I looked a lot of Intel and AMD motherboards over the last year (my ancient AthlonMP might be getting a bit obsolete and all) and all of them had pins on the board for serial ports. Of course, you had to get a cable and wire it up to a backplate... but they're there if you really want 'em. You're SOL on laptops AFAICT though.
Amusingly enough, I discovered that most of them ran the serial and floppy ports from the temperature sensor mcu:-X
And... since I decided to keep that AthlonMP alive with a SATA card (oh yeah PCI baby) I then discovered that KMS, AGP, and SMP are not friends, and now know the joys of using KGDB over a serial console during boot to figure out why in the Hell the machine dies during card to host transfers or sometimes during modesetting... but only when both processors are up. LET ME TELL YOU ABOUT THE FUN I'VE HAD. Not that I've gotten far since whatever problem it is causes even the hardware NMI watchdog to go out for an extended lunch.
Unfortunately, the Endura Pro pointing stick is more of a joystick... it might be a Trackpoint I or some clone... I had one for a long time, and it was useful for throwing the pointer at windows (reasonable under focus follows mouse), but I still had to keep a trackball around for fine operation (boo). Also, the buttons are cheap membrame switches, and my left button wore out after maybe 18 months. That wasn't too bad (I eventually had to remove the button plate and tap the membrane by hand) until it wore out and the left button was stuck on... ended up having to cut the trace at that point and not have pointer control... (and I emailed them, alas, you can't replace that part due to the way it's manufactured).
If only they would build a model with a Trackpoint IV. Come on Unicomp, license it already... until then, boo, living without a trackpoint it shall be. Life could be worse;)
The main reason IPC is slow on GNU/Linux is because TLB flushes are expensive. Modern processors have nice tagged TLBs (designed for virtualization, but nothing stops a kernel from using it as a PID tag) which could eliminate most of the overhead.
That, and Linux suffers the same problem Mach has: it's too damn big to fit into cache so the essential parts of the kernel for IPC/task switching in general don't stay in fast memory.
Wayland works around the "task switching is expensive" problem by papering over the real issues and just moving stuff into the monolithic kernel (terrible idea... the display server need not do modesetting, but it should be a user space task if you ask me) and making the compositor/window-manager/display-server a single task. You gain a performance hack, but you lose proper process interface separation and other nice things. I suspect you could fix a lot of the latency issues in X11 with a synchronous IPC protocol change for window managers and compositors (ala jackd... you could then avoid copying and just use shared memory segments to shove pixel data between the few processes).
But, ey, I guess it'll work. I'm still not convinced Wayland will ever take over: the X11 developers seem to be working on a lot of stuff to negate the need for Wayland (e.g. the multi-pointer input model fixes the broken focus model, glamour replaces DDX with Mesa, a new xkb standard is on the way, etc.).
Hell, I have a dual AthlonMP rig (got it when I graduated high school... about ten years ago) that I just popped a sata card into and a pair of new disks to breathe a few more months life into it as a media and backup server. XBMC's a bit slow (but that's the Radeon 9100's fault, heh) but otherwise... and I used that as my daily workstation (driving a nice 1920x1200 monitor and doing some heavy SBCL hacking) until a couple of years ago. Basically Firefox needing a gig of ram to display "Hello World" is what drove me into using my laptop more...
The limiting factor for using perfectly adequate old hardware now are those pesky acronyms: PATA, PCI, AGP, DDR{266,333}...
RMS has always held that it is acceptable to use non-free software for the purposes of reverse engineering or replacing it. Or if there are no Free alternatives whatsoever (but, if you are capable, you should write that software...).
Probably not, actually. 'Tho my personal view is that both "parties" may as well be one and are roughly equally evil ("how would you like to be screwed over today, sir?")... I listened to the vote as it happened, and it was pretty clearly a "oh no... I hit the wrong button" moment. Then the majority used some parlimentary trickery to prevent a reconsideration vote (which I believe they could have done, since a supposedly anti-fracking person voted for it and everyone can vote to reconsider...), and then, despite a clearly ambiguous voice vote, they closed the session and re-opened a new one at 12:05 a.m. instead of waiting until morning... just to affirm the previous day's actions into law and half-heartedly debate one bill before giving up 20 minutes later.
Also, there was another Democrat who definitely traded her vote... something about tax breaks for the film industry in eastern Carolina made her change her tune from anti-fracking to pro-fracking that night. A terrible combination of unfortunate circumstances I say.
The worst part is that NC has so little natural gas that it seems really pointless. That, and they're going to be doing it under a freaking Nuclear power plant (slated to be expanded to 3 units soon, but with Duke at the helm now... save us all). I look forward to the day when a mild seismic event occurs and triggers a week long "oops we just lost 3GW of baseload to an automatic SCRAM" event.
Double Jeopardy doesn't come into play until the first witness in a trial has been sworn in. Up until that point, charges can be dropped, amended, wholesale changed, etc. It seems unfair, but think about e.g. someone being charged with manslaughter ("I was durnk occifer and that bike came out of nowhere") and the prosecution learning, before trial, that there was pre-meditation involved ("Liloaid drove me over the edge and I HAD TO DO IT"). Or the reverse.
Where it gets dirty is that the district attorney's office can continue your trial in some magical pre-trial state indefinitely, despite judge's orders that the trial go foward next time I-really-mean-it-this-time (unless the daily court session runs out of time before hearing your case, unless your lawyer has another court appearance before the prosecution feels like hearing it that day, unless they amend a couple of words in your charges, unless the fourth moon of Pluto is aligned with the Nemesis that day,...).
Unfortunately, neither video nor audio nor transcripts nor slides of the presentation have appeared yet. They aren't likely to for at least another week... there might even be a quick story on it when most of the talks are online.
Re:It's SENSATIONAL! But also kind of BORING!
on
The 300 km/h Superbus
·
· Score: 3, Insightful
If you build the rail, people will live there. The only reason people live by highways is because... the highway was put there to attract residential development.
Raleigh is currently suffering major growing pains, and unfortunately the state has chosen to build a highway through farmlands to house everyone... yeah, I definitely want to live 40 miles from RTP in the middle of nowhere and commute with $5/gal gas looming on the horizon... no, building a light rail system and encouraging people to live in the Raleigh/Durham corridor is a waste of money I tell you.
Where there is rail, people will use it. See NYC (commuting into NYC from Long Island is a breeze on the train, effectively impossible by car, and that's a good thing), Boston, DC, etc. DC to a lesser extent because the Metro hasn't expanded into areas where people live, but MARC is tolerable if you only need to go into the city in the morning and back out in the evening.
The whole "we're too spread out" argument is perfectly valid in the midwest though... and luckily most of our population exists hugging the coast lines. So... highways for the midwesterners, rail for the rest of us! Unfortunately, U.S. central policy is obsessed with "one solution for every part of our geographically diverse land" for whatever reason.
The copyright assignment contributors sign to the FSF might be violated if the FSF did that to GRUB2. It would also be counter to their goals of spreading software freedom. I mean, if the FSF weren't fighting against DRM why even bother updating the GPLv2?
Source: I had to sign one of those agreements a couple of years ago.
The Foundation promises that all distribution of the Work, or of any
work "based on the Work", that takes place under the control of the
Foundation or its assignees, shall be on terms that explicitly and
perpetually permit anyone possessing a copy of the work to which the terms
apply, and possessing accurate notice of these terms, to redistribute
copies of the work to anyone on the same terms. These terms shall not
restrict which members of the public copies may be distributed to. These
terms shall not require a member of the public to pay any royalty to the
Foundation or to anyone else for any permitted use of the work they apply
to, or to communicate with the Foundation or its agents in any way either
when redistribution is performed or on any other occasion.
Actually, Amtrak has priority over freight travel. The FRA was pretty lax in enforcement, but over the last 3 years things have improved substantially. Private rail companies / many states were getting away with murder for a long time there though... (e.g. building sidings that couldn't fit most freight trains... I used to loathe travelling through Virginia).
I really don't get why we evaluate energy sources based on their "EROI"... they generate power. And we're certainly running out of oil / will destroy the environment if we continue burning coal and gas. I.e. the assumptions behind the "EROI" are utterly flawed and give a useless number that has no reflection on reality (I mean, you build a power plant and you're kind of stuck with for 80 or so years... the notion that natural gas will be available for baseload generation in 80 years is laughable).
Nuclear waste would be less of a problem if everyone weren't so uppity about proliferation risks... no reason to bury almost all of the energy in the fuel because some terrrirst might somehow get a small bit of waste and somehow build an advanced centrifuge system to extra the Pu. In his mom's basement of course using duct tape and paperclips (communist paperclips).
This from someone who lives about 15 miles from a nuclear reactor (they might build two AP1000s there too, providing the entire area's current power use from Nuclear... captain planet is pleased). I'm glad it's a nuclear reactor and not a coal plant (air quality is already crappy enough during the summer).
I think you're underestimating how pleasant it is to kick back on a deck chair and churn out some code for a few hours. I have one of those old core2 whatever low voltage laptops with a 13" screen (eh, every 15" had the same resolution... and I made the mistake of buying a 17" monstrosity before), and it really is nice not being tethered to a desk (for 6+ hours of heavy use even).
Of course, in small screen editing mode I'll have a full screen Emacs frame (~270x96 characters isn't too shabby and can house 3 or 4 windows comfortably) with a tiny font and spread things out onto more desktops so it's only a bit less convenient.
But then I have a 24" monitor, clickety keyboard, and trackball sitting on my desk.
Not sure what's up, but checking your user comment page I see: this which appears to be the comment in question. We're still in the "never delete any comments" business at least.
Sort of, DRI is now DRI2, and still used... but the server itself is doing the direct rendering, and clients are doing GLX through the server. Which is much saner since you consolidate buffer management and only the X server is capable of hosing the entire system.
I find it amusing that the desire for fancy desktop effects got us working AIGLX... you'd have thought folks doing CAD or something would. Oh well, I guess the compiz loving people did something good at least:)
Indirect GLX rendering has been accelerated for something like... five years now (Google for AIGLX). There shouldn't be too much overhead versus running locally -- bandwidth, obviously, is going to be a limiting factor for texture heavy applications, but once the commands etc. hit the X server everything goes through the same path as local rendering.
It used to be that OpenGL apps used DRI to directly send commands to the graphics card. Those were the bad old days when video games could crash the entire display system but hosing the graphics card... and you couldn't really run more than one DRI client at a time. It was a mess.
The flag feature just puts the comment into a report; editors can then choose to ignore the report, or magically downmod to -1. The comment can still be moderated back up if other folks decide to do so. In theory, it's a great way to avoid burning mod points on trolls and to instead use them for modding up insightful commentary.
Hope that clears things up!
Feel free to tell me to choke on a whiffle bat, however;)
Part of that has been a declining submission rate -- since the goal of Slashdot is to post stuff readers want to read and give everyone some time on the frontpage so we rely on your submissions.
I've started following various Free Software project and other tech news sites to try and make up for this a bit (as the resident programmer and GNU/Hippie); suggestions on sites to follow for relevant news are welcome, but more submissions are even more welcome!
Slashdot Mirror
I looked a lot of Intel and AMD motherboards over the last year (my ancient AthlonMP might be getting a bit obsolete and all) and all of them had pins on the board for serial ports. Of course, you had to get a cable and wire it up to a backplate... but they're there if you really want 'em. You're SOL on laptops AFAICT though.
Amusingly enough, I discovered that most of them ran the serial and floppy ports from the temperature sensor mcu :-X
And... since I decided to keep that AthlonMP alive with a SATA card (oh yeah PCI baby) I then discovered that KMS, AGP, and SMP are not friends, and now know the joys of using KGDB over a serial console during boot to figure out why in the Hell the machine dies during card to host transfers or sometimes during modesetting ... but only when both processors are up. LET ME TELL YOU ABOUT THE FUN I'VE HAD. Not that I've gotten far since whatever problem it is causes even the hardware NMI watchdog to go out for an extended lunch.
REAL AMMMUUUUERICANS CAN'T DRINK THAT FANCY HIPPIE BEER.
(I'm an avid homebrewer and only drink craft beer ... life's too short for crappy beer I say! And if I want cheap, I can make cheap and good myself.)
Budweiser is owned by a filthy Belgian company now! Have you drink that ... crap, Miller's South African and Coors is Canadian.
IT'S A CONSPIRACY OF THE LIBERAL MEDIA.
Unfortunately, the Endura Pro pointing stick is more of a joystick... it might be a Trackpoint I or some clone... I had one for a long time, and it was useful for throwing the pointer at windows (reasonable under focus follows mouse), but I still had to keep a trackball around for fine operation (boo). Also, the buttons are cheap membrame switches, and my left button wore out after maybe 18 months. That wasn't too bad (I eventually had to remove the button plate and tap the membrane by hand) until it wore out and the left button was stuck on... ended up having to cut the trace at that point and not have pointer control... (and I emailed them, alas, you can't replace that part due to the way it's manufactured).
If only they would build a model with a Trackpoint IV. Come on Unicomp, license it already... until then, boo, living without a trackpoint it shall be. Life could be worse ;)
The main reason IPC is slow on GNU/Linux is because TLB flushes are expensive. Modern processors have nice tagged TLBs (designed for virtualization, but nothing stops a kernel from using it as a PID tag) which could eliminate most of the overhead.
That, and Linux suffers the same problem Mach has: it's too damn big to fit into cache so the essential parts of the kernel for IPC/task switching in general don't stay in fast memory.
Wayland works around the "task switching is expensive" problem by papering over the real issues and just moving stuff into the monolithic kernel (terrible idea... the display server need not do modesetting, but it should be a user space task if you ask me) and making the compositor/window-manager/display-server a single task. You gain a performance hack, but you lose proper process interface separation and other nice things. I suspect you could fix a lot of the latency issues in X11 with a synchronous IPC protocol change for window managers and compositors (ala jackd... you could then avoid copying and just use shared memory segments to shove pixel data between the few processes).
But, ey, I guess it'll work. I'm still not convinced Wayland will ever take over: the X11 developers seem to be working on a lot of stuff to negate the need for Wayland (e.g. the multi-pointer input model fixes the broken focus model, glamour replaces DDX with Mesa, a new xkb standard is on the way, etc.).
Copyright only covers distribution.
Hell, I have a dual AthlonMP rig (got it when I graduated high school... about ten years ago) that I just popped a sata card into and a pair of new disks to breathe a few more months life into it as a media and backup server. XBMC's a bit slow (but that's the Radeon 9100's fault, heh) but otherwise... and I used that as my daily workstation (driving a nice 1920x1200 monitor and doing some heavy SBCL hacking) until a couple of years ago. Basically Firefox needing a gig of ram to display "Hello World" is what drove me into using my laptop more...
The limiting factor for using perfectly adequate old hardware now are those pesky acronyms: PATA, PCI, AGP, DDR{266,333} ...
RMS has always held that it is acceptable to use non-free software for the purposes of reverse engineering or replacing it. Or if there are no Free alternatives whatsoever (but, if you are capable, you should write that software...).
Probably not, actually. 'Tho my personal view is that both "parties" may as well be one and are roughly equally evil ("how would you like to be screwed over today, sir?") ... I listened to the vote as it happened, and it was pretty clearly a "oh no... I hit the wrong button" moment. Then the majority used some parlimentary trickery to prevent a reconsideration vote (which I believe they could have done, since a supposedly anti-fracking person voted for it and everyone can vote to reconsider...), and then, despite a clearly ambiguous voice vote, they closed the session and re-opened a new one at 12:05 a.m. instead of waiting until morning... just to affirm the previous day's actions into law and half-heartedly debate one bill before giving up 20 minutes later.
Also, there was another Democrat who definitely traded her vote... something about tax breaks for the film industry in eastern Carolina made her change her tune from anti-fracking to pro-fracking that night. A terrible combination of unfortunate circumstances I say.
The worst part is that NC has so little natural gas that it seems really pointless. That, and they're going to be doing it under a freaking Nuclear power plant (slated to be expanded to 3 units soon, but with Duke at the helm now ... save us all). I look forward to the day when a mild seismic event occurs and triggers a week long "oops we just lost 3GW of baseload to an automatic SCRAM" event.
Shouldn't, probably. Do? Yes.
Double Jeopardy doesn't come into play until the first witness in a trial has been sworn in. Up until that point, charges can be dropped, amended, wholesale changed, etc. It seems unfair, but think about e.g. someone being charged with manslaughter ("I was durnk occifer and that bike came out of nowhere") and the prosecution learning, before trial, that there was pre-meditation involved ("Liloaid drove me over the edge and I HAD TO DO IT"). Or the reverse.
Where it gets dirty is that the district attorney's office can continue your trial in some magical pre-trial state indefinitely, despite judge's orders that the trial go foward next time I-really-mean-it-this-time (unless the daily court session runs out of time before hearing your case, unless your lawyer has another court appearance before the prosecution feels like hearing it that day, unless they amend a couple of words in your charges, unless the fourth moon of Pluto is aligned with the Nemesis that day, ...).
Unfortunately, neither video nor audio nor transcripts nor slides of the presentation have appeared yet. They aren't likely to for at least another week... there might even be a quick story on it when most of the talks are online.
Blame Canada.
Yo dawg I heard you liked slashdot so I ...
If you build the rail, people will live there. The only reason people live by highways is because ... the highway was put there to attract residential development.
Raleigh is currently suffering major growing pains, and unfortunately the state has chosen to build a highway through farmlands to house everyone... yeah, I definitely want to live 40 miles from RTP in the middle of nowhere and commute with $5/gal gas looming on the horizon... no, building a light rail system and encouraging people to live in the Raleigh/Durham corridor is a waste of money I tell you.
Where there is rail, people will use it. See NYC (commuting into NYC from Long Island is a breeze on the train, effectively impossible by car, and that's a good thing), Boston, DC, etc. DC to a lesser extent because the Metro hasn't expanded into areas where people live, but MARC is tolerable if you only need to go into the city in the morning and back out in the evening.
The whole "we're too spread out" argument is perfectly valid in the midwest though... and luckily most of our population exists hugging the coast lines. So... highways for the midwesterners, rail for the rest of us! Unfortunately, U.S. central policy is obsessed with "one solution for every part of our geographically diverse land" for whatever reason.
The copyright assignment contributors sign to the FSF might be violated if the FSF did that to GRUB2. It would also be counter to their goals of spreading software freedom. I mean, if the FSF weren't fighting against DRM why even bother updating the GPLv2?
Source: I had to sign one of those agreements a couple of years ago.
This is indeed true, although the moderator FAQ makes no mention of it anymore. An oversight that shall be rectified.
Actually, Amtrak has priority over freight travel. The FRA was pretty lax in enforcement, but over the last 3 years things have improved substantially. Private rail companies / many states were getting away with murder for a long time there though... (e.g. building sidings that couldn't fit most freight trains ... I used to loathe travelling through Virginia).
I really don't get why we evaluate energy sources based on their "EROI" ... they generate power. And we're certainly running out of oil / will destroy the environment if we continue burning coal and gas. I.e. the assumptions behind the "EROI" are utterly flawed and give a useless number that has no reflection on reality (I mean, you build a power plant and you're kind of stuck with for 80 or so years... the notion that natural gas will be available for baseload generation in 80 years is laughable).
Nuclear waste would be less of a problem if everyone weren't so uppity about proliferation risks... no reason to bury almost all of the energy in the fuel because some terrrirst might somehow get a small bit of waste and somehow build an advanced centrifuge system to extra the Pu. In his mom's basement of course using duct tape and paperclips (communist paperclips).
This from someone who lives about 15 miles from a nuclear reactor (they might build two AP1000s there too, providing the entire area's current power use from Nuclear... captain planet is pleased). I'm glad it's a nuclear reactor and not a coal plant (air quality is already crappy enough during the summer).
Behold, the horror.
I think you're underestimating how pleasant it is to kick back on a deck chair and churn out some code for a few hours. I have one of those old core2 whatever low voltage laptops with a 13" screen (eh, every 15" had the same resolution... and I made the mistake of buying a 17" monstrosity before), and it really is nice not being tethered to a desk (for 6+ hours of heavy use even).
Of course, in small screen editing mode I'll have a full screen Emacs frame (~270x96 characters isn't too shabby and can house 3 or 4 windows comfortably) with a tiny font and spread things out onto more desktops so it's only a bit less convenient.
But then I have a 24" monitor, clickety keyboard, and trackball sitting on my desk.
I knew that, but my brain autocorrected it when typing and didn't notice that it was creating a typo. Nothing to see here, move along.
Not sure what's up, but checking your user comment page I see: this which appears to be the comment in question. We're still in the "never delete any comments" business at least.
Sort of, DRI is now DRI2, and still used... but the server itself is doing the direct rendering, and clients are doing GLX through the server. Which is much saner since you consolidate buffer management and only the X server is capable of hosing the entire system.
I find it amusing that the desire for fancy desktop effects got us working AIGLX ... you'd have thought folks doing CAD or something would. Oh well, I guess the compiz loving people did something good at least :)
Indirect GLX rendering has been accelerated for something like ... five years now (Google for AIGLX). There shouldn't be too much overhead versus running locally -- bandwidth, obviously, is going to be a limiting factor for texture heavy applications, but once the commands etc. hit the X server everything goes through the same path as local rendering.
It used to be that OpenGL apps used DRI to directly send commands to the graphics card. Those were the bad old days when video games could crash the entire display system but hosing the graphics card... and you couldn't really run more than one DRI client at a time. It was a mess.
The flag feature just puts the comment into a report; editors can then choose to ignore the report, or magically downmod to -1. The comment can still be moderated back up if other folks decide to do so. In theory, it's a great way to avoid burning mod points on trolls and to instead use them for modding up insightful commentary.
Hope that clears things up!
Feel free to tell me to choke on a whiffle bat, however ;)
Part of that has been a declining submission rate -- since the goal of Slashdot is to post stuff readers want to read and give everyone some time on the frontpage so we rely on your submissions.
I've started following various Free Software project and other tech news sites to try and make up for this a bit (as the resident programmer and GNU/Hippie); suggestions on sites to follow for relevant news are welcome, but more submissions are even more welcome!