Lately I started using the awsome bar, but I used to hate it. What got me is that I have my homedirs exported via NFS. That in itself was annoying for using firefox since I needed different profiles for different machines in order to be able to run more than one firefox at a time on different hosts. But after a while I had a nice system where I could use a few profiles on all my machines and have bookmarks and history follow me around. But then the awesome bar made it so that those sqlite dbfiles were being used for everything. sqlite and nfs simply did not play well together. I ended-up having to symlink my firefox profiles to copies on local disks. So for all that time that I did not use the awesome bar, I lost the easy way of using a few profiles on many different machines. That is what bugged me the most about awesome bar.
No x86 does 0-9 BCD only, 754r would need 0-999 BCD for any speed improvement compared to 11 adds with carry. Also I think that the BCD opcodes never got extended circa 80286 like the other ones did to allow other registers to be used.
Google is using anycast for their DNS servers. There are not just two machines at 8.8.8.8 and 8.8.4.4 as the sole DNS servers. You get a relatively close-by server. This is a tried and true technique for DNS. In fact there is a technical feature about the google approach that is neat. It is likely that google is using many of the same servers it is for search for the DNS servers as well. They are running the caching DNS at each facility, such that if one server at the facility gets a record, then any other DNS server at that facility uses that response. That is one cool way to limit the delays for someone else making a DNS request. I've not seen that mentioned much before, and that is neat. I wish slashdot comments about stories that are trying to be technical would have technical comments on them near the beginning, instead of rehashing of all this privacy stuff, for a third or fourth story.
Another approach that was mentioned a lot before is that after the DNS server provides a response, the server checks to see if time is running-out regarding the TTL. If it is and has not expired yet, it asks again and pretends that the TTL counter has begun again. This again is trying to limit a DNS delay for some poor schmuck.
Another technical detail I have not seen mentioned much is that google DNS servers are returning largely authoritive answers only, often in cases where other DNS servers do not. For example, look-up a private IPv4 such as 192.168.1.1 with google's servers and some others. Others typically return non-authoritive responses, say to RFC1918.private.net. There is a lot of subtly misconfigured software-out there, hopefully this will bring it to the fore front about dealing with non-authoritive answers more carefully.
As to regarding the performance of google DNS, from a few locations for me, seems very fast. Is faster (much) than AT&T, bit slower than comcast, bit slower than work, comparison with OpenDNS is in the noise. What is more important is that they treat all records correctly, so for example kx509 _kca._udp.REALM style SRV records are handled unlike the DNS servers from some ISPs which seem to think that DNS is only for A records.
Another interesting feature is that google DNS is playing tricks with case in DNS queries and replies as yet another stop-gap-measure against DNS cache poisoning attacks. That's clever, I believe it was proposed before, but bind folks presented some issues and left it at that.
That happened to me too. Fox was having issues here. It was more goof-up rather than mal-intent. One time it was an ad that set the flag but did not unset it at the end of the ad. So then nothing recorded for hours until they fixed it.
I avoid this by uninstalling everything from Adobe. I bet you it was a flash ad utilizing an acrobat (or whatever it is they call it now a days) vulnerability. I have no flash and I have no acrobat. There are plenty of open source alternatives to Acrobat (even for Windows) but generally xpdf and gnash under VirtualBox is good enough or makes me decide I don't care about that site.
I don't understand. I typically have an account that is not a sudoer that I use for everything. Using allowusers in the sshd_config that is the only account that can ssh in. Then if I need to do something as root, I do 'login admin' (or slight variations depending on unix flavor), where admin is a sudoer and then I run what I need under sudo. Is that what you mean by two password wall? What is the security risk here? I know that there have been vulns with regards to sudo in the past, is that what you are referring to?
The company that made 68K a-like SoC handhelds. The one where I just built a cross compiler for on FreeBSD to write programs. The company that very good documentation for the time about how to program the PDA. There was a free and an open source emulator. You could buy Code Warrior to dev on if you did not feel like using a gcc+binutils cross compiler. The way to get data to and from your PDA was free for Mac and Windows, plus there were a bunch of open source tools to do the same. If you wanted to distribute your programs, you could just put them on a web or FTP site somewhere. You could sell the programs on CDs or whatever you felt like as well. The only thing that was anything remotely like this, was that there was a web site where you could reserve a 4 character creator code for your applications. As long as it was free, it was yours.
Long ago there was a replacement for AD, it was known as yp. It had to change it's name to NIS, now NIS+. With that and kerberized NFS, the whole Windows approach is just backwards. Where NFS does not apply rsync does. You can get Windows to authenticate from LDAP, even tie that in with kerberos. If you need Windows to see shared drives, on Solaris you can reexport home dirs as smbfs, and that hauls on Solaris 10 being an in kernel implementation with a big ZFS storage pool behind it.
They do, actually if I recall correctly in T2 they have one per core, where each consists of 3 cells each consisting of two n-well resistors (thermal noise essentially) connected to an amp which feeds into a VCO (voltage controlled oscillator) that gets sampled with regards to a different clock.
I'm very happy to hear about your success. I'm a big fan of FreeBSD as well.
Read the syncer man page, there are sysctls that you can tweak to slow down flushing the pagecache so your disk will not spin up as often. If your laptop is very stable you might not worry so much about a panic. The battery should give you enough time for a sync or halt in case of a power outage.
On a laptop, I find I hardly ever care about the stuff in/var after a reboot. Using md was clever.
Did you look into pdflush? In particular dirty_writeback_centisecs? lsof, iostat, vmstat, and strace in a process of elimination are useful as well. It is too bad linux (I am assuming you are on linux) does not have something like dtrace or at least fs_usage. Did you do the other stuff listed above? In particular you could have some stupid process that writes-out its config at something insane like once a second. There was some notorious Seagate firmware for a while too, but not on the small drives as far as I know. Good luck.
Look in cron and disable stuff you do not use, especially locate.
Do not use the optical drive and make sure you are quiescing it.
Turn-off access time modifications for the hard disk.
Turn off fsck on boot.
Turn off periodic SMART status checking (on some drives this spins it up).
Tune the time to idle the drives and the periodic disk flusher (you have basically UPS with a laptop anyway).
Turn off swap.
Use a light simple window manager such as fvwm2 instead of something like gnome where lots of files are being accessed all of the time and you have many procs/threads running and the neat effects burn the battery.
Find the docs to your graphics drivers and tweak the tunables to use as little power as possible (this will give you much more than you likely expect).
Turn off bluetooth and wireless when you are not using it.
Don't use any of the crazy sound daemons.
You probably don't need wake-on-magic-packet for a laptop, turn it off, it helps a lot for some NICs.
Do you use multicast for wireless, most likely not, read the docs and figure-out how to get your driver to ignore that, it can conserve more power on some cards.
With some of the older chips USB was very power hungry in sleep (if that's your case tweak what you can so that it does as little as possible, likely turning off the wake on keyboard and mouse since you shutting and opening the lid should handle that).
Install a flash blocker and/or ad blocker and use gnash where you can instead of the adobe version.
Man I wish I had seen that there had been a reply to my comment earlier. I do embedded/realtime at work. Most of it happens to be on PPC and 68K of various vintage. Then I do work on various ARM and x86 but fewer installations. So this was definitely not a case of x86 is different than everything else and esoteric things not used.
I almost always align things even on systems that do not trap, simply because it is mire efficient, portable, and the compiler does it automatically. What I was referring to with regards to ARM and the way compilers align for it is that the structures become aligned differently than PPC (and power), x86, 68K, SPARC, and MIPS (yes I have used all of those minimally 32-bit cpus for work at least once) in a way that is unexpected at first.
say you have this struct: { int8_t a; int16_t b; int16_t; }
How does it get layed-out on PPC and x86:
a0bbcc
How does it get laid-out in ARM:
a0bbcc00
That was the problem I was referring to.
The thing about ldr is that in my case only the kernel was at first non relocatable, I wanted to have a few tasks (porting 68K pSOS code) and just wanted them to be PIC and loaded where ever so I could have the same kind of scheme where I used to have where some installations would have different number and sorts of task running depending on configuration.
Then the DACR and friends stuff came up since again I wanted to have tasks like in pSOS but have some memory protection as well. It was such a mess I gave-up and I ended-up having one task sharing the kernel and I did my own round robin scheduling of code it different object files instead of separate tasks and a bit more in the ISRs than I would have liked to make it work.
Then when latter ARMs came out I was very glad that I had since all of the DACR stuff have changed. That largely does not happen on PPC or x86, at least not ever in an area that affected me.
I don't know if I would call ARM asm beautiful to look at. Possibly it's nicer to look at than x86, but by the time of the 386 there were practically no more examples left of instructions that were not orthogonal. Sure the byte code for some of those might have been four times as long as the original 8086 version that used AX only, but you hardly ever look at that.
There are lots of annoying little things with ARM. In the area of the instruction set itself, there is the whole issue of what happens when you load a 32-bit word that is unaligned. That depends on the mode you are in, but it usually does some crazy rotation, but that is actually useful to make use of. This also leads to strange alignment relative to everything else out there if you want the same code to be able to fill a structure say. Then there is the whole loading of constants using LDR which is not a real instruction, the assembler does what ever craziness it can manage to get a constant. If your code is PIC that basically means loading from memory, if it is noPIC then it might do crazy pc relative stuff. Oh and what about thumb mode, very non-orthogonal.
Then ARM has the whole issues of all the differences. When they have TLBs and caches they tend to be alien (DACR ASID and PID) though the newest ARM cores are becoming more conventional. The same thing is there with regards to floating point. There have been lots of various ways that it was done.
The article in question is not quite published yet:
Jack, R. E., Blais, C., Scheepers, C., Schyns, P. G., & Caldara, R. (in press) Cultural Confusions Show Facial Expressions are Not Universal Current Biology
Here is an earlier one using the same methodologies (PDF):
It is about where western and eastern people look at faces using eye tracking when for example learning or recognizing a face. There were some subtle differences.
For that user using that profile for that browser. Now consider a typical home computer with 2 or three users each with Firefox and IE or Firefox and Safari. Oh and guess where it stores that you do not wish to accept flash cookies?
Gnash is the solution, just rm -rf the correct dir when you are finished.
IE is not DNS server. What is most likely happening is that with some registry entry a certain way and a certain set of patches, when IE gets a NXDOMAIN when doing a domain name lookup it then does a bing/google/yahoo search (depending on another registry entry for your preferred search engine). It used to show a page with a red X.
This is not DNS hijacking. If somehow Windows now had a caching DNS server that substituted a IP address that then redirected to a bing search or something of that sort, that would be DNS hijacking. This is IE the client trying to handle NXDOMAIN errors in a helpful way. Hopefully this is customizable like I expect. The only thing about this is that if the default is bing that increases exposure and ad revenue for Microsoft. It does not break the internet like DNS hijacking does.
And in fact Sony did make Z series laptops with VT incompatible C2D chips (I know since my cousin has one) plus in some cases they used chipsets that do not support VT (or at least not easily with clever SMI hacking). I bet another concern is that buggy hardware on some of the Z series is made to work with System Management Mode (SMM) code in the BIOS, stuff that was never tested with VT.
I think that the xbox360 d-pad is a bit mushy but other than that not bad. What is bad about it is its location. My thumb is at a nearly 45 degree angle, and so I tend to press NE when I want N or E unusually often. Also it is harder than it should for me to press W, SW, and S because of the location. My thumb needs to rock for those.
The most natural controller that mixed a d-pad and analog stick was for the N64. It was a good idea to split the controller like that so that if you held it one way it was an ideal D-pad (also cross shaped) and if you held it another way you had a natural analog stick (and trigger). The other good idea was to have 4 ports on the console. The failings of the N64 controller were the quality of analog stick and the size of the 4 C-buttons and their distance from the B and A buttons. This should have been the finest controller for fighting games but those stupid C-buttons were useless.
Slashdot Mirror
I xpdf on unix, Preview.app on OS X, and Sumatra on Windows. Weren't there some vulnerabilities in Foxit for Windows about two years ago?
That hides your friend list on your profile, but does not prevent facebook apps from seeing your friend list.
Sure you can, but now any app can see your hometown and friends. I see no way to turn those off.
Lately I started using the awsome bar, but I used to hate it. What got me is that I have my homedirs exported via NFS. That in itself was annoying for using firefox since I needed different profiles for different machines in order to be able to run more than one firefox at a time on different hosts. But after a while I had a nice system where I could use a few profiles on all my machines and have bookmarks and history follow me around. But then the awesome bar made it so that those sqlite dbfiles were being used for everything. sqlite and nfs simply did not play well together. I ended-up having to symlink my firefox profiles to copies on local disks. So for all that time that I did not use the awesome bar, I lost the easy way of using a few profiles on many different machines. That is what bugged me the most about awesome bar.
Actually virtually all Firefox extensions are js and DOM interacting with foo.xul. You can create C++ extensions as well though.
No x86 does 0-9 BCD only, 754r would need 0-999 BCD for any speed improvement compared to 11 adds with carry. Also I think that the BCD opcodes never got extended circa 80286 like the other ones did to allow other registers to be used.
Google is using anycast for their DNS servers. There are not just two machines at 8.8.8.8 and 8.8.4.4 as the sole DNS servers. You get a relatively close-by server. This is a tried and true technique for DNS. In fact there is a technical feature about the google approach that is neat. It is likely that google is using many of the same servers it is for search for the DNS servers as well. They are running the caching DNS at each facility, such that if one server at the facility gets a record, then any other DNS server at that facility uses that response. That is one cool way to limit the delays for someone else making a DNS request. I've not seen that mentioned much before, and that is neat. I wish slashdot comments about stories that are trying to be technical would have technical comments on them near the beginning, instead of rehashing of all this privacy stuff, for a third or fourth story.
Another approach that was mentioned a lot before is that after the DNS server provides a response, the server checks to see if time is running-out regarding the TTL. If it is and has not expired yet, it asks again and pretends that the TTL counter has begun again. This again is trying to limit a DNS delay for some poor schmuck.
Another technical detail I have not seen mentioned much is that google DNS servers are returning largely authoritive answers only, often in cases where other DNS servers do not. For example, look-up a private IPv4 such as 192.168.1.1 with google's servers and some others. Others typically return non-authoritive responses, say to RFC1918.private.net. There is a lot of subtly misconfigured software-out there, hopefully this will bring it to the fore front about dealing with non-authoritive answers more carefully.
As to regarding the performance of google DNS, from a few locations for me, seems very fast. Is faster (much) than AT&T, bit slower than comcast, bit slower than work, comparison with OpenDNS is in the noise. What is more important is that they treat all records correctly, so for example kx509 _kca._udp.REALM style SRV records are handled unlike the DNS servers from some ISPs which seem to think that DNS is only for A records.
Another interesting feature is that google DNS is playing tricks with case in DNS queries and replies as yet another stop-gap-measure against DNS cache poisoning attacks. That's clever, I believe it was proposed before, but bind folks presented some issues and left it at that.
That happened to me too. Fox was having issues here. It was more goof-up rather than mal-intent. One time it was an ad that set the flag but did not unset it at the end of the ad. So then nothing recorded for hours until they fixed it.
I avoid this by uninstalling everything from Adobe. I bet you it was a flash ad utilizing an acrobat (or whatever it is they call it now a days) vulnerability. I have no flash and I have no acrobat. There are plenty of open source alternatives to Acrobat (even for Windows) but generally xpdf and gnash under VirtualBox is good enough or makes me decide I don't care about that site.
There are lots of people named Linus, this is likely not who you think it is unless you find yourself disagreeing with a Mr. Upson regularly.
I don't understand. I typically have an account that is not a sudoer that I use for everything. Using allowusers in the sshd_config that is the only account that can ssh in. Then if I need to do something as root, I do 'login admin' (or slight variations depending on unix flavor), where admin is a sudoer and then I run what I need under sudo. Is that what you mean by two password wall? What is the security risk here? I know that there have been vulns with regards to sudo in the past, is that what you are referring to?
The company that made 68K a-like SoC handhelds. The one where I just built a cross compiler for on FreeBSD to write programs. The company that very good documentation for the time about how to program the PDA. There was a free and an open source emulator. You could buy Code Warrior to dev on if you did not feel like using a gcc+binutils cross compiler. The way to get data to and from your PDA was free for Mac and Windows, plus there were a bunch of open source tools to do the same. If you wanted to distribute your programs, you could just put them on a web or FTP site somewhere. You could sell the programs on CDs or whatever you felt like as well. The only thing that was anything remotely like this, was that there was a web site where you could reserve a 4 character creator code for your applications. As long as it was free, it was yours.
Long ago there was a replacement for AD, it was known as yp. It had to change it's name to NIS, now NIS+. With that and kerberized NFS, the whole Windows approach is just backwards. Where NFS does not apply rsync does. You can get Windows to authenticate from LDAP, even tie that in with kerberos. If you need Windows to see shared drives, on Solaris you can reexport home dirs as smbfs, and that hauls on Solaris 10 being an in kernel implementation with a big ZFS storage pool behind it.
Somebody mod this up.
They do, actually if I recall correctly in T2 they have one per core, where each consists of 3 cells each consisting of two n-well resistors (thermal noise essentially) connected to an amp which feeds into a VCO (voltage controlled oscillator) that gets sampled with regards to a different clock.
I'm very happy to hear about your success. I'm a big fan of FreeBSD as well.
Read the syncer man page, there are sysctls that you can tweak to slow down flushing the pagecache so your disk will not spin up as often. If your laptop is very stable you might not worry so much about a panic. The battery should give you enough time for a sync or halt in case of a power outage.
On a laptop, I find I hardly ever care about the stuff in /var after a reboot. Using md was clever.
Did you look into pdflush? In particular dirty_writeback_centisecs? lsof, iostat, vmstat, and strace in a process of elimination are useful as well. It is too bad linux (I am assuming you are on linux) does not have something like dtrace or at least fs_usage. Did you do the other stuff listed above? In particular you could have some stupid process that writes-out its config at something insane like once a second. There was some notorious Seagate firmware for a while too, but not on the small drives as far as I know. Good luck.
Look in cron and disable stuff you do not use, especially locate.
Do not use the optical drive and make sure you are quiescing it.
Turn-off access time modifications for the hard disk.
Turn off fsck on boot.
Turn off periodic SMART status checking (on some drives this spins it up).
Tune the time to idle the drives and the periodic disk flusher (you have basically UPS with a laptop anyway).
Turn off swap.
Use a light simple window manager such as fvwm2 instead of something like gnome where lots of files are being accessed all of the time and you have many procs/threads running and the neat effects burn the battery.
Find the docs to your graphics drivers and tweak the tunables to use as little power as possible (this will give you much more than you likely expect).
Turn off bluetooth and wireless when you are not using it.
Don't use any of the crazy sound daemons.
You probably don't need wake-on-magic-packet for a laptop, turn it off, it helps a lot for some NICs.
Do you use multicast for wireless, most likely not, read the docs and figure-out how to get your driver to ignore that, it can conserve more power on some cards.
With some of the older chips USB was very power hungry in sleep (if that's your case tweak what you can so that it does as little as possible, likely turning off the wake on keyboard and mouse since you shutting and opening the lid should handle that).
Install a flash blocker and/or ad blocker and use gnash where you can instead of the adobe version.
Man I wish I had seen that there had been a reply to my comment earlier. I do embedded/realtime at work. Most of it happens to be on PPC and 68K of various vintage. Then I do work on various ARM and x86 but fewer installations. So this was definitely not a case of x86 is different than everything else and esoteric things not used.
I almost always align things even on systems that do not trap, simply because it is mire efficient, portable, and the compiler does it automatically. What I was referring to with regards to ARM and the way compilers align for it is that the structures become aligned differently than PPC (and power), x86, 68K, SPARC, and MIPS (yes I have used all of those minimally 32-bit cpus for work at least once) in a way that is unexpected at first.
say you have this struct: { int8_t a; int16_t b; int16_t; }
How does it get layed-out on PPC and x86:
a0bbcc
How does it get laid-out in ARM:
a0bbcc00
That was the problem I was referring to.
The thing about ldr is that in my case only the kernel was at first non relocatable, I wanted to have a few tasks (porting 68K pSOS code) and just wanted them to be PIC and loaded where ever so I could have the same kind of scheme where I used to have where some installations would have different number and sorts of task running depending on configuration.
Then the DACR and friends stuff came up since again I wanted to have tasks like in pSOS but have some memory protection as well. It was such a mess I gave-up and I ended-up having one task sharing the kernel and I did my own round robin scheduling of code it different object files instead of separate tasks and a bit more in the ISRs than I would have liked to make it work.
Then when latter ARMs came out I was very glad that I had since all of the DACR stuff have changed. That largely does not happen on PPC or x86, at least not ever in an area that affected me.
I don't know if I would call ARM asm beautiful to look at. Possibly it's nicer to look at than x86, but by the time of the 386 there were practically no more examples left of instructions that were not orthogonal. Sure the byte code for some of those might have been four times as long as the original 8086 version that used AX only, but you hardly ever look at that.
There are lots of annoying little things with ARM. In the area of the instruction set itself, there is the whole issue of what happens when you load a 32-bit word that is unaligned. That depends on the mode you are in, but it usually does some crazy rotation, but that is actually useful to make use of. This also leads to strange alignment relative to everything else out there if you want the same code to be able to fill a structure say. Then there is the whole loading of constants using LDR which is not a real instruction, the assembler does what ever craziness it can manage to get a constant. If your code is PIC that basically means loading from memory, if it is noPIC then it might do crazy pc relative stuff. Oh and what about thumb mode, very non-orthogonal.
Then ARM has the whole issues of all the differences. When they have TLBs and caches they tend to be alien (DACR ASID and PID) though the newest ARM cores are becoming more conventional. The same thing is there with regards to floating point. There have been lots of various ways that it was done.
Here is the site of one of the co-authors:
http://www.psy.gla.ac.uk/staff/index.php?id=RJ002
The article in question is not quite published yet:
Jack, R. E., Blais, C., Scheepers, C., Schyns, P. G., & Caldara, R. (in press) Cultural Confusions Show Facial Expressions are Not Universal Current Biology
Here is an earlier one using the same methodologies (PDF):
http://www.psy.gla.ac.uk/docs/download.php?type=PUBLS&id=1404
It is about where western and eastern people look at faces using eye tracking when for example learning or recognizing a face. There were some subtle differences.
For that user using that profile for that browser. Now consider a typical home computer with 2 or three users each with Firefox and IE or Firefox and Safari. Oh and guess where it stores that you do not wish to accept flash cookies?
Gnash is the solution, just rm -rf the correct dir when you are finished.
IE is not DNS server. What is most likely happening is that with some registry entry a certain way and a certain set of patches, when IE gets a NXDOMAIN when doing a domain name lookup it then does a bing/google/yahoo search (depending on another registry entry for your preferred search engine). It used to show a page with a red X.
This is not DNS hijacking. If somehow Windows now had a caching DNS server that substituted a IP address that then redirected to a bing search or something of that sort, that would be DNS hijacking. This is IE the client trying to handle NXDOMAIN errors in a helpful way. Hopefully this is customizable like I expect. The only thing about this is that if the default is bing that increases exposure and ad revenue for Microsoft. It does not break the internet like DNS hijacking does.
And in fact Sony did make Z series laptops with VT incompatible C2D chips (I know since my cousin has one) plus in some cases they used chipsets that do not support VT (or at least not easily with clever SMI hacking). I bet another concern is that buggy hardware on some of the Z series is made to work with System Management Mode (SMM) code in the BIOS, stuff that was never tested with VT.
I think that the xbox360 d-pad is a bit mushy but other than that not bad. What is bad about it is its location. My thumb is at a nearly 45 degree angle, and so I tend to press NE when I want N or E unusually often. Also it is harder than it should for me to press W, SW, and S because of the location. My thumb needs to rock for those.
The most natural controller that mixed a d-pad and analog stick was for the N64. It was a good idea to split the controller like that so that if you held it one way it was an ideal D-pad (also cross shaped) and if you held it another way you had a natural analog stick (and trigger). The other good idea was to have 4 ports on the console. The failings of the N64 controller were the quality of analog stick and the size of the 4 C-buttons and their distance from the B and A buttons. This should have been the finest controller for fighting games but those stupid C-buttons were useless.