Do you even know what I meant by that, oh glorious moderator? It is most certainly ON TOPIC as it pertains to how the game Final Fantasy VII DEEPLY AFFECTED ME.
Christ almighty, where's my account with metamod...
That thing is really wicked. It's halfway there (I like how it charges in the cradle... probably a better idea than alkaline batteries). If you could figure out how to lose the pad, and make it wireless... then a firmware update... boom!
If it's too hard to make it truly freehand, it might be acceptable to use a set of magic buttons, "fairy clips", I don't know what you'd call it, maybe it looks like a ruler, but the purpose is to establish an origin, a baseline on the "paper" with which to compare it's position using sonar, RF, whatever. You would like to think that you could do a semi-decent job using dead-reckoning, and maybe a little user guided post-processing to get your scribbles untangled. But maybe that's asking a little much. It's a shame to have to carry around the pad.
It's just the jabber server. Presumably, you would hook it up to a PDA or laptop (like a whip antenna or stylus) and use interactively it that way. The reason for it being a jabber server is so that it can queue up XML messages people send you, and send out pre-programmed responses. Like business cards, presentation notes in MathML, etc. Not so much for the interactive features.
I don't think there would be a need for a seperate "personal server". The device would respond to requests from files, and any server-like features are just response to complex requests. You could organize it around a whole XML-RPC framework with specific types of requests all of which are heavily geared towards processing on the client side at presentation time. You provide the data in a standardized format (or use some fancy XML markup) and keep it in there.
Also, I don't think that using microdrives is a good idea because solid state is better suited for battery operation. Plus you don't need expensive drive electronics to deal with user handling. You can translate those savings into larger storage capacities.
If you were worried about making huge volumes of data available, I would be carrying around a USB-enabled notebook drive, not a repurposed iPod.
But it's not a pen. It contains 256 MB (or so) of flash, which is shared via bluetooth. A 10cm high gain antenna is hidden within the length of the pen itself, and powered by a single AAA battery. Walk by an enabled PC, optionally type in a password, and all your documents, your keyring, etc. are available. Finally, as an added bonus, when you write on paper (or anything for that matter), you can choose to record your scribbles on the flash drive. Tiny gyroscopic sensors determine the motion of the pen across the page, and a pressure sensor determines whether the pen is against a writing surface. Each time you expose the ball point head it creates a new file, and when you retract it, it closes it. You can tell which file is which by the timestamps.
THAT would kick ass. And as embedded logic gets more powerful, you could have a personal web/email/jabber server running in there too. A wireless iPod sounds nifty, but where's the innovation people?
It's the year 2003, I'm writing software...
on
SCO DOS'ed
·
· Score: 1
and at work, we don't have religious wars about operating systems. We think computers are cool. We like our jobs. We use a whole range of stuff. Sure we bitch about Sun's high prices and how we have to buy our personal pizza boxes off ebay, but that's to be expected.
I think the problem is slashdot itself, it's a place where people eventually butt heads. There's nothing to do here but talk, share, debate, argue. That's not bad, but don't think that slashdot is an analogue of what people do, think, or say all day.
Right after I posted that comment I tried to explain to someone the virtues of virtual memory -> physical memory mapping; including the whole "don't allocate 'till you see a page fault on it" thing. He agreed it was quite clever.
If you use linux with BIG_MEM (3GB/1GB split) enabled in the kernel, your.text segment is allocated starting at 0x00010000. I forget why but there was some simplicity to using the 1GB/1GB split otherwise on linux (.text starting at 0x80000000, kernel mem at 0xc0000000). Windows uses a default 2GB/2GB split, so it makes sense it would start at 0x0. I don't think they were thinking about ASCII armor when they made that decision in the NT 3.x days.
On Linux and Windows, the.text segment of the currently rybbubg program starts one page AFTER 0. the page containing the address 0 is marked no-access so that NULL pointer deferences PURPOSEFULLY crash the program. They waste 4k of memory for that feature.
Otherwise, the program would just keep going. YOu wouldn't see the crash until you attempt to write there and clobber your code.
For about 3 and a half grand (which isn't too steep, bout the same for a nice Xeon system). The chips are the most expensive part (about $750 each). Motherboard, $300. RAM, case, monitor, HDs, the rest are whatever you get them for.
Personally, we spent $10,000 on one of these, put it had dual 20" LCDs on a Quadra4 GLX, and 3 each of WD 10k RPM and Seagate Barracuda V SATA HDs. And a Midiman 1010LT for sound, because we're snobs. w000000t. Can't wait to try this tech preview on it.
Did you misread the part where I said LEDs are the winner in flashlight choice? This is because they beat incadescent lights in lumens per watt numbers (the numbers get worse for incadescents too as they get smaller, like in flashlight bulbs). They don't beat flourescent lights, though. That was my point.
Light bulbs emit infrared, radio, microwave, and ultraviolet light (as they are blackbody sources). The heat from the filament is not easily conducted away into the base of the bulb or through the gas inside. Hence, much of it radiates away in the form of photons. The infrared radiation is re-absorbed by the glass in the bulb itself and causes it to heat up some (in addition to heat transfered by conduction). But it still passes much of the infrared, ultraviolet, etc. energy out into the room. They don't heat up as much PER WATT when compared to LEDs.
LEDs don't emit anything but a narrow wavelength. Unfortunately, many of the photons get re-abosrbed by the substrate. All of this gets converted into heat. Some of it is converted into microwaves and radio waves, but not much. This is because the large package area (and small power output compared to a light bulb) allow it to conduct the heat into the environment, either through the plastic package or through the thick metal leads allow into whatever its attached to. They don't get hot to the touch unless you put a lot in a small space.
Very few of them actually linux viruses (or even understand ELF). They are designed to find windows or mac viruses in files, assuming the server houses a hosted file system, database blobs, mail spool, ftp site w/upload, etc.
This is why you can get McAfee for Sun Solaris, you use the same virus signature files you do for the PC version as you would for the Sun installation.
I guess if Linux viruses started popping up more frequently (or Solaris viruses) then those signatures might start appearing in the dats. But right now, that's fringe stuff. (Supposedly the DATs will flag files on unix system that match a few known unix virus/worms, but I'm not sure about that)
A high-efficiency red LED puts out about 2.5 mW of luminous energy in its emission band, and consumes 120 mW of power. That's 3% efficiency. The rest is dissapated as heat. Incidentally, that's why LEDs have a large footprint (the luminous area is very small); so the heat can spread out and the junction characteristics don't change. Incandescents emit light energy outside the visible band, unlike LEDs. This is where most of the power goes, not heat. Thus, incadescent lights achieve about 15 lumens per watt, flourscents get about 50 per watt. Contrary to popular belief, LEDs are in between, the high efficiency models get about 25 lumens per watt.
The most efficient LED right now is %32. You can't buy these yet... they will be used in lights that operate like flourescent lights since they emit UV. This will be the ideal, long-lasting but low power light source.
LEDs are not economical when a flourscent light with electronic ballast can be used in the same situation. In scenarios where the extra electronics required by a flourscent light are too bulky or not enough power is available- this is where LEDs shine. That is why they are the flash-light champs.
1. Set up a ramdisk on a machine with lots of RAM. 2. Set up a network block device to export said ramdisk. 3. Set up client using nbd-client to talk to server with network block device. 4. swapon/dev/nd0 5. profit!!!
Using NFS for disk-based swap is possible but silly since you incur the extra overhead. NBD works on a plain vanilla TCP connection and avoids touchy issues like memory vs. packet fragmentation. If you have a gigabit ethernet card with zero-copy support in the driver, then you are in business.
Thank you for sending us a copy of your book, "The Unix-Haters Handbook" to us. We've taken a look at it, and realized how misguided we have been.
As we are quite pragmatic, we decided to fix these outstanding issues. It's much better now; you would be proud. In fact, we did a good enough job with your guidance that Macs everywhere are now using it too!
Thank you for sending us a copy of your book, "The Unix-Haters Handbook" to us. We've taken a look at it, and realized how misguided we have been.
As we are quite pragmatic, we decided to fix these outstanding issues. It's much better now; you would be proud. In fact, we did a good enough job with your guidance that Macs everywhere are now using it too!
But doing so on the people you can influence (the operators of legitimate mail servers serving local users) will prevent the situation where a RBL captures a whole domain due to the compromise of a local account. You don't need to figure out how to do a full authentication chain yet (that's the role PGP fills right now). Once you get to a certain critical mass acceptance, then you can go full force (forcing the servers to authenticate to each other using shared secrets). Presumably, at this point there would be trusted MXs that allow connections from mail servers not running SMTP AUTH because they can't use it for whatever reason, but they would be whitelists.
That situation doesn't seem to far in the future. My ISP (Cox) already uses cram-md5 SMTP AUTH. At least I don't have to worry about someone impersonating me through their server. That's one step closer.
If this claim is correct, then there should be no further need for discussion. I was dubious about SCO, and this was because I assumed IBM was a bit more "with-it" then SCO was making them out to be. I figured they weren't that naive or sloppy.
Can you give me more information about where you picked that tidbit up?
Slashdot Mirror
I've got lots to trade.
Do you even know what I meant by that, oh glorious moderator? It is most certainly ON TOPIC as it pertains to how the game Final Fantasy VII DEEPLY AFFECTED ME.
Christ almighty, where's my account with metamod...
That thing is really wicked. It's halfway there (I like how it charges in the cradle... probably a better idea than alkaline batteries). If you could figure out how to lose the pad, and make it wireless... then a firmware update... boom!
If it's too hard to make it truly freehand, it might be acceptable to use a set of magic buttons, "fairy clips", I don't know what you'd call it, maybe it looks like a ruler, but the purpose is to establish an origin, a baseline on the "paper" with which to compare it's position using sonar, RF, whatever. You would like to think that you could do a semi-decent job using dead-reckoning, and maybe a little user guided post-processing to get your scribbles untangled. But maybe that's asking a little much.
It's a shame to have to carry around the pad.
It's just the jabber server. Presumably, you would hook it up to a PDA or laptop (like a whip antenna or stylus) and use interactively it that way. The reason for it being a jabber server is so that it can queue up XML messages people send you, and send out pre-programmed responses. Like business cards, presentation notes in MathML, etc. Not so much for the interactive features.
::bawling:: ::hugs d^2/dx^2, cries harder::
... sniffle... got her final limit... sniffle... and everything.
sniffle
I
Buwaaaahhhhaa-haaaaa! sniff
I don't think there would be a need for a seperate "personal server". The device would respond to requests from files, and any server-like features are just response to complex requests. You could organize it around a whole XML-RPC framework with specific types of requests all of which are heavily geared towards processing on the client side at presentation time. You provide the data in a standardized format (or use some fancy XML markup) and keep it in there.
Also, I don't think that using microdrives is a good idea because solid state is better suited for battery operation. Plus you don't need expensive drive electronics to deal with user handling. You can translate those savings into larger storage capacities.
If you were worried about making huge volumes of data available, I would be carrying around a USB-enabled notebook drive, not a repurposed iPod.
Looks like a pen. Writes like a pen.
But it's not a pen. It contains 256 MB (or so) of flash, which is shared via bluetooth. A 10cm high gain antenna is hidden within the length of the pen itself, and powered by a single AAA battery. Walk by an enabled PC, optionally type in a password, and all your documents, your keyring, etc. are available.
Finally, as an added bonus, when you write on paper (or anything for that matter), you can choose to record your scribbles on the flash drive. Tiny gyroscopic sensors determine the motion of the pen across the page, and a pressure sensor determines whether the pen is against a writing surface. Each time you expose the ball point head it creates a new file, and when you retract it, it closes it. You can tell which file is which by the timestamps.
THAT would kick ass. And as embedded logic gets more powerful, you could have a personal web/email/jabber server running in there too.
A wireless iPod sounds nifty, but where's the innovation people?
n\t
and at work, we don't have religious wars about operating systems. We think computers are cool. We like our jobs. We use a whole range of stuff. Sure we bitch about Sun's high prices and how we have to buy our personal pizza boxes off ebay, but that's to be expected.
I think the problem is slashdot itself, it's a place where people eventually butt heads. There's nothing to do here but talk, share, debate, argue. That's not bad, but don't think that slashdot is an analogue of what people do, think, or say all day.
Right after I posted that comment I tried to explain to someone the virtues of virtual memory -> physical memory mapping; including the whole "don't allocate 'till you see a page fault on it" thing. He agreed it was quite clever.
I was typing while under sudden duress. Fosttjrs@@@
I was all better, thanks.
If you use linux with BIG_MEM (3GB/1GB split) enabled in the kernel, your .text segment is allocated starting at 0x00010000. I forget why but there was some simplicity to using the 1GB/1GB split otherwise on linux (.text starting at 0x80000000, kernel mem at 0xc0000000). Windows uses a default 2GB/2GB split, so it makes sense it would start at 0x0. I don't think they were thinking about ASCII armor when they made that decision in the NT 3.x days.
On Linux and Windows, the .text segment of the currently rybbubg program starts one page AFTER 0. the page containing the address 0 is marked no-access so that NULL pointer deferences PURPOSEFULLY crash the program. They waste 4k of memory for that feature.
Otherwise, the program would just keep going. YOu wouldn't see the crash until you attempt to write there and clobber your code.
For about 3 and a half grand (which isn't too steep, bout the same for a nice Xeon system). The chips are the most expensive part (about $750 each). Motherboard, $300. RAM, case, monitor, HDs, the rest are whatever you get them for.
Personally, we spent $10,000 on one of these, put it had dual 20" LCDs on a Quadra4 GLX, and 3 each of WD 10k RPM and Seagate Barracuda V SATA HDs. And a Midiman 1010LT for sound, because we're snobs. w000000t. Can't wait to try this tech preview on it.
and compile your own damn version. Not hard.
Allow users to choose color scheme in preferences? You can enforce a default, and all the whiners can tweak it to their hearts' content.
Hasn't anyone added a patch to slashcode that actually uses CSS? Jeez.
But I do like the idea. games.slashdot.org and science.slashdot.org... and I wonder where does the time go? ^_^
Did you misread the part where I said LEDs are the winner in flashlight choice? This is because they beat incadescent lights in lumens per watt numbers (the numbers get worse for incadescents too as they get smaller, like in flashlight bulbs). They don't beat flourescent lights, though. That was my point.
Light bulbs emit infrared, radio, microwave, and ultraviolet light (as they are blackbody sources). The heat from the filament is not easily conducted away into the base of the bulb or through the gas inside. Hence, much of it radiates away in the form of photons. The infrared radiation is re-absorbed by the glass in the bulb itself and causes it to heat up some (in addition to heat transfered by conduction). But it still passes much of the infrared, ultraviolet, etc. energy out into the room. They don't heat up as much PER WATT when compared to LEDs.
LEDs don't emit anything but a narrow wavelength. Unfortunately, many of the photons get re-abosrbed by the substrate. All of this gets converted into heat. Some of it is converted into microwaves and radio waves, but not much. This is because the large package area (and small power output compared to a light bulb) allow it to conduct the heat into the environment, either through the plastic package or through the thick metal leads allow into whatever its attached to. They don't get hot to the touch unless you put a lot in a small space.
Very few of them actually linux viruses (or even understand ELF). They are designed to find windows or mac viruses in files, assuming the server houses a hosted file system, database blobs, mail spool, ftp site w/upload, etc.
This is why you can get McAfee for Sun Solaris, you use the same virus signature files you do for the PC version as you would for the Sun installation.
I guess if Linux viruses started popping up more frequently (or Solaris viruses) then those signatures might start appearing in the dats. But right now, that's fringe stuff. (Supposedly the DATs will flag files on unix system that match a few known unix virus/worms, but I'm not sure about that)
A high-efficiency red LED puts out about 2.5 mW of luminous energy in its emission band, and consumes 120 mW of power. That's 3% efficiency. The rest is dissapated as heat. Incidentally, that's why LEDs have a large footprint (the luminous area is very small); so the heat can spread out and the junction characteristics don't change. Incandescents emit light energy outside the visible band, unlike LEDs. This is where most of the power goes, not heat. Thus, incadescent lights achieve about 15 lumens per watt, flourscents get about 50 per watt. Contrary to popular belief, LEDs are in between, the high efficiency models get about 25 lumens per watt.
The most efficient LED right now is %32. You can't buy these yet... they will be used in lights that operate like flourescent lights since they emit UV. This will be the ideal, long-lasting but low power light source.
LEDs are not economical when a flourscent light with electronic ballast can be used in the same situation. In scenarios where the extra electronics required by a flourscent light are too bulky or not enough power is available- this is where LEDs shine. That is why they are the flash-light champs.
1. Set up a ramdisk on a machine with lots of RAM. /dev/nd0
2. Set up a network block device to export said ramdisk.
3. Set up client using nbd-client to talk to server with network block device.
4. swapon
5. profit!!!
Using NFS for disk-based swap is possible but silly since you incur the extra overhead. NBD works on a plain vanilla TCP connection and avoids touchy issues like memory vs. packet fragmentation. If you have a gigabit ethernet card with zero-copy support in the driver, then you are in business.
Have another go at it! It's fun
Thank you for sending us a copy of your book, "The Unix-Haters Handbook" to us. We've taken a look at it, and realized how misguided we have been.
As we are quite pragmatic, we decided to fix these outstanding issues. It's much better now; you would be proud. In fact, we did a good enough job with your guidance that Macs everywhere are now using it too!
Thanks again,
Unix Users Everywhere.
Or who's password's been stolen. ^_^
Thank you for sending us a copy of your book, "The Unix-Haters Handbook" to us. We've taken a look at it, and realized how misguided we have been.
As we are quite pragmatic, we decided to fix these outstanding issues. It's much better now; you would be proud. In fact, we did a good enough job with your guidance that Macs everywhere are now using it too!
Thanks again,
Unix Users Everywhere.
But doing so on the people you can influence (the operators of legitimate mail servers serving local users) will prevent the situation where a RBL captures a whole domain due to the compromise of a local account. You don't need to figure out how to do a full authentication chain yet (that's the role PGP fills right now).
Once you get to a certain critical mass acceptance, then you can go full force (forcing the servers to authenticate to each other using shared secrets).
Presumably, at this point there would be trusted MXs that allow connections from mail servers not running SMTP AUTH because they can't use it for whatever reason, but they would be whitelists.
That situation doesn't seem to far in the future. My ISP (Cox) already uses cram-md5 SMTP AUTH. At least I don't have to worry about someone impersonating me through their server. That's one step closer.
If this claim is correct, then there should be no further need for discussion.
I was dubious about SCO, and this was because I assumed IBM was a bit more "with-it" then SCO was making them out to be. I figured they weren't that naive or sloppy.
Can you give me more information about where you picked that tidbit up?