It's my impression that, while 802.11b and X10 cameras use the same 2.4 GHz spread-spectrum technology (as do many other devices like cordless phones), they don't have anything in common at higher levels -- it's not like X10 cameras implement the Ethernet spec or anything (are they even digital?). The Slashdot article I remember about "warwatching" mentioned an apparatus made from an X10 reciever module and a small LCD TV -- no 802.11b equipment or computer involved at all. If there are devices that are capable of monitoring both 802.11b and X10 transmissions (and getting interesting data from them), please post links -- I'd be interested in such a thing.
As for encrypting an X10 camera, it seems really unlikely that it would be worth the effort and expense. It would probably be much more efficient to get a plain old webcam and attach it to an old machine with a WiFi card. Then, you have the flexability to use whatever encryption/security you want. You could also perhaps buy an (expensive) network-attached webcam like the Axis webcams and plug it into a wired Ethernet to wireless adapter with WEP, or find a webcam that has 802.11b built in. If you are going to rely on WEP for security, however, please be aware of its *ahem* less-than-stellar record.
Or a dot-matrix printer. Seriously, I did this for a while; you can turn on console-on-LPT support in your kernel config, and pass a parameter with your bootloader. It takes a while for the stupid thing to display all the kernel messages at boot, but the sound is priceless. Sadly, the 20-some-year-old printer decided to kick the bucket (still lasted longer than my HP DeskJet, thank you very much), so I switched to a 286 laptop running Minix 1.5 and term, which might be a cheap way to implement the serial2IP idea.
anti-aliasing is not pseudo analog, it increases the effective resolution.
This is only really true when you're talking about subpixel antialiasing, which is generally only usable on LCDs and relies on the red, green, and blue subpixels being in a specific order. Plain old antialiasing just surrounds diagonal lines with grey to make them less jagged. You could essentially think of it as an intelligent blur. I think that the parent poster was referring to this as pseudo-analog because when you draw a diagonal line in a non-raster (aka "analog") context, and then try to rasterize it, you will have pixels that are only partially darkened. Normally, these pixels are just left unfilled. (Non-subpixel) antialiasing uses levels of grey (or whatever is appropriate for the combination of foreground and background) to simulate this analog effect a bit better.
That's an incomplete way of thinking about it. It doesn't matter if you prevent someone from writing something, or just preventing anyone else from reading it: either way, you're stopping a potential communication between writer and reader. So in a sense, the voice of the subscribers talking to non-subscribers has been silenced. Of course, these subscribers can just move to another forum to share views with non-subscribers -- indeed, I expect that to happen to a large degree now. But they have lost part of their voice in the official, sponsored forum.
If you own guns, it might be argued that you had the obligation to lock your doors... I certainly wouldn't want to be sued because a gun I owned wound up on the floor of a neighborhood Stop-N-Rob, next to a dead clerk, simply because some crystal-meth user was able to simply wander into my home and steal said gun... I can think of more than a few plaintiffs' attorneys that might argue that angle in a wrongful death suit.
You know, it's interesting to take those comments in a computer-security context. Compromised machines are often used to send spam, conduct DDoS attacks, and otherwise wreak havok on the Internet -- many of them compromised by script kiddies, the "crystal-meth users" of the Internet. It seems odd then that while the average gun owner knows to take at least basic security precautions with his/her weapon, the average computer owner isn't even aware that a broadband internet connection can be used as a weapon.
How can user awareness be raised? Hell if I know. But it needs to be done: right now the Internet is like a row of houses where 90% have a loaded AK-47 lying on the front doorstep.
However, if you don't need more than what the consumer firewall provides it is a very cheap solution.
It is true that OpenBSD systems may be more expensive in terms of TCO than a $50 home router, but only if your time has value. If you're a bored student with too much free time like me, you can get an OpenBSD router/firewall up and running literally for free, using old commodity hardware. I've used OpenBSD as a router on machines from a Pentium MMX 266MHz all the way down to a Pentium 60MHz, with no problem routing the full capacity of my cable modem. I even ran it on an AlphaStation for a brief period of time (though the OpenBSD/Alpha kernel seems to have some nasty crash issues relating to the router/NAT system -- anyone know how to fix this?)
I'm not sure, but I believe that Glider PRO is PowerPC-only, wheras Basilisk only emulates 68k Macintoshes. By the way, Glider 4.0 will run in 16-color mode, at least on my Powerbook.
I don't quite get your point about x86. Most of the code in the Linux kernel is architecture-independent, and will apply equally to whatever we're using five years from now (which could very well be x86, seeing as it's already been around for more than that). Or is SCO targeting specifically the x86 portion of the Linux kernel? If so, a fair amount of the x86 code is reused for other similar processors.. how will this affect them?
32 bits gives you 4 GB of total address space. This is not enough to map a large database file alongside program and library code and variable space. So the program would have to swap different pieces of the file in and out of memory, in a non-transparent way. This is really no better than explicit read()/write(). 64 bits provides about 4 billion times as much memory space as 32 bits. Even if the machine does not have nearly this much physical RAM, the address space can be used by the VM to do file mapping and other fun stuff.
Remember that laserdisc video is analog too (and the audio also, except on newer discs that use CD-like PCM). So the laserdiscs may have degraded much like the film has. As far as which would be better quality, your guess is as good as mine. There are a lot of variables involved: storage conditions, materials used, etc. I do know that the right kind of film can give fantastic resolution if processed and stored right, whereas IIRC the resolution of laserdisc is only somewhat better than VHS and not as good as DVD. So I'm guessing that the decision to use original films was a good one.
Interesting sidenote: perhaps the laserdisc data could be used to restore pieces of film lost to physical degradation, by combining the two streams. Any video processing gurus want to comment on the feasability of this?
IANAL, but as far as I know, Minix is almost public domain: you can do nearly anything with it (licence here).
As far as portability, the older version of Minix (1.5) runs on 68k (Amiga, Atari, and Macintosh) and SPARC, as well as x86 (32- and 16- bit). However, the 6502 used in the Apple 1/2/3 series is quite a bit more primitive than even a MMU-less x86 or 68k. In particular, fewer and smaller registers means more swapping to memory, which in turn reduces speed and increases code size. The 16-bit version of the 6502 used in the IIgs would be somewhat better than the 8-bit 6502s in the older Apples, but maybe not by enough. Also, Minix uses a microkernel architecture, which is all very nice from a theoretical standpoint, but causes a severe performance hit on really old hardware.
Is Minix on IIgs possible? Theoretically yes, since it's a Universal Turing Machine*. However, it's unlikely to fit within the extremely limited memory and speed constraints. Crafting a new OS from scratch would probably be much easier. But yes, it would be damn cool.
(*Yes, a UTM is assumed to have infinite storage space, so for theoretical purposes I'm assuming that we can hook a 6502 IIgs up to 512MB SDRAM or something.)
If it's that important, you shouldn't be transferring it over the internet at all.
Right, but that kinda defeats the purpose of the question.;-)
But yes, I do agree that crypto is nowhere near a bulletproof suggestion (literally, if you're talking DoS attacks). For it to be of any use, you still need basic smart computing practices: don't give out your passphrase, don't install a Trojan horse, keep your computer in a secure physical location, etc. But the fact of the matter is that just using SSL for credit card transactions puts you above the low-hanging fruit (in this case, people sharing their financial data on KaZaA or Windows fileshares, or emailing it in the clear). If you're plotting world domination over the internet, that's another thing.
It's a question of hardware, not software; most current wardriving software has support for other systems such as 802.11a and g as well as b. In theory, they could also sniff on 900 MHz if there were 900 MHz cards that supported the right software interfaces.
Remember that there's much more to a secure connection than just the "last mile" from your ISP to you. Depending on the network itself for security is risky at best. To really ensure security, you should always assume that your neighbors, your local bank robbers, and the NSA are all snooping on you, and simply encrypt anything important. And please use a good algorithm; WEP doesn't count (this is aimed mostly at the 802.11b crowd, obviously).
I haven't tried it, but it seems to me that this could be accomplished by putting a DNS caching router machine between the PS2 and the internet, and then remapping the DNS for Sony's site over to the site mentioned in the article. This should work on American consoles, unless they don't have a Web browser at all. Of course, needing a router machine probably defeats the purpose of a standalone web browser.
This is much different. Yes, those warez disclaimers are bullshit. That's because they're essentially a licence term ("You will not download this if you're a Fed") imposed by someone (l33t w4rez kiddie) on a work that he is not the copyright owner of (Adobe Photoshop or whatnot). You can't impose terms on intellectual property that you don't own. In this modchip case, the person who made the modchip is imposing terms on his own IP: the design of the modchip that he created. IANAL, but it seems just as legal to say "Feds can't download this software that I made" as it does to say "You can't reverse-engineer this software that I made". Now, in the real world, Microsoft's lawyers will find some technicality and sue the hell out of the poor guy. But, in principle, he's allowed to do this to his own IP.
The key to understanding WEP is the phrase "Wired Equivalency". The theory is that WEP, although a fairly weak cypher, provides the same level of privacy as unencrypted wired Ethernet. That is, breaking WEP is judged to be approximately as difficult as finding somewhere to jack into a wired Ethernet (i.e. not very). WEP never was intended to take the place of encryption systems such as SSL and IPSec that are conventionally used to secure connections over wired networks. Rather, it brings WiFi security to the level of security inherent in wired Ethernet. Thus, WiFi using WEP is insecure only because of the way it is marketed: users see it as a catch-all encryption system, rather than a replacement for the (fairly weak) security inherent to wired Ethernet's physical-access requirement.
This certainly seems to be the trend for defeating conole protection systems in software -- unless all code that has been signed is 100% secure from local attacks (ha!), you can simply insert code into a signed program and take control. It seems like this might be a weak point in new computer security systems like Microsoft's Palladium (in fact, it's been rumored that the Xbox's protection system is a testing version of Palladium, and that's been completely broken this way). If Palladium is being used for DRM for example, then a bug in Windows Media Player might let a user slip in code to save the decoded audio stream, thereby defeating the DRM. But it could also be used by malicious programs to defeat security measures imposed on behalf of the user (indeed, as buffer overflows are used now). I guess it comes down to this question: How will systems like Palladium guard against security holes in signed programs that could compromise the whole security model?
Look into a UNIX utility called screen. I'm not sure if it's included in OS X, but I use it all the time on Debian and OpenBSD and it's wonderful. You can detatch and reattach sessions, so if Terminal crashes, you can just open another one and run screen -R to pick up where you were. It also supports creating multiple windows in the same terminal, and a whole ton of other features that I haven't even tried yet. I've even used it to start a process locally, then ssh in later and automagiclly take control of the process. It should compile without problems on OS X; there might even be binaries available.
Keep in mind that the PowerPC architecture can do quite a bit more per clock tick than x86 hardware, so a 1.33 GHz PowerPC can probably perform about the same as at least a 2 GHz Pentium 4. Of course, this would vary quite a bit depending on the task at hand; only benchmarks will show the real numbers. Anyway, the PowerPC(s) in the Xserve have a good amount of raw computing power, and this will only improve as Apple moves to PPC64 chips.
It seems to me that Apple is looking to move back to the older concept of the UNIX server: a high-end server and a (mostly) proprietary UNIX operating system sold together as a unit. For a while, it seemed like Linux and cheap-as-dirt x86 hardware were going to do away with this; now Apple is trying to introduce it as a product, albeit with more of an open soure component. Only time will tell if they can make money on this. My guess? They'll get a steady but not dominating niche market, much like they have with home computers (and for that matter, much as "big iron" UNIX still has.) They may have a problem with people installing Linux on the Xserves and then not paying for OS X Server software or upgrades; then again, their chosen market may not even consider this. Again, only time will tell.
I agree that DoS attacks are really lame, but at least this is fairly complex and has some interesting technological implications. Most DoS attacks are just floods of packets, and can't really be defended against. This attack can be prevented, and maybe will help improve security in other ways. I'd rather see 10 people who can pull off a preventable algorithmic complexity attack than 10,000 who just packet a server to death. People will write tools, but they'll still remain too hard for moron DDoSer script kiddies to pull off.
"Video games don't affect kids. If Pacman had affected us when we were kids, everyone would be running around in darkened rooms, munching on magic pills and listening to repetitive electronic music." -- Davey Whipwreck
(Is that the right source? I know I've seen it elsewhere attributed differently.)
Yeah, I got that too. The title went blue also... #3300cc, not the color used for Games section stories. I'm not a subscriber or anything fancy like that. The article now shows up fine except that the titlebar is still blue. Maybe they just edited Slashcode to have different colored article titles, and messed something up in the process. Screenshot here.
Slashdot Mirror
It's my impression that, while 802.11b and X10 cameras use the same 2.4 GHz spread-spectrum technology (as do many other devices like cordless phones), they don't have anything in common at higher levels -- it's not like X10 cameras implement the Ethernet spec or anything (are they even digital?). The Slashdot article I remember about "warwatching" mentioned an apparatus made from an X10 reciever module and a small LCD TV -- no 802.11b equipment or computer involved at all. If there are devices that are capable of monitoring both 802.11b and X10 transmissions (and getting interesting data from them), please post links -- I'd be interested in such a thing.
As for encrypting an X10 camera, it seems really unlikely that it would be worth the effort and expense. It would probably be much more efficient to get a plain old webcam and attach it to an old machine with a WiFi card. Then, you have the flexability to use whatever encryption/security you want. You could also perhaps buy an (expensive) network-attached webcam like the Axis webcams and plug it into a wired Ethernet to wireless adapter with WEP, or find a webcam that has 802.11b built in. If you are going to rely on WEP for security, however, please be aware of its *ahem* less-than-stellar record.
Hmm, cross-site scripting. Seems harmless enough, but I wonder if VeriSign stores anything important in the verisign.com cookie...
Or a dot-matrix printer. Seriously, I did this for a while; you can turn on console-on-LPT support in your kernel config, and pass a parameter with your bootloader. It takes a while for the stupid thing to display all the kernel messages at boot, but the sound is priceless. Sadly, the 20-some-year-old printer decided to kick the bucket (still lasted longer than my HP DeskJet, thank you very much), so I switched to a 286 laptop running Minix 1.5 and term, which might be a cheap way to implement the serial2IP idea.
This is only really true when you're talking about subpixel antialiasing, which is generally only usable on LCDs and relies on the red, green, and blue subpixels being in a specific order. Plain old antialiasing just surrounds diagonal lines with grey to make them less jagged. You could essentially think of it as an intelligent blur. I think that the parent poster was referring to this as pseudo-analog because when you draw a diagonal line in a non-raster (aka "analog") context, and then try to rasterize it, you will have pixels that are only partially darkened. Normally, these pixels are just left unfilled. (Non-subpixel) antialiasing uses levels of grey (or whatever is appropriate for the combination of foreground and background) to simulate this analog effect a bit better.
Actually, wait, Windows CE can run on MIPS. Maybe it would actually be worthwhile.
That's an incomplete way of thinking about it. It doesn't matter if you prevent someone from writing something, or just preventing anyone else from reading it: either way, you're stopping a potential communication between writer and reader. So in a sense, the voice of the subscribers talking to non-subscribers has been silenced. Of course, these subscribers can just move to another forum to share views with non-subscribers -- indeed, I expect that to happen to a large degree now. But they have lost part of their voice in the official, sponsored forum.
You know, it's interesting to take those comments in a computer-security context. Compromised machines are often used to send spam, conduct DDoS attacks, and otherwise wreak havok on the Internet -- many of them compromised by script kiddies, the "crystal-meth users" of the Internet. It seems odd then that while the average gun owner knows to take at least basic security precautions with his/her weapon, the average computer owner isn't even aware that a broadband internet connection can be used as a weapon.
How can user awareness be raised? Hell if I know. But it needs to be done: right now the Internet is like a row of houses where 90% have a loaded AK-47 lying on the front doorstep.
It is true that OpenBSD systems may be more expensive in terms of TCO than a $50 home router, but only if your time has value. If you're a bored student with too much free time like me, you can get an OpenBSD router/firewall up and running literally for free, using old commodity hardware. I've used OpenBSD as a router on machines from a Pentium MMX 266MHz all the way down to a Pentium 60MHz, with no problem routing the full capacity of my cable modem. I even ran it on an AlphaStation for a brief period of time (though the OpenBSD/Alpha kernel seems to have some nasty crash issues relating to the router/NAT system -- anyone know how to fix this?)
I'm not sure, but I believe that Glider PRO is PowerPC-only, wheras Basilisk only emulates 68k Macintoshes. By the way, Glider 4.0 will run in 16-color mode, at least on my Powerbook.
You could always set up Basilisk II and run the older Mac version (4.0, listed at the bottom of the page) inside an emulated 68k Mac.
I don't quite get your point about x86. Most of the code in the Linux kernel is architecture-independent, and will apply equally to whatever we're using five years from now (which could very well be x86, seeing as it's already been around for more than that). Or is SCO targeting specifically the x86 portion of the Linux kernel? If so, a fair amount of the x86 code is reused for other similar processors.. how will this affect them?
32 bits gives you 4 GB of total address space. This is not enough to map a large database file alongside program and library code and variable space. So the program would have to swap different pieces of the file in and out of memory, in a non-transparent way. This is really no better than explicit read()/write(). 64 bits provides about 4 billion times as much memory space as 32 bits. Even if the machine does not have nearly this much physical RAM, the address space can be used by the VM to do file mapping and other fun stuff.
Interesting sidenote: perhaps the laserdisc data could be used to restore pieces of film lost to physical degradation, by combining the two streams. Any video processing gurus want to comment on the feasability of this?
As far as portability, the older version of Minix (1.5) runs on 68k (Amiga, Atari, and Macintosh) and SPARC, as well as x86 (32- and 16- bit). However, the 6502 used in the Apple 1/2/3 series is quite a bit more primitive than even a MMU-less x86 or 68k. In particular, fewer and smaller registers means more swapping to memory, which in turn reduces speed and increases code size. The 16-bit version of the 6502 used in the IIgs would be somewhat better than the 8-bit 6502s in the older Apples, but maybe not by enough. Also, Minix uses a microkernel architecture, which is all very nice from a theoretical standpoint, but causes a severe performance hit on really old hardware.
Is Minix on IIgs possible? Theoretically yes, since it's a Universal Turing Machine*. However, it's unlikely to fit within the extremely limited memory and speed constraints. Crafting a new OS from scratch would probably be much easier. But yes, it would be damn cool.
(*Yes, a UTM is assumed to have infinite storage space, so for theoretical purposes I'm assuming that we can hook a 6502 IIgs up to 512MB SDRAM or something.)
Right, but that kinda defeats the purpose of the question. ;-)
But yes, I do agree that crypto is nowhere near a bulletproof suggestion (literally, if you're talking DoS attacks). For it to be of any use, you still need basic smart computing practices: don't give out your passphrase, don't install a Trojan horse, keep your computer in a secure physical location, etc. But the fact of the matter is that just using SSL for credit card transactions puts you above the low-hanging fruit (in this case, people sharing their financial data on KaZaA or Windows fileshares, or emailing it in the clear). If you're plotting world domination over the internet, that's another thing.
I haven't tried it, but it seems to me that this could be accomplished by putting a DNS caching router machine between the PS2 and the internet, and then remapping the DNS for Sony's site over to the site mentioned in the article. This should work on American consoles, unless they don't have a Web browser at all. Of course, needing a router machine probably defeats the purpose of a standalone web browser.
This is much different. Yes, those warez disclaimers are bullshit. That's because they're essentially a licence term ("You will not download this if you're a Fed") imposed by someone (l33t w4rez kiddie) on a work that he is not the copyright owner of (Adobe Photoshop or whatnot). You can't impose terms on intellectual property that you don't own. In this modchip case, the person who made the modchip is imposing terms on his own IP: the design of the modchip that he created. IANAL, but it seems just as legal to say "Feds can't download this software that I made" as it does to say "You can't reverse-engineer this software that I made". Now, in the real world, Microsoft's lawyers will find some technicality and sue the hell out of the poor guy. But, in principle, he's allowed to do this to his own IP.
The key to understanding WEP is the phrase "Wired Equivalency". The theory is that WEP, although a fairly weak cypher, provides the same level of privacy as unencrypted wired Ethernet. That is, breaking WEP is judged to be approximately as difficult as finding somewhere to jack into a wired Ethernet (i.e. not very). WEP never was intended to take the place of encryption systems such as SSL and IPSec that are conventionally used to secure connections over wired networks. Rather, it brings WiFi security to the level of security inherent in wired Ethernet. Thus, WiFi using WEP is insecure only because of the way it is marketed: users see it as a catch-all encryption system, rather than a replacement for the (fairly weak) security inherent to wired Ethernet's physical-access requirement.
This certainly seems to be the trend for defeating conole protection systems in software -- unless all code that has been signed is 100% secure from local attacks (ha!), you can simply insert code into a signed program and take control. It seems like this might be a weak point in new computer security systems like Microsoft's Palladium (in fact, it's been rumored that the Xbox's protection system is a testing version of Palladium, and that's been completely broken this way). If Palladium is being used for DRM for example, then a bug in Windows Media Player might let a user slip in code to save the decoded audio stream, thereby defeating the DRM. But it could also be used by malicious programs to defeat security measures imposed on behalf of the user (indeed, as buffer overflows are used now). I guess it comes down to this question: How will systems like Palladium guard against security holes in signed programs that could compromise the whole security model?
Look into a UNIX utility called screen. I'm not sure if it's included in OS X, but I use it all the time on Debian and OpenBSD and it's wonderful. You can detatch and reattach sessions, so if Terminal crashes, you can just open another one and run screen -R to pick up where you were. It also supports creating multiple windows in the same terminal, and a whole ton of other features that I haven't even tried yet. I've even used it to start a process locally, then ssh in later and automagiclly take control of the process. It should compile without problems on OS X; there might even be binaries available.
It seems to me that Apple is looking to move back to the older concept of the UNIX server: a high-end server and a (mostly) proprietary UNIX operating system sold together as a unit. For a while, it seemed like Linux and cheap-as-dirt x86 hardware were going to do away with this; now Apple is trying to introduce it as a product, albeit with more of an open soure component. Only time will tell if they can make money on this. My guess? They'll get a steady but not dominating niche market, much like they have with home computers (and for that matter, much as "big iron" UNIX still has.) They may have a problem with people installing Linux on the Xserves and then not paying for OS X Server software or upgrades; then again, their chosen market may not even consider this. Again, only time will tell.
I agree that DoS attacks are really lame, but at least this is fairly complex and has some interesting technological implications. Most DoS attacks are just floods of packets, and can't really be defended against. This attack can be prevented, and maybe will help improve security in other ways. I'd rather see 10 people who can pull off a preventable algorithmic complexity attack than 10,000 who just packet a server to death. People will write tools, but they'll still remain too hard for moron DDoSer script kiddies to pull off.
(Is that the right source? I know I've seen it elsewhere attributed differently.)
Yeah, I got that too. The title went blue also... #3300cc, not the color used for Games section stories. I'm not a subscriber or anything fancy like that. The article now shows up fine except that the titlebar is still blue. Maybe they just edited Slashcode to have different colored article titles, and messed something up in the process. Screenshot here.