$5,000 net monthly would seem to be around $75,000 anually,
Actually, it would be around $94K gross (in a state with income tax), and it would be closer to $500K mortgage, assuming the interest rates were typical for some of the "shaky" loans that were made.
They still definitely overcommitted, but where I live, $500K doesn't buy a whole lot of house. There's no way I could afford to buy the house I live in right now, but I got lucky and bought a while ago at the bottom of the market and was close to the same state they were for a while ($1,200 house payment on about $2,800 net).
They were at 70% and I was only about 43%, but today even with a lot more pay, I'd be over 60% if I wanted to buy my house today (assuming the same percentage down payment as when I really bought it).
For me this is the paypal killer. Not only is it secure convenient and trustworthy but banks and credit unions, at least in Canada are pretty customer service oriented...toll free 24-hour hot-lines, and genuinely useful staff are the norm in my experience with TD, RBC, and Scotiabank. Contrast that with Paypal.
But, this means you have to actually have the money to send. With PayPal (or other payment services), you can back your payment with a credit card. For the vast amount of people who rack up a lot of debt, this is important.
I use a credit card for PayPal payments because I get the credit card rewards. Plus, I don't like some of the PayPal practices very much. In particular, at one time sending "cash" (i.e., transfer from a bank account) was free. Later, it had a small fee. Now, it is exactly the same as paying with a credit card. So, I figure if PayPal is going to charge money as if you used a credit card, then it might as well actually cost them that much.
I ran Windows 2000 and XP on a machine for a couple of years and averaged one BSOD a month.
Considering that it was running on an Intel motherboard that had been recalled due to memory controller errors (one of the 810 series, IIRC), that's not too bad for an OS that supposedly crashes "all the time" on good hardware.
I didn't use Firefox 2, so I don't know the exact functionality, but I don't think it takes much to get the "Awesome Bar" like people seem to want (matches only at the beginning of URL, no match on titles).
First install the Hide Unvisited extension. Next, set "browser.urlbar.search.chunkSize = 0" in about:config. Last, add the following to your "userChrome.css" file:
Did you read the part where they measured total power expended to accomplish various tasks? Burning more power is OK if you can finish the job faster and get back into a low power state.
Their numbers are messed up, though.
The VIA system uses 17.4 more watts while under load than the Intel system. That means if they are under load for one second, the VIA system has used 17.4 more watt-seconds than the Intel. On the other hand, the Intel only uses 0.9 more watts under load than the VIA uses when idle. So, if the VIA works for one second, then drops back to idle while the Intel is still working to complete the task, it would take 18.33 more seconds before their total power usage becomes the same (17.4 + (-0.9 * 18.33) == 0).
This means that to use the same amount of power over the long term, the VIA has to be able to complete tasks in ~5% of the time that the Intel takes. Put another way, the Intel would take 19 times as long to finish any task.
But, their own graphs show that the Intel system only takes about 2 times as long (at most) to finish tasks. So, there's no way that the VIA uses less energy over the long haul, but that's what they claim.
Their problem is that they are comparing the total power used by the VIA system only while "active" on the task, and not looking at the power that the VIA uses on idle while the Intel is still completing the task. This is like saying that as soon as you finish that task, you immediately power off (or drop to standby).
If your warning light comes on at just one gallon, you've got less than 30 miles for almost every car made, and there's a lot of great big stretches of nothing in Texas, New Mexico, Arizona, Wyoming, Montana, North Dakota, Nevada, or any other large state.
It's not always that simple. With a white box (at least all the ones I've seen) you dont get a nice BMC or ILO card, you dont get hot-swappable dual redundant power supplies, you dont get a nice turnkey hardware management software that will alert you if something starts to go (you can do it, but its all home-grown work, with hp/dell its turnkey).
You dont get a well-engineered airflow/cooling in a slick 2u rackmount. You generally dont get a mchine that is as maintainable without tools, or as well engineered internally.
I didn't buy the management cards for my motherboards because I didn't need them, and price comparisons I did were without them. Plus, based on the Dells we have at work, their "alerting" leaves much to be desired...a hard drive in a RAID-5 array was failed for nearly 6 months with no alert, and when the 2nd drive failed and we did get an alert, it was too late. Power supplies, airflow, hot swappable drives, etc., are all functions of the box, and you can pay for that if you need it (which I mostly do).
I don't go 2U size because that doesn't allow expandability or quiet operation (you can't fit large slow-moving fans in that small of a box), and I like to keep my systems running cool and quiet. I also think it's stupid that I can get a $30 CPU cooler that keeps the processor significantly cooler and is quieter than anything Dell/HP/whoever slaps on.
As for tools, I'm not sure what this fascination is with not using tools. Every system I've seen requires a screwdriver to install the hard drives in the hot swap racks (although the racks themselves just slide in), no CPU cooler can be installed/removed without some tools, etc. Most "no tools" systems just have regular screws replaced with thumbscrews, which generally require a tool unless they are really out in the open. I can do the same thing for about $0.01/screw, and having a #2 Phillips hanging on a wall near the servers isn't really much of a problem.
There's no getting around build quality, and I do think it's silly sometimes that I'm paying over 10% of the system cost for the case and power supply for what I build, but they last forever (I've been recyling ATX-standard chassis for nearly 10 years).
What are the motherboards and chipsets your getting with the whitebox? Sometimes they're the same (vanilla intel), sometimes not.
What kind of automated driver/bios/flash update tools do you get with your whitebox?
Since this is ESX we're talking about, you have to go with motherboards on the HCL, so you really don't get much of a choice. I use AMD on Tyan and Supermicro motherboards because until Intel came out with the Yorkfield processors, you couldn't get as many cores at a good price with low power usage. Even so, the FB-DIMMs required by Intel raises the power, heat, and cost back up again.
For updates, again, this is ESX and you don't update drivers or BIOS on the bare metal unless it is critical, and if you do, the drivers (plus the automated update utility) are written by VMware, not Dell, etc. BIOS would be handled by the motherboard manufacturer, but I really don't want it automated, since such a large percentage of updates go poorly and leave the machine FUBAR.
Really? Plain regular 2U rackmount servers from dell (like the poweredge 2950) are priced competitively with ibm, hp and other major vendors.
So, what you're saying is that they all have outrageous markups?;->
Really, try and outfit a machine with 8 cores at 2GHz, 16GB of RAM (with room for 48GB more), six 1-Gbit NICs, and an 8-port SATA RAID controller with three 500GB drives (and room for 5 more), and two PCIe x4 slots unused (for future expansion) for less than $4,000 at any of the major vendors. You can white box that machine with VMware HCL list hardware for less than $2500.
There's got to be some "Linux stuff" in ESXi, as none of the hardware drivers are part of the vmkernel...they are all part of the kernel that started out as RedHat Enterprise Linux 3 and had some mods done by VMware (and source for all the mods is available, as per the GPL).
I think that the only real difference between ESX and ESXi is that you don't have the extra VM running as the "console", plus no web access running in userland (which reduces the memory footprint dramatically).
But, without that console, you lose a lot. For me, I have added a decent amount of software (a better NTP daemon, for one), modified the firewall rules to add services, and done some recovery when things went bad (my storage controller is technically unsupported). I would suspect you also can't transfer files to the server via sftp, and can't use scripts to maintain the VMs (like this automated backup script).
OK, so which of those remote access systems (X, MS Remote Desktop, or ssh) allow you to plug in a Firewire drive on the client machine and have it show up as a local drive on the the server (remote machine)?
RDP lets you see it as a network drive, and that still works fine even if the remote machine is a VM running under ESX. Neither of the others support any kind of "client drive visible on the remote machine" setup.
The problem is that ESX isn't like other virtualization systems where you have virtualization software running on your "desktop". On those, it makes sense to be able to plug a USB (or Firewire) drive into the "server" and see it on the VM as native. But, when the server machine doesn't have a local display that shows the VM screen, it just doesn't make any sense.
VMware ESX is really designed for virtualization of server operating systems. Although it can run quite a few "desktop" OS versions, there is no good console system...the built-in remote console view is OK for installs, tweaks that must be done from the console, etc., but not really good enough for extensive use.
ESX does support USB connected to the ESX server, but only for use with the "ESX console VM", and you can't store other VMs on USB. Since ESXi doesn't have a console VM, even that isn't supported.
I'm running 20 Windows Server 2003 boxen on a single HP DL385 G3 with 2 AMD 2218's and 16GB RAM, and I'm still only running at about 60-70% utilization.
For the standard version of Virtual Infrastructure you're going to spend around $2500-$6000, plus around $5000-$10000 for 1 or 2 servers to run it.
Although there is no getting around the cost of VI from VMware, the only thing that's really expensive is storage for all the VMs you need to run, and the outrageous markup on the servers on the VMware HCL.
If you "white box" with all VMware-certified hardware, you can build a very nice server (two quad-core 2GHz AMD, 16GB ECC RAM, 6 1-Gbit NICs, PCIe SAS RAID) for less than $3000. But once you start talking SAN, then you start to have very few choices but to pay the outrageous prices.
ESX or ESXi works just fine with a bunch of plain old IDE and SATA controllers...see here for more information.
You can't put virtual machines on an IDE drive, but you can put them on SATA disks with the controllers listed at that link. You don't get RAID on any of them, though, even if they have some sort of RAID available. ESX(i) only officially supports storing VMs on RAID arrays if the disks appear to be SCSI of some sort (including SAS, or SATA on an SAS-capable controller).
You could also use Openfiler to create iSCSI targets that ESXi can use to store VMs, and Openfiler can use any storage that any modern Linux can use, including Linux software RAID. This allows you to have a VMware ESX(i) setup permanently (ESX was available as a free 90-day trial) on some pretty cheap hardware.
This is why punitive damages should be subjected to a means test whereby the damages are adjusted to reflect a fixed percentage of the annual income of a convicted individual. Thus, the poor working mother might only pay several hundred dollars total or perhaps a couple of thousand max whereas the mega corporation could be on the hook for millions.
Although a nice sentiment, the problem here is that a "real" piracy operation (e.g., one that makes counterfeit DVDs and boxes of software, movies, etc.) would use a "low income" front man who would only have to pay a small fraction of the appropriate damages.
Since such an operation wouldn't be expected to keep accurate records, it would again just be the copyright holder guessing at the number of copies, and there wouldn't be much proof around as to how much actual profit the operation made, so it would come down to "poor person, small fine".
You could also extend this to anyplace like a torrent search engine or tracker site, where they likely don't make a large amount of money, but according the the *AA groups, they do a lot of infringement. Because of this, the *AA would never let such a law get passed, even if someone becomes homeless and makes the national news because they had to pay an insane amount of damages for copying a few MP3s.
If you had 100 guys in a room, each of which with their own ideas, how do you know which of those is going to make a movie that will gross 300 million dollars?
This "need to gross 300 million dollars" is the problem with the movie industry today.
Any product that grosses even 5 times the cost to make and advertise it should be considered wildly successful, and movies shouldn't be an exception. And, there have been a lot of movies that have done this, but very few of them grossed $300 million, because they only cost between one and twenty million to make.
The problem is that making lots of movies that gross $50 million (and only cost $10-20 million) is not what the studios want to do...they want $200-300 million blockbusters.
As an aside, since the same production companies that make movies also make TV shows, this is part of the reason that there are so few good new TV shows. Essentially, a season of TV is about a $10-30 million dollar production and can gross $50-200 million in all forms (original broadcast rights, syndication, DVDs, etc.), but it's not as "Hollywood" as making 10 movies with one being a $300 million blockbuster and the other 9 grossing a total of $300 million and with a cost of over $400 million to make.
Cable ops actually can't modify a terrestrial broadcast signal. This includes down converting and I would assume also refers to compression. (Though I could be wrong on the compression).
The only thing that can't be "modified" without getting in quick trouble is content. In other words, no replacement of OTA commercials with a Comcast promo.
There is some question as to whether this part can be modified by a re-transmission agreement, but very few stations would agree to that, anyway.
Otherwise, it's wide open. Cable companies regularly put two HD channels in a single analog slot, and that requires generally re-compression. But, Fox manages to send most of their OTA in less than 10Mbps, because they compress at the network end using very, very expensive equipment, and don't have the local stations re-compress for the actual broadcast (logos are inserted with some tricky MPEG-2 overlay system). DirecTV is using MPEG-4 to re-transmit local digital signals, which is definitely re-compression.
For cable HD, though, it's far worse, as re-scaling from 1920x1080 to 1440x1080 (or less) is common, as is bandwidth starving.
About the only thing that the stations have at their disposal for this is a lawsuit against the cable company, but since most stations foolishly think that we want to see pixelated HD and 3 subchannels of useless stuff, I don't see those lawsuits happening anytime soon. Some of the networks (CBS for one) do protect the quality through affiliate agreements, and that may be the only way that HDTV doesn't turn into "barely better than SD".
There are NO circumstances under which one user should possess another user's password; not even an Administrator.
For VPN (particularly the Cisco VPN client), this isn't 100% true.
The story says these are "departmental passwords", and Cisco VPN has the concept of "group authentication" that drops you into a group of VPN users. To completely connect, you need the group name, the group password, your username, and your password. Where I am, you get the group name and password as part of the admin install of the client, which places it into the connection profile. So, I "know" the same password as a lot of other people (the ones in my group).
It is stored in the connection profile as obfuscated, but it can't be a hash, as it has to be sent as the original when connecting. And, there are utilities that can reverse the obfuscation in a fraction of a second. But, because of the multiple factors, it's not really a large risk, because you have to match up the user with the right group, and then get the user password right. The Cisco VPNs I connect to use an RSA key for the user password, and if that's the case here, it's not a big deal at all to release this info.
So, if these "departmental passwords" are just Cisco VPN group usernames and passwords, it's not much of an issue, as it's pretty easy to get access to them, unless they force their users to memorize them as well as their personal password.
Just to follow up on it, I read through the thread and found that Foxconn linked to a page on Microsoft's site which supposedly explains ACPI compliance. Interestingly enough, that page refused to display on anything but IE.
The page does work in Firefox 3 if you use the User Agent Switcher extension to fool it into thinking you have IE (6 or 7).
Then greylisting was implemented, which stopped the spam dead in its tracks, and completely nullified the spam that Double Dribble couldn't stop. That's when I turned over the account to another party. I still use greylisting personally with great success.
For me, between greylisting and requiring strict RFC compliance for the "HELO" parameter, pretty much no spam gets through to even be looked at by SpamAssassin.
For the "HELO" parameter, almost every spambot uses one of:
something that isn't a fully qualified domain name ("laptop", "Notebook", and "PC-200806211153" are some recent examples)
an IP address
Neither of these are acceptable (according to section 2.3.5 of the SMTP RFC) as the "HELO" parameter.
Then, I throw out a few more bogus things, like:
my host/domain name
my public IP address
domain literals (i.e., an IP address surrounded by square brackets) that have an IP address in a bogon range
At this point, the e-mail gets to face greylisting, ClamAV, and SpamAssassin. About 1 in 100 "bad emails" get through to the end users.
I don't use NoScript, because it turns normal web browsing into a circus of "allow" clicks, and makes UAC look good.
I have to say that I haven't seen this when using NoScript, except on one site that pulled in scripts from its parent domain, the "www.parent" server, and various machines from the company that really owned the site.
Otherwise, it's generally a matter of not having to enable scripting on a site because I do almost all clicking around without it, or else just selecting "Allow example.com" once from the NoScript menu and never worrying about it again. "Temporarily allow all this page" is also handy for just dealing with a site that you might not come back to often, but really need scripting to do something.
It might be because I don't primarily use NoScript as a protection against security issues, but as a speedup tool. The fact that it protects me from random domains is great, but I wouldn't want to use it in such a way that I would be protected from my favorite site getting pwned (i.e., only use "Temporarily Allow" instead of "Allow"). I can't see anybody using NoScript that way, really.
In *my* experience with the new FF I have found the memory leak is still there *but* it appears to be less than before. In prior versions I had issues after just a couple of hours of browsing. Now? It is still eating way more RAM than it should and an undocumented check shows that it steadily increases and doesn't seem to want to let a whole hell of a lot go after closing some tabs.
For me, it's the exact reverse.
Doing the same sort of browsing with Firefox 3 as I did with IE6 (company standard), I find that Firefox uses far less memory. It's using 130MB right now, where before it wasn't unusual to see my "main" IE window using 130MB, with the others taking up another 100MB or so.
I also noticed that every so often IE would start to use the CPU for no reason. Even clicking "Home" (a blank page for me) on all IE windows wouldn't stop it, and it would sit there taking 5-10% of my dual-core 2.8GHz CPU. Only closing the right window (which I couldn't determine in any way other than guessing) would stop the CPU use.
Except for Firefox not playing well with Sharepoint and my virtual credit card number software, and a small bug with an extension, it's been rock solid.
I use IE more than any other browser out there. Yeah... Really. I use it for the add-ons. IE has the add-ons that I want that work how I like them and so I use it.
What add-ons? Part of the reason I wanted to switch to Firefox was the fact that there were so many useful add-ons, but Firefox 2.x just wasn't ready for prime time (at least not for me).
Slashdot Mirror
I think the "CAN SPAM" law pretty much does what it says on the tin...it allows even more people to send spam.
$5,000 net monthly would seem to be around $75,000 anually,
Actually, it would be around $94K gross (in a state with income tax), and it would be closer to $500K mortgage, assuming the interest rates were typical for some of the "shaky" loans that were made.
They still definitely overcommitted, but where I live, $500K doesn't buy a whole lot of house. There's no way I could afford to buy the house I live in right now, but I got lucky and bought a while ago at the bottom of the market and was close to the same state they were for a while ($1,200 house payment on about $2,800 net).
They were at 70% and I was only about 43%, but today even with a lot more pay, I'd be over 60% if I wanted to buy my house today (assuming the same percentage down payment as when I really bought it).
For me this is the paypal killer. Not only is it secure convenient and trustworthy but banks and credit unions, at least in Canada are pretty customer service oriented...toll free 24-hour hot-lines, and genuinely useful staff are the norm in my experience with TD, RBC, and Scotiabank. Contrast that with Paypal.
But, this means you have to actually have the money to send. With PayPal (or other payment services), you can back your payment with a credit card. For the vast amount of people who rack up a lot of debt, this is important.
I use a credit card for PayPal payments because I get the credit card rewards. Plus, I don't like some of the PayPal practices very much. In particular, at one time sending "cash" (i.e., transfer from a bank account) was free. Later, it had a small fee. Now, it is exactly the same as paying with a credit card. So, I figure if PayPal is going to charge money as if you used a credit card, then it might as well actually cost them that much.
I ran Windows 2000 and XP on a machine for a couple of years and averaged one BSOD a month.
Considering that it was running on an Intel motherboard that had been recalled due to memory controller errors (one of the 810 series, IIRC), that's not too bad for an OS that supposedly crashes "all the time" on good hardware.
I didn't use Firefox 2, so I don't know the exact functionality, but I don't think it takes much to get the "Awesome Bar" like people seem to want (matches only at the beginning of URL, no match on titles).
First install the Hide Unvisited extension. Next, set "browser.urlbar.search.chunkSize = 0" in about:config. Last, add the following to your "userChrome.css" file:
Did you read the part where they measured total power expended to accomplish various tasks? Burning more power is OK if you can finish the job faster and get back into a low power state.
Their numbers are messed up, though.
The VIA system uses 17.4 more watts while under load than the Intel system. That means if they are under load for one second, the VIA system has used 17.4 more watt-seconds than the Intel. On the other hand, the Intel only uses 0.9 more watts under load than the VIA uses when idle. So, if the VIA works for one second, then drops back to idle while the Intel is still working to complete the task, it would take 18.33 more seconds before their total power usage becomes the same (17.4 + (-0.9 * 18.33) == 0).
This means that to use the same amount of power over the long term, the VIA has to be able to complete tasks in ~5% of the time that the Intel takes. Put another way, the Intel would take 19 times as long to finish any task.
But, their own graphs show that the Intel system only takes about 2 times as long (at most) to finish tasks. So, there's no way that the VIA uses less energy over the long haul, but that's what they claim.
Their problem is that they are comparing the total power used by the VIA system only while "active" on the task, and not looking at the power that the VIA uses on idle while the Intel is still completing the task. This is like saying that as soon as you finish that task, you immediately power off (or drop to standby).
If your warning light comes on at just one gallon, you've got less than 30 miles for almost every car made, and there's a lot of great big stretches of nothing in Texas, New Mexico, Arizona, Wyoming, Montana, North Dakota, Nevada, or any other large state.
It's not always that simple. With a white box (at least all the ones I've seen) you dont get a nice BMC or ILO card, you dont get hot-swappable dual redundant power supplies, you dont get a nice turnkey hardware management software that will alert you if something starts to go (you can do it, but its all home-grown work, with hp/dell its turnkey).
You dont get a well-engineered airflow/cooling in a slick 2u rackmount. You generally dont get a mchine that is as maintainable without tools, or as well engineered internally.
I didn't buy the management cards for my motherboards because I didn't need them, and price comparisons I did were without them. Plus, based on the Dells we have at work, their "alerting" leaves much to be desired...a hard drive in a RAID-5 array was failed for nearly 6 months with no alert, and when the 2nd drive failed and we did get an alert, it was too late. Power supplies, airflow, hot swappable drives, etc., are all functions of the box, and you can pay for that if you need it (which I mostly do).
I don't go 2U size because that doesn't allow expandability or quiet operation (you can't fit large slow-moving fans in that small of a box), and I like to keep my systems running cool and quiet. I also think it's stupid that I can get a $30 CPU cooler that keeps the processor significantly cooler and is quieter than anything Dell/HP/whoever slaps on.
As for tools, I'm not sure what this fascination is with not using tools. Every system I've seen requires a screwdriver to install the hard drives in the hot swap racks (although the racks themselves just slide in), no CPU cooler can be installed/removed without some tools, etc. Most "no tools" systems just have regular screws replaced with thumbscrews, which generally require a tool unless they are really out in the open. I can do the same thing for about $0.01/screw, and having a #2 Phillips hanging on a wall near the servers isn't really much of a problem.
There's no getting around build quality, and I do think it's silly sometimes that I'm paying over 10% of the system cost for the case and power supply for what I build, but they last forever (I've been recyling ATX-standard chassis for nearly 10 years).
What are the motherboards and chipsets your getting with the whitebox? Sometimes they're the same (vanilla intel), sometimes not.
What kind of automated driver/bios/flash update tools do you get with your whitebox?
Since this is ESX we're talking about, you have to go with motherboards on the HCL, so you really don't get much of a choice. I use AMD on Tyan and Supermicro motherboards because until Intel came out with the Yorkfield processors, you couldn't get as many cores at a good price with low power usage. Even so, the FB-DIMMs required by Intel raises the power, heat, and cost back up again.
For updates, again, this is ESX and you don't update drivers or BIOS on the bare metal unless it is critical, and if you do, the drivers (plus the automated update utility) are written by VMware, not Dell, etc. BIOS would be handled by the motherboard manufacturer, but I really don't want it automated, since such a large percentage of updates go poorly and leave the machine FUBAR.
Really? Plain regular 2U rackmount servers from dell (like the poweredge 2950) are priced competitively with ibm, hp and other major vendors.
So, what you're saying is that they all have outrageous markups? ;->
Really, try and outfit a machine with 8 cores at 2GHz, 16GB of RAM (with room for 48GB more), six 1-Gbit NICs, and an 8-port SATA RAID controller with three 500GB drives (and room for 5 more), and two PCIe x4 slots unused (for future expansion) for less than $4,000 at any of the major vendors. You can white box that machine with VMware HCL list hardware for less than $2500.
VMware Server might be a cheap alternative if you can't shell out the $300 for Workstation.
I take it you're not in the US? There are many legtimate places (e.g., NewEgg) here that sell brand new full VMware workstation for $200.
There's got to be some "Linux stuff" in ESXi, as none of the hardware drivers are part of the vmkernel...they are all part of the kernel that started out as RedHat Enterprise Linux 3 and had some mods done by VMware (and source for all the mods is available, as per the GPL).
I think that the only real difference between ESX and ESXi is that you don't have the extra VM running as the "console", plus no web access running in userland (which reduces the memory footprint dramatically).
But, without that console, you lose a lot. For me, I have added a decent amount of software (a better NTP daemon, for one), modified the firewall rules to add services, and done some recovery when things went bad (my storage controller is technically unsupported). I would suspect you also can't transfer files to the server via sftp, and can't use scripts to maintain the VMs (like this automated backup script).
OK, so which of those remote access systems (X, MS Remote Desktop, or ssh) allow you to plug in a Firewire drive on the client machine and have it show up as a local drive on the the server (remote machine)?
RDP lets you see it as a network drive, and that still works fine even if the remote machine is a VM running under ESX. Neither of the others support any kind of "client drive visible on the remote machine" setup.
The problem is that ESX isn't like other virtualization systems where you have virtualization software running on your "desktop". On those, it makes sense to be able to plug a USB (or Firewire) drive into the "server" and see it on the VM as native. But, when the server machine doesn't have a local display that shows the VM screen, it just doesn't make any sense.
What do you need Firewire support for?
VMware ESX is really designed for virtualization of server operating systems. Although it can run quite a few "desktop" OS versions, there is no good console system...the built-in remote console view is OK for installs, tweaks that must be done from the console, etc., but not really good enough for extensive use.
ESX does support USB connected to the ESX server, but only for use with the "ESX console VM", and you can't store other VMs on USB. Since ESXi doesn't have a console VM, even that isn't supported.
I'm running 20 Windows Server 2003 boxen on a single HP DL385 G3 with 2 AMD 2218's and 16GB RAM, and I'm still only running at about 60-70% utilization.
For the standard version of Virtual Infrastructure you're going to spend around $2500-$6000, plus around $5000-$10000 for 1 or 2 servers to run it.
Although there is no getting around the cost of VI from VMware, the only thing that's really expensive is storage for all the VMs you need to run, and the outrageous markup on the servers on the VMware HCL.
If you "white box" with all VMware-certified hardware, you can build a very nice server (two quad-core 2GHz AMD, 16GB ECC RAM, 6 1-Gbit NICs, PCIe SAS RAID) for less than $3000. But once you start talking SAN, then you start to have very few choices but to pay the outrageous prices.
ESX or ESXi works just fine with a bunch of plain old IDE and SATA controllers...see here for more information.
You can't put virtual machines on an IDE drive, but you can put them on SATA disks with the controllers listed at that link. You don't get RAID on any of them, though, even if they have some sort of RAID available. ESX(i) only officially supports storing VMs on RAID arrays if the disks appear to be SCSI of some sort (including SAS, or SATA on an SAS-capable controller).
You could also use Openfiler to create iSCSI targets that ESXi can use to store VMs, and Openfiler can use any storage that any modern Linux can use, including Linux software RAID. This allows you to have a VMware ESX(i) setup permanently (ESX was available as a free 90-day trial) on some pretty cheap hardware.
This is why punitive damages should be subjected to a means test whereby the damages are adjusted to reflect a fixed percentage of the annual income of a convicted individual. Thus, the poor working mother might only pay several hundred dollars total or perhaps a couple of thousand max whereas the mega corporation could be on the hook for millions.
Although a nice sentiment, the problem here is that a "real" piracy operation (e.g., one that makes counterfeit DVDs and boxes of software, movies, etc.) would use a "low income" front man who would only have to pay a small fraction of the appropriate damages.
Since such an operation wouldn't be expected to keep accurate records, it would again just be the copyright holder guessing at the number of copies, and there wouldn't be much proof around as to how much actual profit the operation made, so it would come down to "poor person, small fine".
You could also extend this to anyplace like a torrent search engine or tracker site, where they likely don't make a large amount of money, but according the the *AA groups, they do a lot of infringement. Because of this, the *AA would never let such a law get passed, even if someone becomes homeless and makes the national news because they had to pay an insane amount of damages for copying a few MP3s.
If you had 100 guys in a room, each of which with their own ideas, how do you know which of those is going to make a movie that will gross 300 million dollars?
This "need to gross 300 million dollars" is the problem with the movie industry today.
Any product that grosses even 5 times the cost to make and advertise it should be considered wildly successful, and movies shouldn't be an exception. And, there have been a lot of movies that have done this, but very few of them grossed $300 million, because they only cost between one and twenty million to make.
The problem is that making lots of movies that gross $50 million (and only cost $10-20 million) is not what the studios want to do...they want $200-300 million blockbusters.
As an aside, since the same production companies that make movies also make TV shows, this is part of the reason that there are so few good new TV shows. Essentially, a season of TV is about a $10-30 million dollar production and can gross $50-200 million in all forms (original broadcast rights, syndication, DVDs, etc.), but it's not as "Hollywood" as making 10 movies with one being a $300 million blockbuster and the other 9 grossing a total of $300 million and with a cost of over $400 million to make.
Cable ops actually can't modify a terrestrial broadcast signal. This includes down converting and I would assume also refers to compression. (Though I could be wrong on the compression).
The only thing that can't be "modified" without getting in quick trouble is content. In other words, no replacement of OTA commercials with a Comcast promo.
There is some question as to whether this part can be modified by a re-transmission agreement, but very few stations would agree to that, anyway.
Otherwise, it's wide open. Cable companies regularly put two HD channels in a single analog slot, and that requires generally re-compression. But, Fox manages to send most of their OTA in less than 10Mbps, because they compress at the network end using very, very expensive equipment, and don't have the local stations re-compress for the actual broadcast (logos are inserted with some tricky MPEG-2 overlay system). DirecTV is using MPEG-4 to re-transmit local digital signals, which is definitely re-compression.
For cable HD, though, it's far worse, as re-scaling from 1920x1080 to 1440x1080 (or less) is common, as is bandwidth starving.
About the only thing that the stations have at their disposal for this is a lawsuit against the cable company, but since most stations foolishly think that we want to see pixelated HD and 3 subchannels of useless stuff, I don't see those lawsuits happening anytime soon. Some of the networks (CBS for one) do protect the quality through affiliate agreements, and that may be the only way that HDTV doesn't turn into "barely better than SD".
There are NO circumstances under which one user should possess another user's password; not even an Administrator.
For VPN (particularly the Cisco VPN client), this isn't 100% true.
The story says these are "departmental passwords", and Cisco VPN has the concept of "group authentication" that drops you into a group of VPN users. To completely connect, you need the group name, the group password, your username, and your password. Where I am, you get the group name and password as part of the admin install of the client, which places it into the connection profile. So, I "know" the same password as a lot of other people (the ones in my group).
It is stored in the connection profile as obfuscated, but it can't be a hash, as it has to be sent as the original when connecting. And, there are utilities that can reverse the obfuscation in a fraction of a second. But, because of the multiple factors, it's not really a large risk, because you have to match up the user with the right group, and then get the user password right. The Cisco VPNs I connect to use an RSA key for the user password, and if that's the case here, it's not a big deal at all to release this info.
So, if these "departmental passwords" are just Cisco VPN group usernames and passwords, it's not much of an issue, as it's pretty easy to get access to them, unless they force their users to memorize them as well as their personal password.
You don't just wave a piece of paper around and get in without being on a list.
You could if it is psychic paper.
You can fix this by changing the setting for keyword.URL in about:config back to a Google search.
Just to follow up on it, I read through the thread and found that Foxconn linked to a page on Microsoft's site which supposedly explains ACPI compliance. Interestingly enough, that page refused to display on anything but IE.
The page does work in Firefox 3 if you use the User Agent Switcher extension to fool it into thinking you have IE (6 or 7).
Then greylisting was implemented, which stopped the spam dead in its tracks, and completely nullified the spam that Double Dribble couldn't stop. That's when I turned over the account to another party. I still use greylisting personally with great success.
For me, between greylisting and requiring strict RFC compliance for the "HELO" parameter, pretty much no spam gets through to even be looked at by SpamAssassin.
For the "HELO" parameter, almost every spambot uses one of:
Neither of these are acceptable (according to section 2.3.5 of the SMTP RFC) as the "HELO" parameter.
Then, I throw out a few more bogus things, like:
At this point, the e-mail gets to face greylisting, ClamAV, and SpamAssassin. About 1 in 100 "bad emails" get through to the end users.
I don't use NoScript, because it turns normal web browsing into a circus of "allow" clicks, and makes UAC look good.
I have to say that I haven't seen this when using NoScript, except on one site that pulled in scripts from its parent domain, the "www.parent" server, and various machines from the company that really owned the site.
Otherwise, it's generally a matter of not having to enable scripting on a site because I do almost all clicking around without it, or else just selecting "Allow example.com" once from the NoScript menu and never worrying about it again. "Temporarily allow all this page" is also handy for just dealing with a site that you might not come back to often, but really need scripting to do something.
It might be because I don't primarily use NoScript as a protection against security issues, but as a speedup tool. The fact that it protects me from random domains is great, but I wouldn't want to use it in such a way that I would be protected from my favorite site getting pwned (i.e., only use "Temporarily Allow" instead of "Allow"). I can't see anybody using NoScript that way, really.
In *my* experience with the new FF I have found the memory leak is still there *but* it appears to be less than before. In prior versions I had issues after just a couple of hours of browsing. Now? It is still eating way more RAM than it should and an undocumented check shows that it steadily increases and doesn't seem to want to let a whole hell of a lot go after closing some tabs.
For me, it's the exact reverse.
Doing the same sort of browsing with Firefox 3 as I did with IE6 (company standard), I find that Firefox uses far less memory. It's using 130MB right now, where before it wasn't unusual to see my "main" IE window using 130MB, with the others taking up another 100MB or so.
I also noticed that every so often IE would start to use the CPU for no reason. Even clicking "Home" (a blank page for me) on all IE windows wouldn't stop it, and it would sit there taking 5-10% of my dual-core 2.8GHz CPU. Only closing the right window (which I couldn't determine in any way other than guessing) would stop the CPU use.
Except for Firefox not playing well with Sharepoint and my virtual credit card number software, and a small bug with an extension, it's been rock solid.
I use IE more than any other browser out there. Yeah... Really. I use it for the add-ons. IE has the add-ons that I want that work how I like them and so I use it.
What add-ons? Part of the reason I wanted to switch to Firefox was the fact that there were so many useful add-ons, but Firefox 2.x just wasn't ready for prime time (at least not for me).