but I still find your comparison insulting.
--S
I actually have a semi-valid reason for wanting the upgrade besides the "it's shiny" factor ;-)
:p
Simply put: my 6800GT doesn't have dual DVI. I have a pair of Samsung 213T's, and I want them both on DVI, dammit. The one on VGA annoys me.
Of course, looking at this, it seems as though PCI Express is the only option. That means a motherboard upgrade, too... grrrr.
Looks like my sis is getting a shiney new computer.
Maybe I should say "Screw it" and just go to a dual-proc Opteron board and slap a couple of dual-core chips in it...
--S
Sometimes I wish I were my own sibling. My sister keeps getting dibs on my "old" hardware when stuff like this happens and forces the geek in me to go on an upgrade spree.
:p
She will shortly have an Ath64 3500+ and a 6800GT.
Must be nice.
--S
Or maybe he was trying to show that all operating systems have associated costs, using the most commonly available target to construct a rather tongue-in-cheek post?
Harping on Linux because there are administrative costs is just plain silly.
Interestingly, one significant cost of adopting any given OS is the ability to hire people that already know the technology. Something tells me qualified Linux people are easier to find than qualified SCO people. Probably cheaper to hire, too.
--S
That's only because he has to stop and give the "actors" direction...
I might agree with this if we didn't have these things called "Law" and "Government". If you're really so upset about these sorts of things, why aren't you out lobbying your government to do something about it, and getting everyone you know to do the same?
Oh, I forgot, we're all too busy buying cool toys.