EVMS ( http://evms.sourceforge.net/ ) could do some of these things and shared the same vision as ZFS: "EVMS provide[d] a single, unified system for handling all of your storage management tasks." The memory is a bit fuzzy but I think it was a port/reimplementation from AIX or else IBM was involved. I think it died because it didn't integrate well into the kernel (like ZFS it smashed many layers into one).
Wasn't HP the original motivating force behing getting nVidia to write the binary-only drivers? How did they do that? They must have some sway with nVidia. What would they recommend to persuade nVidia to open srouce their drivers? Would $10,000 help nVidia to do so? If yes then give the money to nVidia and not Nouveau.
In my experience the added lag of 70ms may make the game unplayable. My connection latency (on DSL) is about 100ms. Anything above 150ms is too much lag for me.
On the bright side, it looks like the grid nodes are wired to be able to talk to any other node in two hops (a -> whiteboard -> b) so that latency isn't proportional to the number of nodes.
I run Jpilot (http://www.jpilot.org/) all the time. It can open popup windows or execute arbitrary commands for events. Plus, it syncs with your (Palm) PDA.
I agree that because they're female they can't hack it in CS. My question is: why? My belief is that the reason is because we've made it so they can't. So they don't, and thus humankind misses out on the benefit they would provide.
We computer geeks like to think that we are enlightened, that there
is no systematic bias against women in computer science. We
are wrong. What's worse is that we cannot admit that we are
wrong. Suddenly, we become unscientific in our analysis of the
situation, using in our personal anecdotal evidence to serve as proof
against the claim. We think "I never did anything to prevent a women
from pursuing a career in computer science." My brothers, we are
wrong. Here is a summary of the evidence for this claim:
http://norfolk.cs.washington.edu/htbin-post/unrest ricted/colloq/details.cgi?id=312
Computer science isn't advancing as well as it could due to the
fact that intelligent women who would otherwise contribute are being
repelled due to male influence. It's humbling to admit that we aren't
as enlightened as we think we are, but we need to. Only then can we
learn with an open mind what we need to do--suppressing our pride and
other misplaced feelings of righteousness and instead focusing on the
data--to rectify the situation.
Re:Reminds me of 'thrashing' the Atari 2600
on
Glitch Art
·
· Score: 2, Interesting
I recall there was some way, using the techniques you describe, where one could get Space Invaders to give the player a double shot.
Googling a bit I see the technique is simply to turn on the system while holding reset. We didn't figure that out.
Is mod_perl 2.0 ready for prime time yet? Last time I checked--a few months ago--the core was there but the mp 1.x emulation didn't work very well and some important modules, e.g., Apache::AuthCookie weren't ported yet. I went back to 1.x.
It took me the entire weekend to install the tape, mostly because my system runs GNU/Linux and I couldn't figure out how to recompile the kernel. The questions I posted on lkml about what brands of black electrical tape are supported in the kernel went unanswered. I ended up booting into Windows and using this wonderful shareware utility I found at download.com.
The problem only reproduces when you're not running under the debugger.
Boy, that sounded awkward. The problem does not reproduce. The problem is reproducible. You reproduce the problem. A bannana is edible. You eat a bannana. The bannana eats? No.
After a few iterations of the tester saying, "Sorry, but the bug's not fixed yet," and the developer saying, "What are you talking about? I don't see the problem!"
Why didn't the developer try running Word without the debugger? Didn't the developer know the debugger changed the OS's open file limit?
Actually, for me it the "configure" output, especially the part at the end when you see that all the cool features have been enabled thanks to your simply adding a few tokens to the USE flag.
Solution: use Redhat's hack to put the gconf lock in/tmp. Result: race condition. Whoever writes to the config preference file last wins. Yay. User logs into computer A. Logs into another--B--, makes config changes on B, logs out of B, logs out of A, and then the changes are gone. Thanks.
Solution two: run the gconfds on the NFS server, set "ORBIIOPIPv4=1" in/etc/orbitrc for all clients. Seems to work OK. Hope that doesn't FUBAR security. Wait, the NFS server runs Redhat 7.3 (and thus gconf-1). What about those new workstations that run RHEL3 (gconf-2)? They're fucked. "Just upgrade gconf-1 to gconf-2 on the file server," you say. After all, the GNOME developers say that gconf-2 handles gconf-1 clients transparently. (Interlude: descend into dependency hell to get gconf-2 installed--in a maintainable way--on a Redhat 7.3 box.) Oh, wait, gconf-2 doesn't actually handle gconf-1 clients properly. Thanks.
Review stations attached to CT and MRI? Do you mean PACS workstations based on Centricity or the consoles that control the modality? The PACS workstations are MS Windows-base. We used to have Solaris-based review stations ("Advantage Windows") but those are being phased out. The PACS backend is till Solaris-based. In fact the DICOM server middle-end Macs are being replaced by Sun blades.
The CT Sim software was not developed by GE. They bought it.
A(lan) Cox may be slow responding to email due to the fact that he is busy getting his MBA. I think the term is near completion, so he may be writing papers and whatnot. In the mean time you may wish to try this processor.
It would make no difference, I just threw in the mention of Linux to get some Karma:). OpenBSD would be my choice. The poster mentioned Solaris, which I would shy away from. No need to have scalable performance for something as simple as this.
Ignoring the fact that DNS wasn't designed to handle this (setting your ttl to a low time (e.g., 5min) generates a good amount of useless traffic when your site is up), here is how you might do it:
First, you need to have a monitoring system on the Internet somewhere, not through your T1 because if that goes down it won't be able to update your DNS. You have that already, I'm sure, to test your web site accessibility from the Internet. Of course, at least one of your name servers must be accessible when the T1 goes down too, so that will have to be somewhere (other than on your T1) on the Internet as well.
On this name server enable dynamic updates. Modify your monitor system that checks availability of your site to use Net::DNS to update the IP address of your web server when the monitor fails.
Going all open source, I'd use Net::DNS and nagios for the monitoring software, bind for the name server (which supports dynamic updates), with Linux as the OS.
Slashdot Mirror
EVMS ( http://evms.sourceforge.net/ ) could do some of these things and shared the same vision as ZFS: "EVMS provide[d] a single, unified system for handling all of your storage management tasks." The memory is a bit fuzzy but I think it was a port/reimplementation from AIX or else IBM was involved. I think it died because it didn't integrate well into the kernel (like ZFS it smashed many layers into one).
http://www.quake3world.com/files/#music
For MS you can apply the patch to all systems. For Solaris before 8 the cost is $400 _per server_ or _$150K_ for more than 375 servers. See
a ris_dst_addendum.jsp
http://www.sun.com/bigadmin/features/techtips/sol
Wasn't HP the original motivating force behing getting nVidia to write the binary-only drivers? How did they do that? They must have some sway with nVidia. What would they recommend to persuade nVidia to open srouce their drivers? Would $10,000 help nVidia to do so? If yes then give the money to nVidia and not Nouveau.
See http://www.ee.washington.edu/research/seal/publica tions.html for publication with numbers.
In particular,
http://www.ee.washington.edu/research/seal/pubfile s/73.pdf
http://www.ee.washington.edu/research/seal/pubfile s/Krichtafovitch_ESA_June_05.pdf
http://www.ee.washington.edu/research/seal/pubfile s/MSEE_Nels.pdf
http://www.smallbusinesscomputing.com/webmaster/ar ticle.php/3498116
It gives pointers to various offerings, including one-time passwords, hardware tokens, smart cards, and biometrics.
In my experience the added lag of 70ms may make the game
unplayable. My connection latency (on DSL) is about 100ms. Anything
above 150ms is too much lag for me.
On the bright side, it looks like the grid nodes are wired to be able
to talk to any other node in two hops (a -> whiteboard -> b) so that
latency isn't proportional to the number of nodes.
I run Jpilot (http://www.jpilot.org/) all the time. It can open popup windows or execute arbitrary commands for events. Plus, it syncs with your (Palm) PDA.
If Windows supports named pipes you can do something like this: http://www.unix.org.ua/orelly/perl/cookbook/ch16_2 3.htm where instead of calling fortune you open a random wave file.
I agree that because they're female they can't hack it in CS. My
question is: why? My belief is that the reason is because we've made
it so they can't. So they don't, and thus humankind misses out on the
benefit they would provide.
We computer geeks like to think that we are enlightened, that there is no systematic bias against women in computer science. We are wrong. What's worse is that we cannot admit that we are wrong. Suddenly, we become unscientific in our analysis of the situation, using in our personal anecdotal evidence to serve as proof against the claim. We think "I never did anything to prevent a women from pursuing a career in computer science." My brothers, we are wrong. Here is a summary of the evidence for this claim: http://norfolk.cs.washington.edu/htbin-post/unrest ricted/colloq/details.cgi?id=312
Computer science isn't advancing as well as it could due to the fact that intelligent women who would otherwise contribute are being repelled due to male influence. It's humbling to admit that we aren't as enlightened as we think we are, but we need to. Only then can we learn with an open mind what we need to do--suppressing our pride and other misplaced feelings of righteousness and instead focusing on the data--to rectify the situation.
I recall there was some way, using the techniques you describe, where one could get Space Invaders to give the player a double shot.
Googling a bit I see the technique is simply to turn on the system while holding reset. We didn't figure that out.
Is mod_perl 2.0 ready for prime time yet? Last time I checked--a few months ago--the core was there but the mp 1.x emulation didn't work very well and some important modules, e.g., Apache::AuthCookie weren't ported yet. I went back to 1.x.
In case one is unclear on this difficult concept, I have a picture of my implementation:
photo
It took me the entire weekend to install the tape, mostly because my system runs GNU/Linux and I couldn't figure out how to recompile the kernel. The questions I posted on lkml about what brands of black electrical tape are supported in the kernel went unanswered. I ended up booting into Windows and using this wonderful shareware utility I found at download.com.
Boy, that sounded awkward. The problem does not reproduce. The problem is reproducible. You reproduce the problem. A bannana is edible. You eat a bannana. The bannana eats? No.
Why didn't the developer try running Word without the debugger? Didn't the developer know the debugger changed the OS's open file limit?
Actually, for me it the "configure" output, especially the part at the end when you see that all the cool features have been enabled thanks to your simply adding a few tokens to the USE flag.
The Soul of a New Machine by Tracy Kidder.
I'm eating pizza right now; otherwise I'd say more.
Gconf + NFS = less hair.
/tmp. Result: race condition. Whoever writes to the config preference file last wins. Yay. User logs into computer A. Logs into another--B--, makes config changes on B, logs out of B, logs out of A, and then the changes are gone. Thanks.
/etc/orbitrc for all clients. Seems to work OK. Hope that doesn't FUBAR security. Wait, the NFS server runs Redhat 7.3 (and thus gconf-1). What about those new workstations that run RHEL3 (gconf-2)? They're fucked. "Just upgrade gconf-1 to gconf-2 on the file server," you say. After all, the GNOME developers say that gconf-2 handles gconf-1 clients transparently. (Interlude: descend into dependency hell to get gconf-2 installed--in a maintainable way--on a Redhat 7.3 box.) Oh, wait, gconf-2 doesn't actually handle gconf-1 clients properly. Thanks.
Solution: use Redhat's hack to put the gconf lock in
Solution two: run the gconfds on the NFS server, set "ORBIIOPIPv4=1" in
Review stations attached to CT and MRI? Do you mean PACS workstations based on Centricity or the consoles that control the modality? The PACS workstations are MS Windows-base. We used to have Solaris-based review stations ("Advantage Windows") but those are being phased out. The PACS backend is till Solaris-based. In fact the DICOM server middle-end Macs are being replaced by Sun blades.
The CT Sim software was not developed by GE. They bought it.
Sorry, no. GE's CT Sim software runs on Solaris. Their CT and MR scanners (the consoles) run Irix (and now Linux).
Now, if you think Windows is bad, have you ever examined the default install of Irix (say 6.5.12)? Jesus wept.
A(lan) Cox may be slow responding to email due to the fact that he is busy getting his MBA. I think the term is near completion, so he may be writing papers and whatnot. In the mean time you may wish to try this processor.
It would make no difference, I just threw in the mention of Linux to get some Karma :). OpenBSD would be my choice. The poster mentioned Solaris, which I would shy away from. No need to have scalable performance for something as simple as this.
Ignoring the fact that DNS wasn't designed to handle this (setting your ttl to a low time (e.g., 5min) generates a good amount of useless traffic when your site is up), here is how you might do it:
First, you need to have a monitoring system on the Internet somewhere, not through your T1 because if that goes down it won't be able to update your DNS. You have that already, I'm sure, to test your web site accessibility from the Internet. Of course, at least one of your name servers must be accessible when the T1 goes down too, so that will have to be somewhere (other than on your T1) on the Internet as well.
On this name server enable dynamic updates. Modify your monitor system that checks availability of your site to use Net::DNS to update the IP address of your web server when the monitor fails.
Going all open source, I'd use Net::DNS and nagios for the monitoring software, bind for the name server (which supports dynamic updates), with Linux as the OS.
I haven't slept for more then 24 hours but I didn't see a link to the large format trailer on the site so here it is:
Slashdot Me
Of course, I already grabbed it myself.
Damn, I can't spell. I should have used mozex to edit the textbox with emacs and ran a spell checker.