Offloading rasterization has great benefits though. For every postscript printer there seems to be at least one file that causes it to reboot or (worse) get stuck. Such problems are much easier to fix when the postscript interpreter runs on the PC.
cyanogenmod claims to have 745,259 users currently. I am not sure if it is completely free or not but clearly more than 10000 users are using unofficial and modified version.
Hmm, can you give a link to the source code? Afaik we don't have source code to the sun java plugin. Openjdk's java plugin is a completely different project and does not work on many sites that are only tested with sun java.
Says who?;-)
http://www.gnu.org/licenses/license-list.html
specifically says that apache2 has some requirements that are not in gpl2:
"Please note that this license is not compatible with GPL version 2, because it has some requirements that are not in the older version. These include certain patent termination and indemnification provisions."
You can adjust the duration of the password caching in/etc/sudoers. The default is fifteen minutes -- I prefer a duration of two or three minutes, which is long enough to two a couple of commands in a row. If I actually need more than that, then I'd use sudo -i, but I rarely find that useful.
Unfortunately that does not really improve security in typical usage patterns. If you leave your terminal unlocked (or there's a bug in your web browser / pdf reader / etc. that allows arbitrary code execution) it is easy to escalate to root by doing echo alias sudo="/usr/bin/sudo/tmp/rootkit" >>.bashrc and patiently waiting for the user to use sudo for the next time.
My thoughts exactly, I don't want to use non-free adobe flash and slideshare does not work with gnash. Fortunately a very similar looking PDF does seem to be available at
I admit that I don't know much about this either but currently at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=608428 we are wondering if it is possible to support mount paths longer than 80 characters with breaking the freebsd kernel ABI.
http://iki.fi/lindi/usb/usbsnoop.txt has my step-by-step instructions for getting started with usb reverse engineering. It uses a perl script that reads a traffic dump and generates a C program that reproduces the traffic by calling libusb appropriately.
Of course it is a challenge to educate users but at least this gives you the possibility to access a trusted path if you care? I don't use Windows but as I understood it this is one of the features that gnu/linux systems really should support, at least as an option.
Doesn't "hardware-related" also forbid you from selling preinstalled computers that can decode MP3? Or what do you exactly mean? Openmoko has no hardware acceleration for MP3 decoding.
"A shipment to a German distributor was accidently mislabel with "MP3" (The Openmoko can't play MP3's unless modified). That was enough for the German customs to seize it according to European piracy protection laws."
-- http://info.openmoko.se/Home/mp3
"The Linux-based phone, OpenMoko is currently in a patent dispute with Sisvel, as known as Società Italiana per lo Sviluppo dell'Elettronica."
Syscall "emulation" is a very interesting idea since it will spare us from the archive rebuild and will let the user to just reboot to switch kernels without having to touch userland.
However, last time I tried this I couldn't get gnu screen run. I guess some terminal magic is not implemented in the syscall emulation properly? Can you run gnu screen?
My desktop uses nfs as its root filesystem so it is easy to measure how much data it will need to read on boot by measuring network traffic. A complete reboot with "shutdown -r now" generated only 44 megabytes of traffic (including both read/written data and ethernet overhead) so there is clearly no need to read a GB. The system runs debian gnu/linux 3.0 with linux 2.6.18-4-486.
for i in/usr/lib/python*; do if [ -e $i/trace.py ]; then python $i/trace.py --trace --ignore-dir $i "$@" exit 0; fi; done
While these usually produce a lot of output it is usually easier to grep them than to start adding debug prints to the scripts (which may even be in read-only/usr).
FDL 1.3 also adds proxy support: "now licensors can choose a proxy who is allowed to decide whether or not a work can be licensed under the terms of future versions of the FDL." -- http://www.gnu.org/licenses/fdl-1.3-faq.html
procedure proc is type distance_feet is new float; type distance_meters is new float; x : distance_meters:= 2.0; y : distance_feet; begin y:= x; end proc;
causes GCC to print
prog.adb:18:145: expected type "distance_feet" defined at line 18 prog.adb:18:145: found type "distance_meters" defined at line 18
I use the "zhone" package to make and receive calls (the same app has also a primitive address book). However, you are correct in that there is probably not much point in using the phone if you have already decided that you don't plan to do any development.
Slashdot Mirror
Offloading rasterization has great benefits though. For every postscript printer there seems to be at least one file that causes it to reboot or (worse) get stuck. Such problems are much easier to fix when the postscript interpreter runs on the PC.
cyanogenmod claims to have 745,259 users currently. I am not sure if it is completely free or not but clearly more than 10000 users are using unofficial and modified version.
Hmm, can you give a link to the source code? Afaik we don't have source code to the sun java plugin. Openjdk's java plugin is a completely different project and does not work on many sites that are only tested with sun java.
Indeed. Using rtklib you can get 10 cm accuracy even with two cheap openmoko phones (~200 EUR) and cheap antennas (~20 EUR): http://lindi.iki.fi/lindi/finhack/finhack2010-rtklib-lindfors.pdf -- you can read more about it on foss-gps mailing list.
Openmoko seems to be missing from the list. According to http://people.openmoko.org/openmoko/certificate/gta/gta02/certificate/CE/EA832514_R01_CE%20SAR_FIC_GTA02.pdf the SAR is 1.05 W/kg for GSM.
Says who? ;-)
http://www.gnu.org/licenses/license-list.html
specifically says that apache2 has some requirements that are not in gpl2:
"Please note that this license is not compatible with GPL version 2, because it has some requirements that are not in the older version. These include certain patent termination and indemnification provisions."
You can adjust the duration of the password caching in /etc/sudoers. The default is fifteen minutes -- I prefer a duration of two or three minutes, which is long enough to two a couple of commands in a row. If I actually need more than that, then I'd use sudo -i, but I rarely find that useful.
Unfortunately that does not really improve security in typical usage patterns. If you leave your terminal unlocked (or there's a bug in your web browser / pdf reader / etc. that allows arbitrary code execution) it is easy to escalate to root by doing echo alias sudo="/usr/bin/sudo /tmp/rootkit" >> .bashrc and patiently waiting for the user to use sudo for the next time.
My thoughts exactly, I don't want to use non-free adobe flash and slideshare does not work with gnash. Fortunately a very similar looking PDF does seem to be available at
http://elinux.org/images/f/f7/RightApproachMinimalBootTimes.pdf
I admit that I don't know much about this either but currently at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=608428 we are wondering if it is possible to support mount paths longer than 80 characters with breaking the freebsd kernel ABI.
That's kind of the point. Distributions make installing software easier since you don't need to figure it out yourself.
(If you look at sqlninja-0.2.5.tgz you see you can just run the perl script with "./sqlninja".)
Doesn't work in this case since they don't ship a makefile.
http://iki.fi/lindi/usb/usbsnoop.txt has my step-by-step instructions for getting started with usb reverse engineering. It uses a perl script that reads a traffic dump and generates a C program that reproduces the traffic by calling libusb appropriately.
Of course it is a challenge to educate users but at least this gives you the possibility to access a trusted path if you care? I don't use Windows but as I understood it this is one of the features that gnu/linux systems really should support, at least as an option.
Doesn't "hardware-related" also forbid you from selling preinstalled computers that can decode MP3? Or what do you exactly mean? Openmoko has no hardware acceleration for MP3 decoding.
"A shipment to a German distributor was accidently mislabel with "MP3" (The Openmoko can't play MP3's unless modified). That was enough for the German customs to seize it according to European piracy protection laws."
-- http://info.openmoko.se/Home/mp3
"The Linux-based phone, OpenMoko is currently in a patent dispute with Sisvel, as known as Società Italiana per lo Sviluppo dell'Elettronica."
-- http://www.infoq.com/news/2008/11/OpenMoko-MPEG
Enabling all codecs by default is not a good idea. Think security vulnerabilities in ffmpeg:
http://www.debian.org/security/2010/dsa-2000
http://www.debian.org/security/2009/dsa-1781
That isn't enough. In bash you can use DEBUG traps to override any command:
lindi@sauna:~$ function f() {
> echo executing evil sudo...
> }
lindi@sauna:~$ trap f DEBUG
lindi@sauna:~$ sudo iptraf
executing evil sudo...
[sudo] password for lindi:
Syscall "emulation" is a very interesting idea since it will spare us from the archive rebuild and will let the user to just reboot to switch kernels without having to touch userland.
However, last time I tried this I couldn't get gnu screen run. I guess some terminal magic is not implemented in the syscall emulation properly? Can you run gnu screen?
My desktop uses nfs as its root filesystem so it is easy to measure how much data it will need to read on boot by measuring network traffic. A complete reboot with "shutdown -r now" generated only 44 megabytes of traffic (including both read/written data and ethernet overhead) so there is clearly no need to read a GB. The system runs debian gnu/linux 3.0 with linux 2.6.18-4-486.
"bash -x foo.sh" produces nice execution trace of a shell script. To debug perl and python I thus have wrapper scripts "perl-x":
/usr/lib/python*; do if [ -e $i/trace.py ]; then python $i/trace.py --trace --ignore-dir $i "$@" exit 0; fi; done
/usr).
PERLDB_OPTS="NonStop=1 AutoTrace=1 frame=2" exec perl -dS $@
and "python-x":
for i in
While these usually produce a lot of output it is usually easier to grep them than to start adding debug prints to the scripts (which may even be in read-only
FDL 1.3 also adds proxy support: "now licensors can choose a proxy who is allowed to decide whether or not a work can be licensed under the terms of future versions of the FDL." -- http://www.gnu.org/licenses/fdl-1.3-faq.html
procedure proc := 2.0; := x;
is
type distance_feet is new float;
type distance_meters is new float;
x : distance_meters
y : distance_feet;
begin
y
end proc;
causes GCC to print
prog.adb:18:145: expected type "distance_feet" defined at line 18
prog.adb:18:145: found type "distance_meters" defined at line 18
Does wifi in N810 still require a proprietary driver to operate (something with "HAL" in its name if I recall correctly)?
I use the "zhone" package to make and receive calls (the same app has also a primitive address book). However, you are correct in that there is probably not much point in using the phone if you have already decided that you don't plan to do any development.
It has been attempted. See "Silver Needle in the Skype" presentation at http://www.blackhat.com/presentations/bh-europe-06/bh-eu-06-biondi/bh-eu-06-biondi-up.pdf -- The impression I got was that it was deliberately made difficult to understand by adding all sorts of checksums and encryption layers.