Oh, and while I'm ranting about the horribleness of Open Source security stuff, why is it that there is STILL no well-integrated filesystem crypto in any of the Open Source operating systems, including the security-oriented OpenBSD? No, loopback crypto kludges don't count at all.
I find that whenever I say "Why can't such and such be easily done?" I get blasted with "Because you can. you... and........ then..."
That's either Open Source software policy, or the personal nature of many of the open source keystones.
You hint at the main argument against Itanium in your argument for it.
During the main phase of processing, the GIMP and Photoshop probably wouldn't see a major increase in performance from a two-bundle Itanium over a two-system cluster, and definately a dual-processor Athlon or Pentium 4. There simply isn't a need for the low latency that having a multithreading(Dunno if that's the proper term.) CPU provides.
Physics and weather simulations are excellent candidates for multithreading processors, though, since things like gravitational fields affect all regions of the simulation, meaning communication is necessary between that segment and other segments. (Assuming thread division by spatial relation.)
...in particular chips that may require different manufacturing processes.
Or at least portions of more complex circuits where part of the circuit may not warrant the added cost of SOI, 90nm, or strained silicon.
But then, those divisions are already made. AMD, for one, is working on recombining those parts. As an example, consider AMD's putting the memory controller on the CPU die.
I am curious, however, as to whether you could have more than one silicon die in the same ceramic casing. This would let you combine different parts made using different techniques. The CPU and L1/L2 caches could be on a high-cost process, with the L3 cache and memory controller being built with cheaper processes.
Placing more emphasis on cost savings than on performance, you could build the L1 instruction cache with a slower process than the L1 data cache. Or you could leave all of L1 on the same process and split L2 into instruction and data caches, with different processes.
If you wanted to ramp up CPU speed without a major hit to your performance, you could reduce the feature size of the core and L1 caches, and put L2 on a slower, more reliable process. That way, you could ramp up core speeds without having as much worry about yield loss from cache failure.
Helluva way to increase yield, and it gives the designer a LOT of options.
I'm concerned about new applications in surveillance.
While I'm sure remote gap-dropping is a long ways away, what if someone were able to place a sensing device on top of the gap between the chips? Who needs to decrypt someone's hard drive if they can just log memory transfers?
It's not like similar techniques aren't already in place. For example, I'm certain I heard about a case where someone put a keylogger on a mobster's computer. I know someone's used keyloggers at my college campus.
That "gay cooling duct module" provides dedicated space for whatever cooling apparatus you may want. That means air or water cooling, and I'm sure someone will think to put a noise supressor in there, with technology akin to noise-canceling headphones.
And there's lots of proffesional equipment that'll go in PCI-X slots. Just open a NASA Tech-Briefs magazine, and you'll see all sorts of stuff whose data wouldn't fit across AGP 8x.
Hell, I'm still happy on my 750MHz Duron w/ 768MB of PC133 RAM.
Runs Linux just fine.
And I don't even need a swap partition.:P (That's a type of pagefile, for the technicly declined.)
I might need one in the future, though. Having lots of memory is going to really bite it when it comes to "software mode suspend" which is Linux's answer to S4 suspension. (Coming in kernel 2.6)
If what's at question is the copyrightable nature of functional structure, then we're free to draw from as many different fields as we'd like to prove prior art. Or, at least, that the subject is irrelevant.
As a very basic example, every english student is taught pretty much the same way to write an essay. Does that mean that whoever wrote the first essay can now file a lawsuit against all students across the world and history?
From a more recent perspective: Cars have four wheels, a power source, and a passenger compartment. Does that mean the inventor of the first "horseless carriage" can file a lawsuit against everyone one supplies a product satisfying those requirements?
From the doomsayer's department: SCO, here we go...again
This strikes of the old guild systems. If you left your guild (say, a carpenter's guild) and joined another (say, a thatcher's guild), you were damned by both the guild you left and the guild you joined.
...for a community-wide class action lawsuit against SCO.
Talk about a hare-brained law.
on
Slackware Turns 10
·
· Score: 0, Offtopic
21 computer, enables 3rd parties to store data on that com- 22 puter, or use that computer to search other computers' 23 contents over the Internet.''.
They're making search engines of any sort illegal. Think about it.
I've never seen people on Slashdot, as a group, claim to be open-minded.
The administrators are known to be heavy-handed responding to things they don't like. This gets keeps the dissenting extremists off the system. It also causes an all-around discomfort for people who have something to say, and a general distaste for the system for many(most? all?) other people on the system.
Moderators are often quick to moderate "troll" to anything that might spur disagreement. This prevents flamewars. It also prevents a great deal of insightful discussion, and discourages newbies from participating.
A couple months ago, I lost my moderator priviledges. I've never said anything intentionally inflammatory before on Slashdot, so I don't know for certain what specifically caused the priviledges to be revoked.
Recently, I subscribed to Slashdot. I block all ads, and view 10 ad-free pages per day. I had all the plums at first, but now some of them are gone. For example, I can no longer see articles from "The Mysterious Future," and read Slashdot every fifteen minutes to twenty minutes, for most of the day.
I'm not expecting this account to escape unscathed from this rant, but I don't care. I've been slowly maturing beyond the community fascism for a while, and I guess it's time to move on.
Slashdot Mirror
Dupes are bad enough. Now you want to incite panic?
Oh, and while I'm ranting about the horribleness of Open Source security stuff, why is it that there is STILL no well-integrated filesystem crypto in any of the Open Source operating systems, including the security-oriented OpenBSD? No, loopback crypto kludges don't count at all.
... and .... .... then ..."
I find that whenever I say "Why can't such and such be easily done?" I get blasted with "Because you can. you
That's either Open Source software policy, or the personal nature of many of the open source keystones.
You hint at the main argument against Itanium in your argument for it.
During the main phase of processing, the GIMP and Photoshop probably wouldn't see a major increase in performance from a two-bundle Itanium over a two-system cluster, and definately a dual-processor Athlon or Pentium 4. There simply isn't a need for the low latency that having a multithreading(Dunno if that's the proper term.) CPU provides.
Physics and weather simulations are excellent candidates for multithreading processors, though, since things like gravitational fields affect all regions of the simulation, meaning communication is necessary between that segment and other segments. (Assuming thread division by spatial relation.)
...in particular chips that may require different manufacturing processes.
Or at least portions of more complex circuits where part of the circuit may not warrant the added cost of SOI, 90nm, or strained silicon.
But then, those divisions are already made. AMD, for one, is working on recombining those parts. As an example, consider AMD's putting the memory controller on the CPU die.
I am curious, however, as to whether you could have more than one silicon die in the same ceramic casing. This would let you combine different parts made using different techniques. The CPU and L1/L2 caches could be on a high-cost process, with the L3 cache and memory controller being built with cheaper processes.
Placing more emphasis on cost savings than on performance, you could build the L1 instruction cache with a slower process than the L1 data cache. Or you could leave all of L1 on the same process and split L2 into instruction and data caches, with different processes.
If you wanted to ramp up CPU speed without a major hit to your performance, you could reduce the feature size of the core and L1 caches, and put L2 on a slower, more reliable process. That way, you could ramp up core speeds without having as much worry about yield loss from cache failure.
Helluva way to increase yield, and it gives the designer a LOT of options.
I'm concerned about new applications in surveillance.
While I'm sure remote gap-dropping is a long ways away, what if someone were able to place a sensing device on top of the gap between the chips? Who needs to decrypt someone's hard drive if they can just log memory transfers?
It's not like similar techniques aren't already in place. For example, I'm certain I heard about a case where someone put a keylogger on a mobster's computer. I know someone's used keyloggers at my college campus.
Why? So I can add it to my resume, of course. 20 years old, and I know the C64 inside and out. :)
That "gay cooling duct module" provides dedicated space for whatever cooling apparatus you may want. That means air or water cooling, and I'm sure someone will think to put a noise supressor in there, with technology akin to noise-canceling headphones.
And there's lots of proffesional equipment that'll go in PCI-X slots. Just open a NASA Tech-Briefs magazine, and you'll see all sorts of stuff whose data wouldn't fit across AGP 8x.
There's lots and lots of USB serial and parallel adapters out there. Supported by both Windows and Linux. (And, I'm sure, FreeBSD)
Hell, I'm still happy on my 750MHz Duron w/ 768MB of PC133 RAM.
:P (That's a type of pagefile, for the technicly declined.)
Runs Linux just fine.
And I don't even need a swap partition.
I might need one in the future, though. Having lots of memory is going to really bite it when it comes to "software mode suspend" which is Linux's answer to S4 suspension. (Coming in kernel 2.6)
If what's at question is the copyrightable nature of functional structure, then we're free to draw from as many different fields as we'd like to prove prior art. Or, at least, that the subject is irrelevant.
As a very basic example, every english student is taught pretty much the same way to write an essay. Does that mean that whoever wrote the first essay can now file a lawsuit against all students across the world and history?
From a more recent perspective: Cars have four wheels, a power source, and a passenger compartment. Does that mean the inventor of the first "horseless carriage" can file a lawsuit against everyone one supplies a product satisfying those requirements?
From the doomsayer's department: SCO, here we go...again
cat /vmlinuz > /dev/dsp
:P
The sound of Linux.
Both debian and gentoo (and Red Hat) have security mailing lists that list packages/ebuilds that have been updated for security reasons.
I haven't used Gentoo, but Debian offers not only that, but apt-get.
Just add this script to your root crontab:
apt-get update || mail root -s "apt-get update error"
apt-get install --assume-yes || mail root -s "apt-get install error"
I'm not 100% certain that's correct...would someone else look that over for me?
This strikes of the old guild systems. If you left your guild (say, a carpenter's guild) and joined another (say, a thatcher's guild), you were damned by both the guild you left and the guild you joined.
Sounds like a regression in civilization, to me.
There's no difference.
I'd have to point out a counterexample to Netscape: Internet Explorer.
Comes bundled with every version of Windows, and you don't have to download it.
I doubt it. They're using DirectX, after all.
How is objectivity of this study any different from, say, a study by Microsoft promoting Windows?
99% of browsers, or 99% of market share? There's a difference...
Any civilized thought requires many many brain cells. So they don't qualify...
With a Republican government? You must be mad.
Come to think of it, it wouldn't stand much chance with a government run by Democrats, either.
...for a community-wide class action lawsuit against SCO.
21 computer, enables 3rd parties to store data on that com-
22 puter, or use that computer to search other computers'
23 contents over the Internet.''.
They're making search engines of any sort illegal. Think about it.
I've never seen people on Slashdot, as a group, claim to be open-minded.
The administrators are known to be heavy-handed responding to things they don't like. This gets keeps the dissenting extremists off the system. It also causes an all-around discomfort for people who have something to say, and a general distaste for the system for many(most? all?) other people on the system.
Moderators are often quick to moderate "troll" to anything that might spur disagreement. This prevents flamewars. It also prevents a great deal of insightful discussion, and discourages newbies from participating.
A couple months ago, I lost my moderator priviledges. I've never said anything intentionally inflammatory before on Slashdot, so I don't know for certain what specifically caused the priviledges to be revoked.
Recently, I subscribed to Slashdot. I block all ads, and view 10 ad-free pages per day. I had all the plums at first, but now some of them are gone. For example, I can no longer see articles from "The Mysterious Future," and read Slashdot every fifteen minutes to twenty minutes, for most of the day.
I'm not expecting this account to escape unscathed from this rant, but I don't care. I've been slowly maturing beyond the community fascism for a while, and I guess it's time to move on.
It sounds to me like it'd be more useful for truly long-distance connections, like from here to Pluto.
I thought it was called OSDN. Or IBM. Or Red Hat, Suse, and Mandrake. Oh, and don't forget Paypal!