I just put a firewall rule in locking my PS3 off the Internet. I'm tired of everything that wants to find stuff on the 'net. I don't care, I don't play on-line, and I can switch the Internet back on when I actually want to use it.
It's particularly annoying that they don't have a "Never connect for BD-LIVE content" in th Blu-Ray setup. There's only "Ask every time" and "Always allow". But why would I want an Internet connection to watch a damn movie?!
I'd just pull the Ethernet plug, except then media streaming on my LAN won't work either. So the router's firewall will do.
And when it goes wrong, we-the-customer all pay, and the companies that saved Big Bucks by using Taiwanese and Chinese manufacturing and parts supplies get to just say, "Sorry, the warranty is over, buy a new one!"
The LGPL code that remains can't be closed. But it still isn't the blanket get-out-of-open-source-free card that a lot of people think it is.
For example, libraries based on LGPL libraries remain LGPL and therefore open source. (LGPL 2.1 #2.)
Applications must use a pluggable library system (such as shared objects with a dynamic or runtime linker), or provide re-linkable binaries, otherwise the application must be open source. (LGPL 2.1 #6. Failing to qualify for an exception means you go back to the other requirements in the license.) BTW, C++ with templates makes qualifying under this section pretty much impossible, as the templates must appear as source in the translation unit that references them. (Or your library only provides fully-instantiated templates of pre-defined types, which kind of defeats the point.)
Remember, the GNU licenses are about the end-user getting freedom to modify the code they receive, and not the developer/vendor having rights.
I don't know about the network providers, but I know IBM Rational's ClearQuest product has _absolutely no provision for testing_. Sure, you can designate a test database, but that's not what I mean, and I don't think it's what you mean, either.
What _I_ mean by testing is taking a snapshot of the production system and then run our planned changes against it.
This product has no provision for doing that. And the DB schemas are opaque (or at least undocumented), and doing a copy of the underlying database does NOT result in the copy being used: it goes back to the original. (Ignoring the thousands of files you need 'cause the blasted thing is written under Eclipse now.)
They aren't alone. Near as I can tell, the trend to "rapid deployment" has made creating a representative test copy nearly impossible.
The best thing I've been able to do with some products is dump all the filesystems on the original server, restore them into a VM, and put the VM into "Host-only" networking mode so it can't get to the original server.
How about opening the POP3S and IMAPS ports, instead? And put some sensible password restrictions in. And use TLS SMTP with mandatory login on the SMTPS port. I have users on Palm Treo, Outlook, Apple Mail, Thunderbird, iPod Touch, and more.
The server I run can also be set to require client SSL certificates. I believe the iPhone "Profile" feature is where this would be set up for corporate clients; when I imported my SSL CA certificate, it created a profile automatically.
As was (once again) pointed out on Groklaw recently, this sort of language is a restriction that is incompatible with the GPL. (GPLv2 section 6, much more explicit about patents in GPLv3 section 11.)
Far safer to avoid Microsoft patented technology than to rely on such a promise.
While definitely a heavyweight compared to VLC, I've been using XBMC for Mac instead of VLC for fiddling with stuff before it goes to transcoding. VLC has better direct navigation of DVDs, but XBMC seems to be a bit more robust at playing them. And anyway, it's going to wind up played on XBMC for Linux connected to the big screen.
Frankly, if Apple's DVD player would just ignore prohibited user operations, I'd use that. I rarely watch a DVD from even a full rip these days, they get unwrapped, ripped, transcoded, and the physical disc put in storage before ever making it to the home theatre screen.
But XBMC can also handle Blu-Ray transport streams which have LPCM and DTS audio tracks, VLC choked on those. (Though it seemed happy with ones which had AC3 audio; though it could be some other issue about those discs which XBMC can deal with and VLC can't. Obviously, these are ripped to HDD, 'cause there's no BD+ or AACS support for Mac. That also means no waiting around for 47 different copyright warnings and previews.)
On the other hand, I'd hate to lose VLC, and I've been learning Objective C... if only I had more spare time.
Microsoft hates UTF-8, which was designed carefully so it wouldn't be damaged by existing 8-bit clean code. Microsoft actually uses UCS-16LE as an on-disk and on-wire format (not just an in-memory format).
This isn't the only case where they write out in-memory structures directly to disk; if you've been following the OOXML debacle at all, it's littered with in-memory blogs with XML decoration.
OS X apps that use the system widget set should all support scrolling without focus. The Firefox and Thunderbird at 2.0 were a bit of a roll-your-own thing in some areas. This was visible in a jarring discontinuity in UI conventions. Like scrolling, and how selection and arrow + qualifier keys behave in text boxes, and so on.
Another little-known MacOS no-focus-interaction thing is, Cmd-Drag moves a window without activating it. Handy for making part of another app visible without affecting your current window (and its selections or edit state, especially during Finder rename).
I worked at Big Blue in the 90s, and they had a fixed rate per "shift" for being on-call, they called it "standby". Though, at the time, I was the first one in the particular office who said, "How much do I get?" when asked to carry a pager.
This money was very, very appealing right out of University. Six years later, since the rate is independent of your salary (or even salary band or rank... actually or even inflation), it didn't seem quite so impressive.
An "on-call" shift cannot be a shift where you are otherwise expected to be in the office. (Or working normally, given the "flexiplace" work at home plan.) Should you actually get called, then you bill time worked, in addition to the standby pay for that shift. Compensation rates follow the normal overtime rules for your jurisdiction. (Time-and-a-half, with a four hour "deductible" for Ontario. So your first four hours of overtime aren't paid.)
My PERSONAL rule is, as long as I have to worry about the company needing me, they need to be paying something. Like if I can't go out Friday night and get drunk, they need to be paying. (Normally, I'd expect being sober by Monday morning is no problem. Being sober for an emergency page at 1 AM Saturday morning.... not so easy.)
Or if it interferes with vacation plans or anything like that. Or even being able to go see a movie.
zone "doubleclick.net" IN { type master; file "master.d/null"; }; zone "gostats.com" IN { type master; file "master.d/null"; }; zone "hitbox.com" IN { type master; file "master.d/null"; }; zone "google-analytics.com" IN { type master; file "master.d/null"; };
and so on. 'master.d/null' contains SOA and NS records identifying your own DNS server, and nothing else.
Any domain I've got in there is either something that has caused repeated page-load delays, or "cookie bombs" where 30 "The site somesiteotherthantheoneyouaretryingtovisit.com wishes to set a cookie" to come up before I get to say "no" to the first one.
Merely providing ads isn't enough to get banned from my LAN.
Just out of curiosity, what files do you have that you care about which are NOT owned by your user account?
All my photos, videos, music files, word processing and spreadsheet files, FileMaker databases (yes I'm old), source code, and so on are all owned by me.
The stuff that's protected from my account is the stuff I can recover with the Fedora or Mac OS DVD and a visit to the appropriate patch site for updates.
Of course, Time Machine, Retrospect, and cron+dumpe2fs make sure stuff I care about is in more than one place.
The phrase you're looking for is "vendor lock-in". You can only access your files and your data with The Chosen And Approved Tools.
And, frankly, once your company was on the annual maintenance bus, there's no way the vendor would want to provide extra features, like generic export... they know you're screwed if you try to leave.
Occasionally, that tactic fails, and they lose a maintenance stream. But it usually works. Especially if the next software release has Magic Special Feature everyone at the client site wants. (Which will need all new servers, and upgraded workstations, and more SAN, and better networking, and when you finally get it deployed, you'll find out they lied right at your C?O's face. Again. And yet he's not held accountable for continuing to pay for their crap.)
Note the difference between "ones you've bought" and "ones provided by the cable Internet vendor".
My experience has been, any software provided by an ISP is to be treated as worse than malware.
Since I never used 16-bit Windows, I never understood "Internet Dialler" software that Windows users seemed to always install from their ISP... and was always the first thing in the way when trying to fix a busted system. But it has served to convince legions that ISP-provided software is necessary to get on the Internet. (Whereas even Windows 95 had enough stuff to dial a modem and set up PPP or SLIP. For non-PPPoE broadband--typically cable--you need no extra software on any Ethernet capable OS.)
More recently, "Internet Security Suites" provided by ISP should be never installed. If found installed, it should be removed by re-formatting the system from read-only media. (Should of serious registry hacking, that's the only way to get rid of the stuff Bell Sympaticrap provides.)
So I'm not surprised they ship modems or routers with "PWN ME!" as the default setting.
Buying your own router immediately puts you in the top percentiles of "tech skill".
It's possible it's even worse for guys, because they AREN'T expected to be embarrassed about how they look when going to the gym, so no-one gives them any advice about what to do about it.
Like; wear baggy clothes; pick a "utilitarian" gym that's just got normal work-a-day people in it, neither the pretty cardio-bunnies nor the hard-core bodybuilders; everyone else is there to get their workout done, too; and so on.
But most importantly: You've got to start somewhere.
On the other hand, I'm a big fan of "everyday" kinds of exercise. If you drive, start parking at the back of the lot. If you take the bus or subway, get off a stop early, then two, then three. That sort of thing. That can be difficult in Suburbia, but it's just that much more important to make the effort. At the big-box mall, don't drive from store to store, for example.
For a little more, you can do a lot at home with some dumbbells and a simple bench. Don't get lured in by fancy chrome rigs. Check out Craigslist and other for-sale sites. Classifieds. Freecycle. There's a lot of people who give up on their exercise plans....
Whatever the choice, the important thing, and the very hardest thing, is to get up off my fat butt and actually do it. I mean your butts.
When a traffic signal in Toronto loses contact with the computer, it goes to flashing 4-way red, to remind people it's an all-way-stop now. They only go to "no signal" if there's a power failure. (In which case, they're still all-way-stop but despite the radio saying so every time there's a power failure, 70% of drivers just blow through a dark traffic light at speed.)
I think all-way-stop on multilane roads are MUCH MUCH MUCH MUCH worse than untimed lights. People have no idea how to deal with multilane roads that have a stop sign; almost no-one will let a left turn through, all sorts of problems like that.
Compare "Major Appliances" to "Appliances". Hand mixer, blender, kettle, coffee maker, bread machine, coffee grinder, countertop boiler, ice cream maker, and so on. Plenty are made without a grounded cord.
Ground isn't "instant safety", though. Depending on circumstances, ground can make a fault worse. That's why the shift to double-insulated power tools with ungrounded plugs. (Say on a drill: the chuck is insulated from the motor, and the motor is insulated from the housing. So if you drill into a live wire, the circuit DOES NOT COMPLETE through the power tool to either ground or you--or both. (If you hit a Big Cable, it will be too much for the 16 gauge ground, so there'll be plenty of current to go through you, too.))
Mind you, I had a paper shredder "fail dangerous" when the double-insulating piece that insulated the cutters from the motor failed... and the motor fell out of its mounts... and turned on... and shorted hot to the control panel. (That unit should not have received ULC and CSA safety approval with a design that brittle. The motor should have been bolted to the case, not "propped" in place by a plastic widget.)
Well, one thing, iSCSI has to transfer disk blocks, and NFS actually knows about files. So the NFS server can do read-ahead knowing the structure of a directory, or the inode, or the file itself. But iSCSI may not read-ahead in the right pattern, because first you have to read the directory blocks, then the inode blocks, then the file blocks, and they may not be adjacent and sequential.
You may also need to transfer more than you need using iSCSI. "Open this file for me" on NFS doesn't take a very large packet at all, as the client doesn't need to know anything about the server's underlying filesystem.
Right tool for the job and all that....
(This is not to be taken as an endorsement of NFS.)
And, of course, distributions are STILL free to put whatever scheduler they feel like maintaining in Linux. Linus doesn't have to accept it for a distribution to include it.
That's the great thing... and sometimes the worst thing... about Linux. You can bolt in anything you're prepared to code, and it doesn't MATTER if the "vendor" doesn't want it. That just means you have to handle merges and testing on your own when new kernels come out.
For starters, people sweat and computers do not. So, airflow helps cool people by increasing evaporation, in addition to direct thermal transfer. Even when you think you aren't sweating, your skin is still moist and evaporative cooling still works.
Unless someone invents a CPU swamp cooler, that's just not happening on a computer. You do need airflow to keep the hot air from remaining close to the hot component (this can be convection or forced), but you don't get that extra... let's call it "wind chill" effect that humans feel.
His lawsuit and license threats kept our customers from switching over to Red Hat Enterprise Linux systems. Consequently, we have to support a mish-mash of HP-UX, AIX, and Solaris (on two CPUs). Linux is finally catching on, but it's been delayed, and gave Solaris on Intel an edge it did not deserve, which has us running duplicate systems that shouldn't be necessary, and duplicate test load.
I'm also left wondering if OpenSolaris is licensed correctly, given who Sun bought some of the licenses from. That, however, I don't have to worry about professionally, as our customers won't run an OS without a big, juicy support contract.
Slashdot Mirror
This article reminded me....
I just put a firewall rule in locking my PS3 off the Internet. I'm tired of everything that wants to find stuff on the 'net. I don't care, I don't play on-line, and I can switch the Internet back on when I actually want to use it.
It's particularly annoying that they don't have a "Never connect for BD-LIVE content" in th Blu-Ray setup. There's only "Ask every time" and "Always allow". But why would I want an Internet connection to watch a damn movie?!
I'd just pull the Ethernet plug, except then media streaming on my LAN won't work either. So the router's firewall will do.
It could be worse. It could be a movie where the Bat Computer is nothing but a 20th Anniversary Macintosh with Kaleidescope to re-skin the UI.
And when it goes wrong, we-the-customer all pay, and the companies that saved Big Bucks by using Taiwanese and Chinese manufacturing and parts supplies get to just say, "Sorry, the warranty is over, buy a new one!"
The LGPL code that remains can't be closed. But it still isn't the blanket get-out-of-open-source-free card that a lot of people think it is.
For example, libraries based on LGPL libraries remain LGPL and therefore open source. (LGPL 2.1 #2.)
Applications must use a pluggable library system (such as shared objects with a dynamic or runtime linker), or provide re-linkable binaries, otherwise the application must be open source. (LGPL 2.1 #6. Failing to qualify for an exception means you go back to the other requirements in the license.) BTW, C++ with templates makes qualifying under this section pretty much impossible, as the templates must appear as source in the translation unit that references them. (Or your library only provides fully-instantiated templates of pre-defined types, which kind of defeats the point.)
Remember, the GNU licenses are about the end-user getting freedom to modify the code they receive, and not the developer/vendor having rights.
I don't know about the network providers, but I know IBM Rational's ClearQuest product has _absolutely no provision for testing_. Sure, you can designate a test database, but that's not what I mean, and I don't think it's what you mean, either.
What _I_ mean by testing is taking a snapshot of the production system and then run our planned changes against it.
This product has no provision for doing that. And the DB schemas are opaque (or at least undocumented), and doing a copy of the underlying database does NOT result in the copy being used: it goes back to the original. (Ignoring the thousands of files you need 'cause the blasted thing is written under Eclipse now.)
They aren't alone. Near as I can tell, the trend to "rapid deployment" has made creating a representative test copy nearly impossible.
The best thing I've been able to do with some products is dump all the filesystems on the original server, restore them into a VM, and put the VM into "Host-only" networking mode so it can't get to the original server.
How about opening the POP3S and IMAPS ports, instead? And put some sensible password restrictions in. And use TLS SMTP with mandatory login on the SMTPS port. I have users on Palm Treo, Outlook, Apple Mail, Thunderbird, iPod Touch, and more.
The server I run can also be set to require client SSL certificates. I believe the iPhone "Profile" feature is where this would be set up for corporate clients; when I imported my SSL CA certificate, it created a profile automatically.
Hmmm....
As was (once again) pointed out on Groklaw recently, this sort of language is a restriction that is incompatible with the GPL. (GPLv2 section 6, much more explicit about patents in GPLv3 section 11.)
Far safer to avoid Microsoft patented technology than to rely on such a promise.
While definitely a heavyweight compared to VLC, I've been using XBMC for Mac instead of VLC for fiddling with stuff before it goes to transcoding. VLC has better direct navigation of DVDs, but XBMC seems to be a bit more robust at playing them. And anyway, it's going to wind up played on XBMC for Linux connected to the big screen.
Frankly, if Apple's DVD player would just ignore prohibited user operations, I'd use that. I rarely watch a DVD from even a full rip these days, they get unwrapped, ripped, transcoded, and the physical disc put in storage before ever making it to the home theatre screen.
But XBMC can also handle Blu-Ray transport streams which have LPCM and DTS audio tracks, VLC choked on those. (Though it seemed happy with ones which had AC3 audio; though it could be some other issue about those discs which XBMC can deal with and VLC can't. Obviously, these are ripped to HDD, 'cause there's no BD+ or AACS support for Mac. That also means no waiting around for 47 different copyright warnings and previews.)
On the other hand, I'd hate to lose VLC, and I've been learning Objective C... if only I had more spare time.
Microsoft hates UTF-8, which was designed carefully so it wouldn't be damaged by existing 8-bit clean code. Microsoft actually uses UCS-16LE as an on-disk and on-wire format (not just an in-memory format).
This isn't the only case where they write out in-memory structures directly to disk; if you've been following the OOXML debacle at all, it's littered with in-memory blogs with XML decoration.
OS X apps that use the system widget set should all support scrolling without focus. The Firefox and Thunderbird at 2.0 were a bit of a roll-your-own thing in some areas. This was visible in a jarring discontinuity in UI conventions. Like scrolling, and how selection and arrow + qualifier keys behave in text boxes, and so on.
Another little-known MacOS no-focus-interaction thing is, Cmd-Drag moves a window without activating it. Handy for making part of another app visible without affecting your current window (and its selections or edit state, especially during Finder rename).
I worked at Big Blue in the 90s, and they had a fixed rate per "shift" for being on-call, they called it "standby". Though, at the time, I was the first one in the particular office who said, "How much do I get?" when asked to carry a pager.
This money was very, very appealing right out of University. Six years later, since the rate is independent of your salary (or even salary band or rank... actually or even inflation), it didn't seem quite so impressive.
An "on-call" shift cannot be a shift where you are otherwise expected to be in the office. (Or working normally, given the "flexiplace" work at home plan.) Should you actually get called, then you bill time worked, in addition to the standby pay for that shift. Compensation rates follow the normal overtime rules for your jurisdiction. (Time-and-a-half, with a four hour "deductible" for Ontario. So your first four hours of overtime aren't paid.)
My PERSONAL rule is, as long as I have to worry about the company needing me, they need to be paying something. Like if I can't go out Friday night and get drunk, they need to be paying. (Normally, I'd expect being sober by Monday morning is no problem. Being sober for an emergency page at 1 AM Saturday morning.... not so easy.)
Or if it interferes with vacation plans or anything like that. Or even being able to go see a movie.
Set up your own DNS, and do:
zone "doubleclick.net" IN { type master; file "master.d/null"; };
zone "gostats.com" IN { type master; file "master.d/null"; };
zone "hitbox.com" IN { type master; file "master.d/null"; };
zone "google-analytics.com" IN { type master; file "master.d/null"; };
and so on. 'master.d/null' contains SOA and NS records identifying your own DNS server, and nothing else.
Any domain I've got in there is either something that has caused repeated page-load delays, or "cookie bombs" where 30 "The site somesiteotherthantheoneyouaretryingtovisit.com wishes to set a cookie" to come up before I get to say "no" to the first one.
Merely providing ads isn't enough to get banned from my LAN.
Just out of curiosity, what files do you have that you care about which are NOT owned by your user account?
All my photos, videos, music files, word processing and spreadsheet files, FileMaker databases (yes I'm old), source code, and so on are all owned by me.
The stuff that's protected from my account is the stuff I can recover with the Fedora or Mac OS DVD and a visit to the appropriate patch site for updates.
Of course, Time Machine, Retrospect, and cron+dumpe2fs make sure stuff I care about is in more than one place.
Because the real world is a line from (-INF,0) to (+INF,0). The imaginary world is the entire complex plane EXCEPT that line where y=0i.
The phrase you're looking for is "vendor lock-in". You can only access your files and your data with The Chosen And Approved Tools.
And, frankly, once your company was on the annual maintenance bus, there's no way the vendor would want to provide extra features, like generic export... they know you're screwed if you try to leave.
Occasionally, that tactic fails, and they lose a maintenance stream. But it usually works. Especially if the next software release has Magic Special Feature everyone at the client site wants. (Which will need all new servers, and upgraded workstations, and more SAN, and better networking, and when you finally get it deployed, you'll find out they lied right at your C?O's face. Again. And yet he's not held accountable for continuing to pay for their crap.)
Dr. Watson, I need you....
Note the difference between "ones you've bought" and "ones provided by the cable Internet vendor".
My experience has been, any software provided by an ISP is to be treated as worse than malware.
Since I never used 16-bit Windows, I never understood "Internet Dialler" software that Windows users seemed to always install from their ISP... and was always the first thing in the way when trying to fix a busted system. But it has served to convince legions that ISP-provided software is necessary to get on the Internet. (Whereas even Windows 95 had enough stuff to dial a modem and set up PPP or SLIP. For non-PPPoE broadband--typically cable--you need no extra software on any Ethernet capable OS.)
More recently, "Internet Security Suites" provided by ISP should be never installed. If found installed, it should be removed by re-formatting the system from read-only media. (Should of serious registry hacking, that's the only way to get rid of the stuff Bell Sympaticrap provides.)
So I'm not surprised they ship modems or routers with "PWN ME!" as the default setting.
Buying your own router immediately puts you in the top percentiles of "tech skill".
It's possible it's even worse for guys, because they AREN'T expected to be embarrassed about how they look when going to the gym, so no-one gives them any advice about what to do about it.
Like; wear baggy clothes; pick a "utilitarian" gym that's just got normal work-a-day people in it, neither the pretty cardio-bunnies nor the hard-core bodybuilders; everyone else is there to get their workout done, too; and so on.
But most importantly: You've got to start somewhere.
On the other hand, I'm a big fan of "everyday" kinds of exercise. If you drive, start parking at the back of the lot. If you take the bus or subway, get off a stop early, then two, then three. That sort of thing. That can be difficult in Suburbia, but it's just that much more important to make the effort. At the big-box mall, don't drive from store to store, for example.
For a little more, you can do a lot at home with some dumbbells and a simple bench. Don't get lured in by fancy chrome rigs. Check out Craigslist and other for-sale sites. Classifieds. Freecycle. There's a lot of people who give up on their exercise plans....
Whatever the choice, the important thing, and the very hardest thing, is to get up off my fat butt and actually do it. I mean your butts.
When a traffic signal in Toronto loses contact with the computer, it goes to flashing 4-way red, to remind people it's an all-way-stop now. They only go to "no signal" if there's a power failure. (In which case, they're still all-way-stop but despite the radio saying so every time there's a power failure, 70% of drivers just blow through a dark traffic light at speed.)
I think all-way-stop on multilane roads are MUCH MUCH MUCH MUCH worse than untimed lights. People have no idea how to deal with multilane roads that have a stop sign; almost no-one will let a left turn through, all sorts of problems like that.
Compare "Major Appliances" to "Appliances". Hand mixer, blender, kettle, coffee maker, bread machine, coffee grinder, countertop boiler, ice cream maker, and so on. Plenty are made without a grounded cord.
Ground isn't "instant safety", though. Depending on circumstances, ground can make a fault worse. That's why the shift to double-insulated power tools with ungrounded plugs. (Say on a drill: the chuck is insulated from the motor, and the motor is insulated from the housing. So if you drill into a live wire, the circuit DOES NOT COMPLETE through the power tool to either ground or you--or both. (If you hit a Big Cable, it will be too much for the 16 gauge ground, so there'll be plenty of current to go through you, too.))
Mind you, I had a paper shredder "fail dangerous" when the double-insulating piece that insulated the cutters from the motor failed... and the motor fell out of its mounts... and turned on... and shorted hot to the control panel. (That unit should not have received ULC and CSA safety approval with a design that brittle. The motor should have been bolted to the case, not "propped" in place by a plastic widget.)
And it's real fun to do that 'dd' on a filesystem which doesn't support sparse files.
It takes a LOT LOT LOT LOT LOT longer: the 'seek' actually extends the file on-disk by the specified amount, so it really does take up the 8.0GB.
I like leaving large sparse files around to trap bad backup software before people put their faith in it.
Well, one thing, iSCSI has to transfer disk blocks, and NFS actually knows about files. So the NFS server can do read-ahead knowing the structure of a directory, or the inode, or the file itself. But iSCSI may not read-ahead in the right pattern, because first you have to read the directory blocks, then the inode blocks, then the file blocks, and they may not be adjacent and sequential.
You may also need to transfer more than you need using iSCSI. "Open this file for me" on NFS doesn't take a very large packet at all, as the client doesn't need to know anything about the server's underlying filesystem.
Right tool for the job and all that....
(This is not to be taken as an endorsement of NFS.)
And, of course, distributions are STILL free to put whatever scheduler they feel like maintaining in Linux. Linus doesn't have to accept it for a distribution to include it.
That's the great thing... and sometimes the worst thing... about Linux. You can bolt in anything you're prepared to code, and it doesn't MATTER if the "vendor" doesn't want it. That just means you have to handle merges and testing on your own when new kernels come out.
For starters, people sweat and computers do not. So, airflow helps cool people by increasing evaporation, in addition to direct thermal transfer. Even when you think you aren't sweating, your skin is still moist and evaporative cooling still works.
Unless someone invents a CPU swamp cooler, that's just not happening on a computer. You do need airflow to keep the hot air from remaining close to the hot component (this can be convection or forced), but you don't get that extra... let's call it "wind chill" effect that humans feel.
His lawsuit and license threats kept our customers from switching over to Red Hat Enterprise Linux systems. Consequently, we have to support a mish-mash of HP-UX, AIX, and Solaris (on two CPUs). Linux is finally catching on, but it's been delayed, and gave Solaris on Intel an edge it did not deserve, which has us running duplicate systems that shouldn't be necessary, and duplicate test load.
I'm also left wondering if OpenSolaris is licensed correctly, given who Sun bought some of the licenses from. That, however, I don't have to worry about professionally, as our customers won't run an OS without a big, juicy support contract.