I've got OO 1.0.1 installed on FreeBSD and Windows XP (for my wife) and I don't know why anyone who knew anything about it would call it "Java based." I believe that java can be used in some sort of browser component burried deep inside, but from what I can see it would run perfectly well without a JRE at all.
The reverse engineering steps are (likely) the same for playback and fair-use backup. Again, I simply want the reverse engineering for the purpose of exercising fair use to remain legal -- and in particular to be treated differently than reverse engineering for the purpose of violating copyright.
I realize that that the actual steps taken are the same. The net result is that actual conduct is what should be judged. Did you really just make a personal backup for yourself? Or are you selling VideoCD copies on eBay?
Really, this is no different than the battle that was fought over encryption. Encryption can be used to keep the sysadmin from reading your private mail, and it can also be used by terrorists to coordinate their activities without fear of interception (of course, one man's terrorists are another man's freedom fighters). Guns can be used for home defense or home invasion. And on and on and on.
So this should really be the battle cry: "It's the conduct, stupid."
There is no need to force content providers to allow copying to be easy. It is sufficient to make (or keep, depending on your point of view) it legal to reverse-engineer such schemes for the purposes of exercising fair use rights. This is the part the DMCA got wrong - it does not make a distinction between reverse engineering CSS for the purpose of ripping a copy, DIVXing it and putting it up for P2P and reverse engineering CSS for the purpose of using XINE to play DVDs.
It would be rather complicated to manufacture TVs, radios, etc. if the RF bandwidth weren't standardized... 50 different tv tuners in one would be complicated today, and probably impractical around the time they added the UHF system...
It would be no more complex to allow TVs to tune to arbitrary frequencies than to make them cable-ready. Besides, even this problem can be solved by market forces. It is in the mutual interest of both the consumer electronics and broadcasting industries for there to be broad agreement on such matters. That hardly means we require government intervention to achieve it.
I thought of one other area where interstate commerce is involved in VHF - aeronautical mobile services. The emissions from an airplane at cruising altitude can cover vast territory and clearly impact interstate commerce. I believe it's an area for the FAA, but if we really need an FCC to govern interstate communications, then that is one more chunk of real estate I can't argue too much about.
Someone please remind me again why we need the FCC to regulate all of this spectrum?
Here's my point... It is, I suppose, appropriate for the FCC to regulate interstate or international communications. But RF not aimed at a satellite that is higher than 50 MHz (most of the time) is not going to leave the state (yes, there are exceptions, but neighboring states have been cooperating ever since there *were* states in border issues without any federal intervention).
The band plan in Minnesota has absolutely zero impact on me here in California. Heck, where I am neither does the bandplan in Oregon, Idaho, Nevada, Arizona or Mexico. Why is it that a bunch of beaurocrats in Washington DC should have complete and total say over issues involving strictly local transmission and reception of radio signals (we can thank FDR and his buddies, of course, for the question of why they do as opposed to why they should)?
Is there a tool that allows you to force the switch to forward ethernet frames so they can be sniffed without switch administrator access?
Look at how a switch works. When you transmit a frame, it associates your MAC address with the port you're on. Next time a frame is sent to that MAC address, it sends it only to that port, becuase it knows that address is going to be found only on that port. If the switch does not know where a particular MAC address is to be found, it must send the frame to every port. Therein lies the weakness.
Switches tend to have finite sized MAC tables. If you overload them, they throw away older data in a least-recently-used manner. So the way to turn a switch into a hub is to send a non-stop stream of frames with random Ethernet source addresses. Make them small frames so you don't chew up all of the bandwidth of the switch, and don't send them too frequently (unfortunately, how frequently is too frequently is dependent on how big the switches table is). Address them to a known non-existent MAC address and make them a known unused Ethernet protocol. You will flood the switch's table and it will be forced to broadcast all frames. QED.
I think you have it backwards - TV stations are leaving the upper UHF (channels 52-69) for the "Core TV Channels" 2-51.
None of the digital broadcasters around here are in VHF. IMHO they'd have a hell of a time doing that. Their digital broadcasts are all in UHF, and until they're done with analog broadcasting, they're going to be using those VHF slots for that. Once they've turned the analog VHF off, are they going to then move back down to VHF? Why?
The lower 700 MHz Auction begins tommorow!. This is for UHF channels 52-59.
I believe they're shoving high UHF translators downwards to make more room for cell phones. Look at the bandplan. It's set up as 3 channels (that is, 6 MHz blocks) that air paired, and 2 in the middle that are unpaired.
One of the things we have to look forward to once broadcast television has gone fully digital is the give-back of the VHF TV spectrum. IMHO, VHF TV was one of the biggest wastes of spectrum ever conceived. The 12 channels of TV take up more than a quarter of the VHF bandwidth. By contrast, the 56 channels of UHF take up about 12.5% of UHF.
So what?
In general, lower frequencies tend to suffer a bit less from multipath distortion, suffer less from feedline losses, are easier to engineer, and more efficient to generate.
Channels 2-6 are very low in frequency indeed. They start at 54 MHz (TV channels are 6 MHz wide), but there is a 4 MHz gap between 4 and 5 for various low power services (mostly RC cars and planes), with channel 6 abutting the bottom of the FM radio channels (88 MHz). Now, I think channels 5 and 6 should be dedicated to an amateur broadcasting service, and the rest perhaps to land-mobile activities, but channels 7-13 are the perfect place for low power data services.
Of course, it's going to be years before the VHF TV transmitters are finally turned off, but I do believe it will happen eventually, and if we don't plan well in advance, there will be a smoke-filled-room give-away of this prime spectrum to someone with a lot of money, which isn't necessarily in the best interests of everyone.
The assertion made is that digital sampling of the inputs on the piano, recording the digital samples and playing them back is different enough from a live performance to be inferrior. I will stipulate that the visual presence of the artist adds to a live performance, but set that aside for a moment.
It is a lot like the Turing test in some regard. If you blindfold an audience and play them the same song first by a live musician and second by recording the *same* live musician, and if the audience can't reliably pick out which is the recording and which is the musician, then I'd say the recording technology in question passes the test.
Legitimate arguments can still be made that the performer may change his performance to suit different circumstances, and such adaptations would not be possible with a recording. But those arguments lie outside the confines of whether or not the recording mechanism is sufficiently accurate to capture and reproduce a single performance by the artist.
But in the context of a competition, I believe it may be actually be a benefit to the objectivity of the judging. The judges won't know who is who, so no favoritism will be possible. The artists' inability to tailor their performances for the judges will be a hinderance borne equally by all of the contestants. But unlike an audio recording, no extraneous noise other than that made by the performer will be recorded.
I'm all for Microsoft adopting open source. It would be the best thing for everyone except Microsoft. Which is precisely why it will never happen.
Microsoft OSes have historically been a technical laughing stock (I am willing to stipulate for the purpose of this discussion that Windows 2000 was their first serious contender. I actually think it's the best piece of code they ever released. I put it more or less on a par with Mac OS X, but I digress). So if they were so bad, why did they succeed? Because they had monopolistic market control that they misused (remember: A federal court has actually found them guilty of this. There's no longer any reason for debate on this point).
How do they maintain their stranglehold? It's quite simple: Embrace, Enhance, Exclude. For those of you who have not had this tactic properly explained, it goes like this: Microsoft sees an open source protocol succeed, it then adds support for that protocol to one of its product suites, but in doing so makes a very slight modification the specification for which it does not release. Microsoft uses this modification as a compatibility barrier for other products.
This is why Microsoft is so frightened of the GPL, but is not threatened by non-GPL open source. They cannot take a GPLed program and perform EEE with it, because they must release the source to their modifications. It then becomes easy to insure compatibility with the Microsoft variant.
Of course, Microsoft turns this around and claims that it eliminates their "freedom to innovate." Nothing could be further from the truth. They are free to innovate within the GPL. They simply would be disallowed from doing so in a manner whose sole purpose is to screw anyone not using their software.
Without their playbook of anticompetitive behavior, Microsoft deep in its soul believes it cannot succeed. A Microsoft that would be willing to participate in any meaningful way in Open Source would be so different as to hardly be comparable to what exists today.
Two too many voltage conversion steps
on
Do-it-yourself UPS
·
· Score: 1
Surely someone out there makes an ATX power supply that takes 12 VDC as an input. You could take a rather expensive component (the AC inverter) out of the picture and increase the overall efficiency by a measurable amount.
Oh, and the greeniacs out there will be no doubt pointing out that there's no particular reason that the AC->DC step couldn't be replaced by a bunch of solar panels and a charge controller.:-)
Very few, but some significant changes
on
Fair IP Laws?
·
· Score: 2, Insightful
1. Eliminate software and business method patents.
2. Strictly limit the timespan of copyrights. Holywood has a habit of getting the expiration of copyrights pushed out whenever they come close to their expiration date. That sort of nonsense has got to stop. Copyrights should be granted for 50 years to the original creater, his estate or assignee. Period. That's all. Finito.
3. A media consumer's bill of rights that formally states that a consumer may do *ANYTHING* with a copyrighted work for which they have purchased a copy, *except* for distributing additional copies of same. Too often the debate has centered around what rights should be granted to copy owners (as opposed to copyRIGHT owners). BS. They should have *all* rights *except* for a short list.
4. Note that case number 3 talks about the simple sale of a copy. It does not preclude a seller from including restrictions in the purchasing contract on the use by the recipient. But such restrictions *must* be part of the contract agreed to by *both* parties *before* performance of the contract. Click-through or shrink-wrap licenses or other shenanigans that are tacked on *after* both parties have performed the simple exchange of money for a copy *explicitely* should not be allowed to be binding.
There may be reasons folks want to jam cell phones. Reducing RF almost certainly wouldn't be one of them. If you start interfering with it, the first thing it's going to do is turn the power up to try and talk over the interference. Congratulations: You have now made the cell phone transmit more RF, never mind the fact that your own jammer emits RF as well.
I've been complaining about this one for years now.
Outlook can be used in one of two configurations (that matter): "Internet only", which has IMAP and POP capabilities, and "Corporate", which has Exchange and POP capabilities. This Hobson's choice is an anti-competitive barrier they've put in place to keep people from using IMAP servers for e-mail.
One way to transition away from Exchange is to move the mail to an IMAP server and leave the contacts, calendar stuff, etc on the Exchange server...... Except, of course, for the fact that Microsoft has made that impossible.
Another way that Microsoft could fix this would be to allow Outlook to store non-email content as MIME attachments in IMAP servers. That is, a particular IMAP folder could be designated as a Contacts or Calendar folder and Outlook could store that type of information as MIME attachments to e-mail messages. But, of course, that would make it possible to get all of the functionality of Outlook without an Exchange server, which is something Microsoft Can Never Allow.
A very long time ago (Think: SunOS 4.x), you had to have more swap than RAM because the amount of virtual memory you had was EQUAL to the amount of swapspace you had. That is, every page of RAM had to be backed by a page of swap, or else it wouldn't end up being useful (I'm oversimplifying a bit).
Now, in order for FreeBSD to be willing to save a core image, you have to have a swap partition with more space than you have in RAM, otherwise savecore will refuse to set things up. But for FreeBSD, the amount of virtual memory you have is equal to the amount of RAM you have PLUS the amount of swap space you've got set up (again, there is some RAM that gets used to hold the kernel image, so this is a bit of a simplification). Given that, it is perfectly ok to run a machine without any swap at all, provided you have a sufficient amount of memory to do everything you want to do. But having swap is good because it gives you some cushion, plus if you want to save cores from panics, you must, as I said, configure a swap partition with at least as much space as you have RAM.
I dunno.... Looking at the pictures it looks like there's an awful lot of unshielded plastic. The reason PCs go into metal boxes is so that you don't wreck your next door neighbor's TV reception (or worse).
Two jobs ago, I worked for Enlighten Software, whose EnlightenDSM product was supposed to do all this with some bells and whistles. They went bankrupt trying.
The only way I see it happening is if some common configuration API gets some traction and then everyone agrees to write shims to get it to talk to their favorite OS or piece of middleware. I am not optimistic.:-)
They can't insure that a party hack isn't standing behind absentee voters watching them fill in their ballots from their kitchen tables. What's the difference?
All that said, I think the security questions are a lot more important. It's probably a lot easier for dead former Chicago residents to vote over the Internet than... Oh wait...
Just recently I had to go from 0 knowledge of RDBMS to a working web app. The elephant postgresql book by Bruce Momjian (ISBN: 0201703319) was just about perfect. It explained from reasonably basic concepts using lots of examples, but wasn't really hugely wordy. It went from the basics through some fairly advanced usage, but left out the really arcane stuff (but did reference where you could find that if you wanted it).
More important than anything else, IMHO, was that it has a really, really good index.
Speaking at least for myself, that book should be used as a model for other books of its class.
First, like so many others in this thread, let me say that the force is strong in this one.
From what I can figure from the pictures, it does ejection the same way that it does picking -- counting on the little lip on the bottom of the cartridge to stick out far enough that the matching hump on the end of the spatula can grab it.
The problem is that since I have an HP DDS-3 drive, I won't be able to count on that mechanism because unfortunately when it ejects the cartrdige, the bottom slide is still in the open position. Actually pulling the cartridge out is what closes it. So the "lip" isn't there. One would have to give the picker some sort of horizontal tweezers to pinch the cartridge and back it out. That sounds rather difficult, unfortunately.
Along the same lines, I think one thing that would largely simplify everyone's life would be to finally get rid of the ISA bus and its associated peripherals. Even if it's a virtual one, just about every PC out there still has an ISA bus that has one or two serial ports, a parallel port, an AT/PS2 keyboard controller, a timer, speaker gizmo, and a floppy controller.
Certainly the floppy, serial, parallel and keyboard/mouse parts could go and not be missed. There are USB versions of all of those. Couldn't the timer/RTC and speaker be made into a PCI device by making some small changes to the OS?
DOS is finally gone. Isn't it now time to flush away the rest of the 8088 compatibility layers as well?
I mailed a CD-R to my brother. When it arrived, it looked almost exactly like a CD-R that had been intentionally destroyed by being put in a microwave for 5 seconds. I ended up hand-delivering a replacement (this was over the Christmas holidays) since I was going to be in town anyway. I can't imagine that they would be using radiation strong enough to do this, but what else could account for this level of damage?
Rotating the keys quickly will not solve the only problems with WEP. In general, encrypting stuff isn't enough. The receiver of encrypted data must insure that the data was not changed in route. Since the packet validator in WEP is CRC rather than something cryptographically strong, it is possible to perform replay attacks, or even worse, replay modified packets and have them be received as if they were legitimate. Even if it is not possible to decrypt packets, the ability to modify and reinject packets and have them be received creates some big problems.
I have had very bad luck with HP Optical products in the past. The net effect really is that you're not buying the product in question, you're renting it for the warranty period. This is because they self destruct at day # (warrenty + 1), and their out-of-warranty repair charge is (price of new drive * 0.99).
Consider keeping everything you've done the same, except change Linux + ext2 to FreeBSD + UFS + softupdates. It will likely make a big improvement. Oracle runs just fine under FreeBSD's Linux emulator (I have heard that you must install Oracle on a Linux machine, then tar up the installation and move it to a FreeBSD machine).
In another year when FreeBSD 5 comes out we hope to give Solaris a run for its money on SMP hardware, believe it or not.
Slashdot Mirror
I've got OO 1.0.1 installed on FreeBSD and Windows XP (for my wife) and I don't know why anyone who knew anything about it would call it "Java based." I believe that java can be used in some sort of browser component burried deep inside, but from what I can see it would run perfectly well without a JRE at all.
I realize that that the actual steps taken are the same. The net result is that actual conduct is what should be judged. Did you really just make a personal backup for yourself? Or are you selling VideoCD copies on eBay?
Really, this is no different than the battle that was fought over encryption. Encryption can be used to keep the sysadmin from reading your private mail, and it can also be used by terrorists to coordinate their activities without fear of interception (of course, one man's terrorists are another man's freedom fighters). Guns can be used for home defense or home invasion. And on and on and on.
So this should really be the battle cry: "It's the conduct, stupid."
It would be no more complex to allow TVs to tune to arbitrary frequencies than to make them cable-ready. Besides, even this problem can be solved by market forces. It is in the mutual interest of both the consumer electronics and broadcasting industries for there to be broad agreement on such matters. That hardly means we require government intervention to achieve it.
I thought of one other area where interstate commerce is involved in VHF - aeronautical mobile services. The emissions from an airplane at cruising altitude can cover vast territory and clearly impact interstate commerce. I believe it's an area for the FAA, but if we really need an FCC to govern interstate communications, then that is one more chunk of real estate I can't argue too much about.
Someone please remind me again why we need the FCC to regulate all of this spectrum?
Here's my point... It is, I suppose, appropriate for the FCC to regulate interstate or international communications. But RF not aimed at a satellite that is higher than 50 MHz (most of the time) is not going to leave the state (yes, there are exceptions, but neighboring states have been cooperating ever since there *were* states in border issues without any federal intervention).
The band plan in Minnesota has absolutely zero impact on me here in California. Heck, where I am neither does the bandplan in Oregon, Idaho, Nevada, Arizona or Mexico. Why is it that a bunch of beaurocrats in Washington DC should have complete and total say over issues involving strictly local transmission and reception of radio signals (we can thank FDR and his buddies, of course, for the question of why they do as opposed to why they should)?
Look at how a switch works. When you transmit a frame, it associates your MAC address with the port you're on. Next time a frame is sent to that MAC address, it sends it only to that port, becuase it knows that address is going to be found only on that port. If the switch does not know where a particular MAC address is to be found, it must send the frame to every port. Therein lies the weakness.
Switches tend to have finite sized MAC tables. If you overload them, they throw away older data in a least-recently-used manner. So the way to turn a switch into a hub is to send a non-stop stream of frames with random Ethernet source addresses. Make them small frames so you don't chew up all of the bandwidth of the switch, and don't send them too frequently (unfortunately, how frequently is too frequently is dependent on how big the switches table is). Address them to a known non-existent MAC address and make them a known unused Ethernet protocol. You will flood the switch's table and it will be forced to broadcast all frames. QED.
None of the digital broadcasters around here are in VHF. IMHO they'd have a hell of a time doing that. Their digital broadcasts are all in UHF, and until they're done with analog broadcasting, they're going to be using those VHF slots for that. Once they've turned the analog VHF off, are they going to then move back down to VHF? Why?
The lower 700 MHz Auction begins tommorow!. This is for UHF channels 52-59.
I believe they're shoving high UHF translators downwards to make more room for cell phones. Look at the bandplan. It's set up as 3 channels (that is, 6 MHz blocks) that air paired, and 2 in the middle that are unpaired.
So what?
In general, lower frequencies tend to suffer a bit less from multipath distortion, suffer less from feedline losses, are easier to engineer, and more efficient to generate.
Channels 2-6 are very low in frequency indeed. They start at 54 MHz (TV channels are 6 MHz wide), but there is a 4 MHz gap between 4 and 5 for various low power services (mostly RC cars and planes), with channel 6 abutting the bottom of the FM radio channels (88 MHz). Now, I think channels 5 and 6 should be dedicated to an amateur broadcasting service, and the rest perhaps to land-mobile activities, but channels 7-13 are the perfect place for low power data services.
Of course, it's going to be years before the VHF TV transmitters are finally turned off, but I do believe it will happen eventually, and if we don't plan well in advance, there will be a smoke-filled-room give-away of this prime spectrum to someone with a lot of money, which isn't necessarily in the best interests of everyone.
The assertion made is that digital sampling of the inputs on the piano, recording the digital samples and playing them back is different enough from a live performance to be inferrior. I will stipulate that the visual presence of the artist adds to a live performance, but set that aside for a moment.
It is a lot like the Turing test in some regard. If you blindfold an audience and play them the same song first by a live musician and second by recording the *same* live musician, and if the audience can't reliably pick out which is the recording and which is the musician, then I'd say the recording technology in question passes the test.
Legitimate arguments can still be made that the performer may change his performance to suit different circumstances, and such adaptations would not be possible with a recording. But those arguments lie outside the confines of whether or not the recording mechanism is sufficiently accurate to capture and reproduce a single performance by the artist.
But in the context of a competition, I believe it may be actually be a benefit to the objectivity of the judging. The judges won't know who is who, so no favoritism will be possible. The artists' inability to tailor their performances for the judges will be a hinderance borne equally by all of the contestants. But unlike an audio recording, no extraneous noise other than that made by the performer will be recorded.
I'm all for Microsoft adopting open source. It would be the best thing for everyone except Microsoft. Which is precisely why it will never happen.
Microsoft OSes have historically been a technical laughing stock (I am willing to stipulate for the purpose of this discussion that Windows 2000 was their first serious contender. I actually think it's the best piece of code they ever released. I put it more or less on a par with Mac OS X, but I digress). So if they were so bad, why did they succeed? Because they had monopolistic market control that they misused (remember: A federal court has actually found them guilty of this. There's no longer any reason for debate on this point).
How do they maintain their stranglehold? It's quite simple: Embrace, Enhance, Exclude. For those of you who have not had this tactic properly explained, it goes like this: Microsoft sees an open source protocol succeed, it then adds support for that protocol to one of its product suites, but in doing so makes a very slight modification the specification for which it does not release. Microsoft uses this modification as a compatibility barrier for other products.
This is why Microsoft is so frightened of the GPL, but is not threatened by non-GPL open source. They cannot take a GPLed program and perform EEE with it, because they must release the source to their modifications. It then becomes easy to insure compatibility with the Microsoft variant.
Of course, Microsoft turns this around and claims that it eliminates their "freedom to innovate." Nothing could be further from the truth. They are free to innovate within the GPL. They simply would be disallowed from doing so in a manner whose sole purpose is to screw anyone not using their software.
Without their playbook of anticompetitive behavior, Microsoft deep in its soul believes it cannot succeed. A Microsoft that would be willing to participate in any meaningful way in Open Source would be so different as to hardly be comparable to what exists today.
Surely someone out there makes an ATX power supply that takes 12 VDC as an input. You could take a rather expensive component (the AC inverter) out of the picture and increase the overall efficiency by a measurable amount.
:-)
Oh, and the greeniacs out there will be no doubt pointing out that there's no particular reason that the AC->DC step couldn't be replaced by a bunch of solar panels and a charge controller.
1. Eliminate software and business method patents.
2. Strictly limit the timespan of copyrights. Holywood has a habit of getting the expiration of copyrights pushed out whenever they come close to their expiration date. That sort of nonsense has got to stop. Copyrights should be granted for 50 years to the original creater, his estate or assignee. Period. That's all. Finito.
3. A media consumer's bill of rights that formally states that a consumer may do *ANYTHING* with a copyrighted work for which they have purchased a copy, *except* for distributing additional copies of same. Too often the debate has centered around what rights should be granted to copy owners (as opposed to copyRIGHT owners). BS. They should have *all* rights *except* for a short list.
4. Note that case number 3 talks about the simple sale of a copy. It does not preclude a seller from including restrictions in the purchasing contract on the use by the recipient. But such restrictions *must* be part of the contract agreed to by *both* parties *before* performance of the contract. Click-through or shrink-wrap licenses or other shenanigans that are tacked on *after* both parties have performed the simple exchange of money for a copy *explicitely* should not be allowed to be binding.
There may be reasons folks want to jam cell phones. Reducing RF almost certainly wouldn't be one of them. If you start interfering with it, the first thing it's going to do is turn the power up to try and talk over the interference. Congratulations: You have now made the cell phone transmit more RF, never mind the fact that your own jammer emits RF as well.
I've been complaining about this one for years now.
Outlook can be used in one of two configurations (that matter): "Internet only", which has IMAP and POP capabilities, and "Corporate", which has Exchange and POP capabilities. This Hobson's choice is an anti-competitive barrier they've put in place to keep people from using IMAP servers for e-mail.
One way to transition away from Exchange is to move the mail to an IMAP server and leave the contacts, calendar stuff, etc on the Exchange server...... Except, of course, for the fact that Microsoft has made that impossible.
Another way that Microsoft could fix this would be to allow Outlook to store non-email content as MIME attachments in IMAP servers. That is, a particular IMAP folder could be designated as a Contacts or Calendar folder and Outlook could store that type of information as MIME attachments to e-mail messages. But, of course, that would make it possible to get all of the functionality of Outlook without an Exchange server, which is something Microsoft Can Never Allow.
A very long time ago (Think: SunOS 4.x), you had to have more swap than RAM because the amount of virtual memory you had was EQUAL to the amount of swapspace you had. That is, every page of RAM had to be backed by a page of swap, or else it wouldn't end up being useful (I'm oversimplifying a bit).
Now, in order for FreeBSD to be willing to save a core image, you have to have a swap partition with more space than you have in RAM, otherwise savecore will refuse to set things up. But for FreeBSD, the amount of virtual memory you have is equal to the amount of RAM you have PLUS the amount of swap space you've got set up (again, there is some RAM that gets used to hold the kernel image, so this is a bit of a simplification). Given that, it is perfectly ok to run a machine without any swap at all, provided you have a sufficient amount of memory to do everything you want to do. But having swap is good because it gives you some cushion, plus if you want to save cores from panics, you must, as I said, configure a swap partition with at least as much space as you have RAM.
I dunno.... Looking at the pictures it looks like there's an awful lot of unshielded plastic. The reason PCs go into metal boxes is so that you don't wreck your next door neighbor's TV reception (or worse).
Two jobs ago, I worked for Enlighten Software, whose EnlightenDSM product was supposed to do all this with some bells and whistles. They went bankrupt trying.
:-)
The only way I see it happening is if some common configuration API gets some traction and then everyone agrees to write shims to get it to talk to their favorite OS or piece of middleware. I am not optimistic.
They can't insure that a party hack isn't standing behind absentee voters watching them fill in their ballots from their kitchen tables. What's the difference?
All that said, I think the security questions are a lot more important. It's probably a lot easier for dead former Chicago residents to vote over the Internet than... Oh wait...
Just recently I had to go from 0 knowledge of RDBMS to a working web app. The elephant postgresql book by Bruce Momjian (ISBN: 0201703319) was just about perfect. It explained from reasonably basic concepts using lots of examples, but wasn't really hugely wordy. It went from the basics through some fairly advanced usage, but left out the really arcane stuff (but did reference where you could find that if you wanted it).
More important than anything else, IMHO, was that it has a really, really good index.
Speaking at least for myself, that book should be used as a model for other books of its class.
First, like so many others in this thread, let me say that the force is strong in this one.
From what I can figure from the pictures, it does ejection the same way that it does picking -- counting on the little lip on the bottom of the cartridge to stick out far enough that the matching hump on the end of the spatula can grab it.
The problem is that since I have an HP DDS-3 drive, I won't be able to count on that mechanism because unfortunately when it ejects the cartrdige, the bottom slide is still in the open position. Actually pulling the cartridge out is what closes it. So the "lip" isn't there. One would have to give the picker some sort of horizontal tweezers to pinch the cartridge and back it out. That sounds rather difficult, unfortunately.
Along the same lines, I think one thing that would largely simplify everyone's life would be to finally get rid of the ISA bus and its associated peripherals. Even if it's a virtual one, just about every PC out there still has an ISA bus that has one or two serial ports, a parallel port, an AT/PS2 keyboard controller, a timer, speaker gizmo, and a floppy controller.
Certainly the floppy, serial, parallel and keyboard/mouse parts could go and not be missed. There are USB versions of all of those. Couldn't the timer/RTC and speaker be made into a PCI device by making some small changes to the OS?
DOS is finally gone. Isn't it now time to flush away the rest of the 8088 compatibility layers as well?
I mailed a CD-R to my brother. When it arrived, it looked almost exactly like a CD-R that had been intentionally destroyed by being put in a microwave for 5 seconds. I ended up hand-delivering a replacement (this was over the Christmas holidays) since I was going to be in town anyway. I can't imagine that they would be using radiation strong enough to do this, but what else could account for this level of damage?
Rotating the keys quickly will not solve the only problems with WEP. In general, encrypting stuff isn't enough. The receiver of encrypted data must insure that the data was not changed in route. Since the packet validator in WEP is CRC rather than something cryptographically strong, it is possible to perform replay attacks, or even worse, replay modified packets and have them be received as if they were legitimate. Even if it is not possible to decrypt packets, the ability to modify and reinject packets and have them be received creates some big problems.
I have had very bad luck with HP Optical products in the past. The net effect really is that you're not buying the product in question, you're renting it for the warranty period. This is because they self destruct at day # (warrenty + 1), and their out-of-warranty repair charge is (price of new drive * 0.99).
Consider keeping everything you've done the same, except change Linux + ext2 to FreeBSD + UFS + softupdates. It will likely make a big improvement. Oracle runs just fine under FreeBSD's Linux emulator (I have heard that you must install Oracle on a Linux machine, then tar up the installation and move it to a FreeBSD machine).
In another year when FreeBSD 5 comes out we hope to give Solaris a run for its money on SMP hardware, believe it or not.