That's a good question. Probably because people don't think of it that way. There are frameworks for this in place... first OLE, now COM, probably soon.NET(?) on Windows, Bonobo for GNOME (which is built on top of CORBA, a more generalized standard for making "requests" from "objects")... I'm sure other platforms have other standards, I bet at least some are CORBA-based.
Why don't people do this? I dunno. I can't say that it seems like it would be bad or very difficult, as a developer. Provided you had a decent implementation of your object frameworks (CORBA, or COM, or whatever). However, I don't have any direct experience developing for these frameworks, so I can't say much.. Somebody here know COM or GNOME?
I'm not sure I'd be willing to pay for a GHz PDA unless AMD made a huge turnaround from their current reputation. That would be one really hot beast to carry around in your hand. AMD would have to put a lot more into making their chips run cooler... though they've come a long way so far.
I think that's an interesting idea though... and it could probably come to fruition easy enough in about a year. Hell, it probably wouldn't be even *that* difficult to pull off. AMD has the price advantage now, I say they just pull their current Athlon core, downclock it (it'll run cooler than the Thunderbird, I bet) and work from there. You'd have to convince people that it's more than just a "cool toy", though.
Ooh! That must make DEC the somewhat older, yet highly sexy^Wattractive and intelligent lover who dies in a tragic "accident". And AMD can be the distant steprelative that fills in for her!
Consolidation might or might not be interesting... or good. I dunno about Apple and Motorola merging though. That'd be kind of like Digi- oh wait. Collaboration could be fun. Maybe CPUs with onboard SDRAM. Somebody does that now, don't they?
Microsoft became responsible for other people's code the moment they got into the business of signing other people's code.
Good point. I hadn't thought about that at all. Although, it looks like all you need to do to get your driver signed is submit it to Microsoft's "Windows Hardware Quality Labs" for testing, and do some packaging.
Unfortunately, MS' website doesn't seem to say much about what WHQL actually does. Further investigation on that website seems to reveal that the signature only means that a driver is "compatible" with a Windows release, and that not installing unsigned drivers "may prevent problems such as... system instability". So, in the end, signed drivers are compatible with Windows and should be more stable than the unsigned. Oh, and that this copy is unmodified from the copy WHQL got. It seems Microsoft doesn't read too much into what its signing policy means for driver quality beyond "does it install, and will it crash?"...
I was hoping I could find some sort of agreement for the end-user on what driver signing means to them, but not so quickly, at least.
Yes, but so is reading code comments out of context:) All "We have a race condition here" implies is that without proper locking or other handling, bad things might happen. If you actually read the code the comment talks about though, that code seems to deal with it.
Also, the kernel 8139 driver has some obscure note ("Note: the chip doesn't have auto-pad!") and writes (basically) the packet data length to some hardware register... I wonder what that does. (Disclaimer, read the following paragraph; also, I don't know this driver or this hardware. I do use it though:3 )
Interestingly enough, I can't find the "We have a race condition" code doesn't exist in 2.4.18. Weren't they basing this off that kernel? If not, the file they link to off scyld.com in the PDF doesn't exist either... hmmm. That code segment is a bit different in 2.4.18's rtl8139_start_xmit... and lxr tells me that rtl8129_start_xmit doesn't exist.
I think it's unfortunate. Slashdot could be a nice statement about independent journalism. Instead it's a nice statement on the laziness of our societies:)
So here M$ gets look innocent and say, "It's nothing to do with us", while actually being guilty as sin for having an OS that uses loads of vulnerable drivers.
How is MS guilty for crappy code in other people's drivers now? They're not responsible for writing drivers for every single piece of hardware someone wants to run under Microsoft Windows. That would be the responsibility of the hardware vendors.... and like it or not (and there are bits of the Win32 API I really do dislike), not all code that comes from Microsoft is utter crap with no redeeming qualities.
Oh, I really wish I could find a link to the Penny Arcade about "M$" right now too.:) Although I just realized I'm responding to an AC...
In my experience, this works really well for actual video files. I admit I don't have a good solution for DVDs ^^; and still drop to the CLI to watch those, which is a small pain now that I've gotten used to being lazy to play "videos". Maybe if someone starts hacking libnav again and we get good DVD navigation for MPlayer. (If only I had more spare time:) Well who knows, maybe if it's not too difficult... )
I don't run KDE, but to do this for GNOME for known media types is fairly easy, you just have to add to or modify a text file that has MIME type associations in it that the configurator references. One of two files, depending on if you want the changes system-wide (requires root, of course) or just for the current user. I assume KDE is similar 'cause that makes lots of sense to me.
Try "CC=gcc-3.2./configure" instead of "./configure". It looks like the new gcc won't install as the default compiler... maybe? Instead it installs as "gcc-3.2", leaving the 2.96 version as "gcc". That's how Debian 3.0 is set up right now (except they use 2.95.4, bless their hearts), also how I have my LFS set up. Don't have a RH system, so I can't directly test it. (Hehe, late response, but maybe (I hope?) this will be of use to you.)
The article submitters now can't be bothered to read the articles? You know, the bit in the article where it says "The Linux, NetBSD and Microsoft Windows operating systems are known to have vulnerable link layer implementations, and it is extremely likely that other operating systems are also affected."!?
Back to more important issues, like the actual content of the article... I'm reading linux/net right now, anyone have any definite answers, and could point me to (say) source for where we do do this?
Uhhhh, no. MCE does [claim to] require a TV card that does its own MPEG-2 encoding.
<span style="flavor: conspiracy-theory;">Although, it wouldn't surprise me too much if they don't actually use it for the first release, then turn it on, come back, and say "Oh look! This release of MCE is so much faster than the first one. Won't you upgrade... ?"</span>
Seriously though. The article says it uses MPEG-2 onboard the TV-in card. The article, in fact, makes a big deal about it and says that ATI and NVidia are complaining about it. So next time you should probably read the article:) Although yes, doing stuff in software generally sucks if there's a hardware solution.
Okay, somehow I got really stupid and forgot that parent was a review of said XPMCE... sorry for the slashbotting. I usually play much nicer:3
But... having read the article, it says "... if you strip away the Media Center part of the equation you will be left with a fully functional copy of Windows XP Professional. What Media Center adds to the OS is primarily support for the Media Center application, which appears as nothing more than a regular program in your Start Menu."
Great. So for a Media Center PC I'm paying for Microsoft to add another program to my Start Menu? With an "incredible interface" no less... but really. So it knows how to talk to a TV card. Whee. Actually, the program UI looks nice, but if we could get one of these MPEG TV cards with open standards (good luck!) I think an equivalent could be made. Especially one better integrating the music bits.
So. "Media Center" PCs require TV cards that do onboard MPEG-2 encoding. Can't I get those now? Granted, there are, what, two of them? But still. Nothing's stopping ATI from liscensing a hardware MPEG-2 solution from someone. Hell, if they'd do that and let the driver toggle MPEG-2 encoding it'd be great. That should be more a message to ATI and NVidia to stop complaining about MS' move in this case and make more advanced products. Sheesh.
One thing about the article. MPEG-2 isn't "bloated". It's different. (I know, but bear with me a moment... ) MPEG-1 was designed for high-bandwidth, high-quality applications. MPEG-2 was designed for lower-bandwidth. MPEG-4 is for even lower-bandwidth applications. I assume that the lower the bandwidth, the higher the time complexity of the encoder. (Decoders are supposed to be relatively simple.) Lo, this is what speaketh the MPEG Standards. So encoding TV to MPEG-2 seems reasonable. Good to know Anandtech knows MPEG.:P
The article calls the decision to require onboard MPEG-2 encoding in its TV-card "unique".... I think it makes sense for a consumer product. I think it makes MORE sense than limiting who can make Media Center PCs. But hey, I'm not Microsoft. I also was somewhat disappointed that, while they found that the performance even under idle load was "unacceptable", their solution was essentially to say "Microsoft, wait for HyperThreading CPUs" instead of "Microsoft, optimize your code".
Hehehe... for all the times I've seen someone tell other people "don't irc as root" on opn... "Don't watch TV as the Administrator!"
The picture of the screwed-over VGA connector was good for a laugh... "DO NOT REMOVE".:3
Okay, I'll give you that I live under a rock (kernel hacking for one class and dealing with the other classes has been my life for the past few months), but I have a few questions...
Why the "new operating system" tag? This seems no more a new OS version than (say) Plus! was. Even Microsoft calls XPMCE a "new member of the Windows XP family". This isn't even a step from 95 to 98 as I understand it.:)
Second, what's so special about XPMCE? Other than it's designed to run on a Media Center PC. On that topic, does that mean that the only thing special about this version is that it will only run on machines that label themselves Media Center PCs? I ask because Microsoft's FAQ page says...
Q.Can I just buy the Windows XP Media Center Edition system software and install it on my current machine? A. No. Windows XP Media Center Edition is only available on a new kind of PC, called a Media Center PC. Media Center PCs provide the best of home computing and entertainment, and are built by Microsoft's partner PC manufacturers. The operating system software is not available as a stand-alone product, and a custom Media Center PC cannot be built at this time with off-the-shelf products.
So. What entails a Media Center PC, if you can't buy one off with OTS components? Cause I know you sure as hell can get "powerful processors, high-capacity hard disks, CD-ROM/DVD drives, and rich graphics and audio capabilities... a remote control, [and a] TV tuner card" off the shelf.
Let's see here: Redmond giant creates a native client for their media formats to run on a competing operating system.
Ah, I see. Like they did for Mac OS already. Nobody seemed really excited about that, or maybe it's just so long ago I don't remember ^^;
If they release a Linux client, Microsoft will be acknowledging the sizable (and growing) user base of Linux and would show by their actions that ignoring the Linux crowd would be harmful to their business.
Oh? My interpretation was that it would be closer to saying "There, all 'major' platforms support our Great[tm] New[tm] media 'standard', so what's the big deal?" Or maybe something else.
As much as I'd like to believe what you just said, I think it's a bit optimistic to be saying that now. My guess is that Microsoft is feeling pressure from Linux on the server end, but I seriously doubt they feel anything (yet!) on the desktop market. Regardless of whether or not they should be.
I agree. MPlayer is nice. Unfortunately, most of my hardware isn't x86 so I can only use it on one machine. *laughs*
I'm actually somewhat confused about which codecs it supports from the code and which it supports from other operating systems' binary drivers. (Although the list of.dll and.ax files in codecs-status.html does make some blatantly obvious:D ). Although I do know that if you happen to be running Linux on ia32, if you just drop the other system codec files into a known location you get support for pretty much everything.
So set her up a Konqueror file type extension that opens mplayer when you double-click media files. I don't use KDE (don't use GNOME either, but I do use a couple of GNOME apps), but it can't be that difficult if it's easy in Nautilus:D
That's what I did when I decided I was sick of typing in long pathnames for media files.
Ah yes. The infamous gcc 2.96. Recommend 2.95 or 3.2.1. You can also tell MPlayer to disable sanity checks in configure, that will let it build with your version of gcc. I would recommend a different compiler since that one's been known to be wonky, but if you're using 2.96-110 you're a bit more likely to be okay.
The perl configure script is the greatest. "Seems like you're not running Eunice." Every time I build perl, I never use configure.gnu, that would be too boring. Always got to be Configure. ^_^ It's also the best-written plain-language configure script I've seen, in terms of understandability.
Well, the thing is, "elegant", well-written code doesn't look "awesome to the eye". Take a look at the new scheduler code, for example. It's very well written, and easy to read, but because of this it doesn't "look impressive". But that's the way it should be.
Someone should have told him to look at the code in drivers/... for all the bitching Al Viro does about that code, there's got to be some absolutely horrid examples in there that would be perfect for scaring people.
I thought the green guy (H2G2 guy?) and cow in the kernel panic messages were a nice touch. Unfortunately I can't put them in here because Slash complains about "junk characters", even with lots of rambling, semi-off-topic filler.
More people should read lkml to find the gems like this. Amusing things will get aired if you read the endless liscensing debates to the point where people get pissed too. Especially the asbestos suits and flamethrowers bits.
Updated versions of... !? Unless you're running ecgs pre-2.90, you don't really need to update GCC with every release:) egcs-2.91 will still compile the kernel. (Thank akpm for this one.) Otherwise, I'm guessing you've got 2.95? (Unlucky you if it's 2.96 before.96-85.) If you had 3.0, I could see wanting to upgrade to 3.2, since it's an obvious improvement. But really, if you have 2.95, there's no clear need to upgrade. 2.95 still works fine, thanks.
Of course, if you have something before 2.95, you probably should have upgraded it a long time ago anyway -_^
Notation. Our professor always gave rest energy as E0, subscripted. (Forgot the CSS, too lazy to look it up. ^^; ) I like that because it's slightly clearer (to me, a programmer, who thinks everything starts with zero:3 ). And yeah, usually it's gamma.
(I almost replied to grandparent to remark that he sounds like my physics professor might when drunk. It'd be quite easy to imagine that if you knew my physics professor. He'd ofte^Wsometimes talk about random stuff in lecture, like how he and his wife accidentally stayed at this famous gay resort in Greece, and how being in physics made reading Greek road signs easier.)
Slashdot Mirror
That's a good question. Probably because people don't think of it that way. There are frameworks for this in place ... first OLE, now COM, probably soon .NET(?) on Windows, Bonobo for GNOME (which is built on top of CORBA, a more generalized standard for making "requests" from "objects") ... I'm sure other platforms have other standards, I bet at least some are CORBA-based.
Why don't people do this? I dunno. I can't say that it seems like it would be bad or very difficult, as a developer. Provided you had a decent implementation of your object frameworks (CORBA, or COM, or whatever). However, I don't have any direct experience developing for these frameworks, so I can't say much.. Somebody here know COM or GNOME?
I'm not sure I'd be willing to pay for a GHz PDA unless AMD made a huge turnaround from their current reputation. That would be one really hot beast to carry around in your hand. AMD would have to put a lot more into making their chips run cooler ... though they've come a long way so far.
I think that's an interesting idea though ... and it could probably come to fruition easy enough in about a year. Hell, it probably wouldn't be even *that* difficult to pull off. AMD has the price advantage now, I say they just pull their current Athlon core, downclock it (it'll run cooler than the Thunderbird, I bet) and work from there. You'd have to convince people that it's more than just a "cool toy", though.
Semiconductors is a soap opera for nerds.
Ooh! That must make DEC the somewhat older, yet highly sexy^Wattractive and intelligent lover who dies in a tragic "accident". And AMD can be the distant steprelative that fills in for her!
Consolidation might or might not be interesting ... or good. I dunno about Apple and Motorola merging though. That'd be kind of like Digi- oh wait. Collaboration could be fun. Maybe CPUs with onboard SDRAM. Somebody does that now, don't they?
Microsoft became responsible for other people's code the moment they got into the business of signing other people's code.
Good point. I hadn't thought about that at all. Although, it looks like all you need to do to get your driver signed is submit it to Microsoft's "Windows Hardware Quality Labs" for testing, and do some packaging.
Unfortunately, MS' website doesn't seem to say much about what WHQL actually does. Further investigation on that website seems to reveal that the signature only means that a driver is "compatible" with a Windows release, and that not installing unsigned drivers "may prevent problems such as ... system instability". So, in the end, signed drivers are compatible with Windows and should be more stable than the unsigned. Oh, and that this copy is unmodified from the copy WHQL got. It seems Microsoft doesn't read too much into what its signing policy means for driver quality beyond "does it install, and will it crash?" ...
I was hoping I could find some sort of agreement for the end-user on what driver signing means to them, but not so quickly, at least.
("could lead to a race though..." is BAD)
Yes, but so is reading code comments out of context :) All "We have a race condition here" implies is that without proper locking or other handling, bad things might happen. If you actually read the code the comment talks about though, that code seems to deal with it.
Also, the kernel 8139 driver has some obscure note ("Note: the chip doesn't have auto-pad!") and writes (basically) the packet data length to some hardware register ... I wonder what that does. (Disclaimer, read the following paragraph; also, I don't know this driver or this hardware. I do use it though :3 )
Interestingly enough, I can't find the "We have a race condition" code doesn't exist in 2.4.18. Weren't they basing this off that kernel? If not, the file they link to off scyld.com in the PDF doesn't exist either ... hmmm. That code segment is a bit different in 2.4.18's rtl8139_start_xmit ... and lxr tells me that rtl8129_start_xmit doesn't exist.
Admittedly this is worst on the unedited front page. I've seen a few redeeming articles recenty. Don't remember what they are :)
I think it's unfortunate. Slashdot could be a nice statement about independent journalism. Instead it's a nice statement on the laziness of our societies :)
So here M$ gets look innocent and say, "It's nothing to do with us", while actually being guilty as sin for having an OS that uses loads of vulnerable drivers.
How is MS guilty for crappy code in other people's drivers now? They're not responsible for writing drivers for every single piece of hardware someone wants to run under Microsoft Windows. That would be the responsibility of the hardware vendors. ... and like it or not (and there are bits of the Win32 API I really do dislike), not all code that comes from Microsoft is utter crap with no redeeming qualities.
Oh, I really wish I could find a link to the Penny Arcade about "M$" right now too. :) Although I just realized I'm responding to an AC ...
In my experience, this works really well for actual video files. I admit I don't have a good solution for DVDs ^^; and still drop to the CLI to watch those, which is a small pain now that I've gotten used to being lazy to play "videos". Maybe if someone starts hacking libnav again and we get good DVD navigation for MPlayer. (If only I had more spare time :) Well who knows, maybe if it's not too difficult ... )
I don't run KDE, but to do this for GNOME for known media types is fairly easy, you just have to add to or modify a text file that has MIME type associations in it that the configurator references. One of two files, depending on if you want the changes system-wide (requires root, of course) or just for the current user. I assume KDE is similar 'cause that makes lots of sense to me.
Try "CC=gcc-3.2 ./configure" instead of "./configure". It looks like the new gcc won't install as the default compiler ... maybe? Instead it installs as "gcc-3.2", leaving the 2.96 version as "gcc". That's how Debian 3.0 is set up right now (except they use 2.95.4, bless their hearts), also how I have my LFS set up. Don't have a RH system, so I can't directly test it. (Hehe, late response, but maybe (I hope?) this will be of use to you.)
By the way, you misspelled "blatant", and the use of HTML tags obviates the need of ugly all-caps messages :)
Sorry, I usually don't feed the trolls ... it's 730am, I've been up for almost 24 hours, and I'm waiting for a drive to badblock ^^;
The article submitters now can't be bothered to read the articles? You know, the bit in the article where it says "The Linux, NetBSD and Microsoft Windows operating systems are known to have vulnerable link layer implementations, and it is extremely likely that other operating systems are also affected."!?
Back to more important issues, like the actual content of the article ... I'm reading linux/net right now, anyone have any definite answers, and could point me to (say) source for where we do do this?
Uhhhh, no. MCE does [claim to] require a TV card that does its own MPEG-2 encoding.
<span style="flavor: conspiracy-theory;">Although, it wouldn't surprise me too much if they don't actually use it for the first release, then turn it on, come back, and say "Oh look! This release of MCE is so much faster than the first one. Won't you upgrade ... ?"</span>
Seriously though. The article says it uses MPEG-2 onboard the TV-in card. The article, in fact, makes a big deal about it and says that ATI and NVidia are complaining about it. So next time you should probably read the article :) Although yes, doing stuff in software generally sucks if there's a hardware solution.
Okay, somehow I got really stupid and forgot that parent was a review of said XPMCE ... sorry for the slashbotting. I usually play much nicer :3
But ... having read the article, it says " ... if you strip away the Media Center part of the equation you will be left with a fully functional copy of Windows XP Professional. What Media Center adds to the OS is primarily support for the Media Center application, which appears as nothing more than a regular program in your Start Menu."
Great. So for a Media Center PC I'm paying for Microsoft to add another program to my Start Menu? With an "incredible interface" no less ... but really. So it knows how to talk to a TV card. Whee. Actually, the program UI looks nice, but if we could get one of these MPEG TV cards with open standards (good luck!) I think an equivalent could be made. Especially one better integrating the music bits.
So. "Media Center" PCs require TV cards that do onboard MPEG-2 encoding. Can't I get those now? Granted, there are, what, two of them? But still. Nothing's stopping ATI from liscensing a hardware MPEG-2 solution from someone. Hell, if they'd do that and let the driver toggle MPEG-2 encoding it'd be great. That should be more a message to ATI and NVidia to stop complaining about MS' move in this case and make more advanced products. Sheesh.
One thing about the article. MPEG-2 isn't "bloated". It's different. (I know, but bear with me a moment ... ) MPEG-1 was designed for high-bandwidth, high-quality applications. MPEG-2 was designed for lower-bandwidth. MPEG-4 is for even lower-bandwidth applications. I assume that the lower the bandwidth, the higher the time complexity of the encoder. (Decoders are supposed to be relatively simple.) Lo, this is what speaketh the MPEG Standards. So encoding TV to MPEG-2 seems reasonable. Good to know Anandtech knows MPEG. :P
The article calls the decision to require onboard MPEG-2 encoding in its TV-card "unique". ... I think it makes sense for a consumer product. I think it makes MORE sense than limiting who can make Media Center PCs. But hey, I'm not Microsoft. I also was somewhat disappointed that, while they found that the performance even under idle load was "unacceptable", their solution was essentially to say "Microsoft, wait for HyperThreading CPUs" instead of "Microsoft, optimize your code".
Hehehe ... for all the times I've seen someone tell other people "don't irc as root" on opn ... "Don't watch TV as the Administrator!"
The picture of the screwed-over VGA connector was good for a laugh ... "DO NOT REMOVE". :3
Okay, I'll give you that I live under a rock (kernel hacking for one class and dealing with the other classes has been my life for the past few months), but I have a few questions ...
Why the "new operating system" tag? This seems no more a new OS version than (say) Plus! was. Even Microsoft calls XPMCE a "new member of the Windows XP family". This isn't even a step from 95 to 98 as I understand it. :)
Second, what's so special about XPMCE? Other than it's designed to run on a Media Center PC. On that topic, does that mean that the only thing special about this version is that it will only run on machines that label themselves Media Center PCs? I ask because Microsoft's FAQ page says ...
Q.Can I just buy the Windows XP Media Center Edition system software and install it on my current machine?
A. No. Windows XP Media Center Edition is only available on a new kind of PC, called a Media Center PC. Media Center PCs provide the best of home computing and entertainment, and are built by Microsoft's partner PC manufacturers. The operating system software is not available as a stand-alone product, and a custom Media Center PC cannot be built at this time with off-the-shelf products.
So. What entails a Media Center PC, if you can't buy one off with OTS components? Cause I know you sure as hell can get "powerful processors, high-capacity hard disks, CD-ROM/DVD drives, and rich graphics and audio capabilities ... a remote control, [and a] TV tuner card" off the shelf.
Something smells fishy.
Let's see here: Redmond giant creates a native client for their media formats to run on a competing operating system.
Ah, I see. Like they did for Mac OS already. Nobody seemed really excited about that, or maybe it's just so long ago I don't remember ^^;
If they release a Linux client, Microsoft will be acknowledging the sizable (and growing) user base of Linux and would show by their actions that ignoring the Linux crowd would be harmful to their business.
Oh? My interpretation was that it would be closer to saying "There, all 'major' platforms support our Great[tm] New[tm] media 'standard', so what's the big deal?" Or maybe something else.
As much as I'd like to believe what you just said, I think it's a bit optimistic to be saying that now. My guess is that Microsoft is feeling pressure from Linux on the server end, but I seriously doubt they feel anything (yet!) on the desktop market. Regardless of whether or not they should be.
I agree. MPlayer is nice. Unfortunately, most of my hardware isn't x86 so I can only use it on one machine. *laughs*
I'm actually somewhat confused about which codecs it supports from the code and which it supports from other operating systems' binary drivers. (Although the list of .dll and .ax files in codecs-status.html does make some blatantly obvious :D ). Although I do know that if you happen to be running Linux on ia32, if you just drop the other system codec files into a known location you get support for pretty much everything.
So set her up a Konqueror file type extension that opens mplayer when you double-click media files. I don't use KDE (don't use GNOME either, but I do use a couple of GNOME apps), but it can't be that difficult if it's easy in Nautilus :D
That's what I did when I decided I was sick of typing in long pathnames for media files.
Ah yes. The infamous gcc 2.96. Recommend 2.95 or 3.2.1. You can also tell MPlayer to disable sanity checks in configure, that will let it build with your version of gcc. I would recommend a different compiler since that one's been known to be wonky, but if you're using 2.96-110 you're a bit more likely to be okay.
Nah, I like the cow better. Slash won't let me post it, so here's a link.
The perl configure script is the greatest. "Seems like you're not running Eunice." Every time I build perl, I never use configure.gnu, that would be too boring. Always got to be Configure. ^_^ It's also the best-written plain-language configure script I've seen, in terms of understandability.
Well, the thing is, "elegant", well-written code doesn't look "awesome to the eye". Take a look at the new scheduler code, for example. It's very well written, and easy to read, but because of this it doesn't "look impressive". But that's the way it should be. Someone should have told him to look at the code in drivers/ ... for all the bitching Al Viro does about that code, there's got to be some absolutely horrid examples in there that would be perfect for scaring people.
I thought the green guy (H2G2 guy?) and cow in the kernel panic messages were a nice touch. Unfortunately I can't put them in here because Slash complains about "junk characters", even with lots of rambling, semi-off-topic filler.
More people should read lkml to find the gems like this. Amusing things will get aired if you read the endless liscensing debates to the point where people get pissed too. Especially the asbestos suits and flamethrowers bits.
Updated versions of ... !? Unless you're running ecgs pre-2.90, you don't really need to update GCC with every release :) egcs-2.91 will still compile the kernel. (Thank akpm for this one.) Otherwise, I'm guessing you've got 2.95? (Unlucky you if it's 2.96 before .96-85.) If you had 3.0, I could see wanting to upgrade to 3.2, since it's an obvious improvement. But really, if you have 2.95, there's no clear need to upgrade. 2.95 still works fine, thanks.
Of course, if you have something before 2.95, you probably should have upgraded it a long time ago anyway -_^
Notation. Our professor always gave rest energy as E0, subscripted. (Forgot the CSS, too lazy to look it up. ^^; ) I like that because it's slightly clearer (to me, a programmer, who thinks everything starts with zero :3 ). And yeah, usually it's gamma.
(I almost replied to grandparent to remark that he sounds like my physics professor might when drunk. It'd be quite easy to imagine that if you knew my physics professor. He'd ofte^Wsometimes talk about random stuff in lecture, like how he and his wife accidentally stayed at this famous gay resort in Greece, and how being in physics made reading Greek road signs easier.)