Well, I need tools that work on the computer I use. I can assure you that what Microsoft makes can do no better than "works sometimes". Hint: I don't run Windows. Something web based might help, but some people are saying even that is bad.
Why don't you just write up a complete requirements document that specifies everything you want in a calendar scheduling system. Be sure to explain things as if they were the first time ever done. In other words, absolutely do not reference any other software or system. Don't say "do such-and-such the same way as foo-bar does it". Don't make it necessary for the developer to look at any existing implementation or even ever have experienced using one.
Thanks for posting a detailed coverage of the different forms of SSD. I knew about the Flash based disks, but I don't call them as "SSD". If their marketing people do, that's certainly going to be an area of confusion. I can see why they would. I do know that at least some RAM based SSDs will retain everything stored in them across soft reboots, and a few I've heard of have short term capacitor backup that can hold the content for a minute or two powered off (for when you need to do a hard reset).
But in what I was posting, SSD meant to me a RAM based device. I certainly wouldn't consider a Flash based device for swapping purposes. But I have looked at Compact Flash and SD (if someone comes up with an SD-to-SATA interface, I'm probably good to go on that plan) for system images. Given the slow speeds involved, such a system may need to store compressed images there and run special initialization to uncompress and load up an SSD or RAMdisk with the/usr and/opt portions of the root file system.
I have used past swap space formulas based on some percentage of RAM size. I remember when the recommended ratio was at least 20 times RAM. But more recently I've decided it is actually more important to figure it based on factors like extreme peak memory demands (to avoid the OOM killer) and how much swap time one is willing to tolerate to shift what programs are swapped in and out. It can vary from person to person. I still consider that a peak is going to be potentially at least double what one usually ends up running. The ritual formulas are certainly not optimal, but they do keep most people out of trouble, even if wasteful in many cases.
I do agree that once you have more RAM than the running processes need, more gains nothing significant. But I do see periodic spikes in RAM demand, like when I open several Slashdot stories at the same time and follow each through their referenced articles and beyond. Firefox can get bloated, and ironically is more bloated as a single process than if I start multiple instances separately (which doesn't happen if I'm just following the links). With 100 windows open, Firefox is not just running with a large VM, it walks all over that large VM all the time, too. Then when most of those windows get closed down, pieces of memory still appear to be allocated in scattered ways that preclude shrinking that VM footprint (I've never seen FF shrink below 85% of what it is at any time I've made A-B measurements on it).
My ultimate point is that even if RAM based SSD reaches the point where it can run at the best SATA or SCSI speed, and at the same price as the RAM you put in the RAM slots on the mainboard, it's still better to just put the RAM itself in, if you have the capacity for it. It will change the dynamics of the system and the way the needs are calculated. For example real I/O won't have to compete with swapping (since there won't be any swapping). FYI, a couple decades ago when I administered IBM mainframes, we'd often dedicate not just a whole drive or two for swapping, even though it was far more space than ever needed, we'd frequently dedicate a whole channel or two for the purpose, just to manage the I/O contentions (mainframes specialized in lots of I/O overlapping).
Back to a Linux PC. Having no swap at all means you must have enough RAM to handle all the situations where you want to avoid OOM. But you won't readily see just how much you will need because the system should operate rather smoothly right up to the brick wall (as opposed the current level of degraded performance due to heavier and heavier swap work). But you can't just put in "sufficient RAM" and enable swap space you don't expect to need because the system has a propensity to go ahead and do the swapping when I/O output activity gets heavy (trying to use the "enough RAM for process needs" for the potentially unlimited I/O buffering it could do).
One thing I find my computer quite often busy doing is swapping. With only 512MB of RAM, and many bloated programs running, it can't hold everything in RAM all at once. But worse, I find, is when a program is doing a lot of I/O output, which gets buffered in RAM more than it should. If the data being copied is a 40GB HD video file, the assumption that I might be reading the file back in soon (so it should be cached in RAM) just doesn't cut it. An SSD dedicated just for swapping might be faster (eliminates the seeks, but still uses I/O bus bandwidth). And it won't prevent existing pages from being swapped out, requiring them to be swapped back in again (usually a lot sooner than I would be reading those large files back in, which obviously cannot be read in whole).
But is SSD the answer for this (swapping)? If it were significantly cheaper than regular RAM, I might think so. For other uses (live copies of/usr, and such) it certainly could help. What I think is the answer for my case is to go overboard on RAM. My current estimate of normal RAM usage I need for my next computer build (in progress... 1/3 of the parts already purchased) is 2GB. But what I plan to do in this case, however, is go with 8GB of RAM... and not enable any swap space at all. Normally, the amount of swap space I would allocate is the lesser of 1: 2x the RAM... and 2: the amount of data that can be transferred in one direction in 30 seconds. I'm switching to SATA so the latter figure will be larger. Still, the 8GB figure well exceeds the 2GB I expect to need for a while.
Suppose with that 2GB of RAM I deploy 6GB of swap space. That gives me a total of 8GB of space for dirty pages (not counting I/O output buffers which have a destination elsewhere). But during the course of normal use, dirty pages often get forced out to swap because of things like I/O output buffering, which also in turn slows down that I/O (more so if it's in the same disk as the swap space, due to head seek times). Now compare that to 8GB of RAM with no swap space at all. The capacity for keeping dirty pages is the same. But when heavy I/O starts to get pushy, there's no where else for those dirty pages to go (to make room to needlessly overbuffer the I/O). The end result should simply be that the I/O can do nothing more than be written where it belongs as fast as it can (and it can be faster since swapping isn't using up any I/O bus bandwidth nor tying up the disk heads into other locations in the case of non-SSD).
So what else is SSD good for? Maybe for/usr if the price is right. But if SSD is just RAM, bottled up through a SATA/SCSI/IDE/etc, how is that any better than RAM? Is 16GB (high end of what/usr needs for nearly everyone) of SSD cheaper than 16GB of RAM by enough to make it worthwhile? I suspect not, unless the SSD is just using cheap RAM.
For every bit of traffic sent over P2P, that's actually TWO traffic handling bits of cost in the last mile space. Someone somewhere was sending a bit, and someone else somewhere was was receiving that bit. That would not affect things when technology with separate-but-equal lanes exist for each direction (such as a T1, OC-3, etc). Broadband, however, does not have this. While there are separate lanes in most parts, it is not equal. The upstream lanes are quite constrained. Imagine a highway with 3 well paved lanes heading west and a dirt road heading east. That's what we have now.
And we have that because the cable companies, telcos (because they want a piece of the action the cable companies have), and people like Mark Cuban, think strictly in terms of producer/provider vs. consumer/sheep in the model of communications. They don't want to see the community model ever be allowed to work because that means they are nothing more than equal to us when that happens.
Good broadband is faster than T1. Mark Cuban most certainly can afford his own personal T3 or even better than that. If I had the money he has, I would definitely have a couple OC-3's at home. He probably does.
But it is not about that. His complaint is that he doesn't like the "unfair" competition from video distributors that make use of P2P so they don't have to invest in a massive concentration of bandwidth that is capable of delivering video to all who want it, all from the several distribution points they would use. That, and he doesn't like the fact that his own videos would get distributed on P2P without him getting a cut of the action through fees charged and/or ads presented.
His complaints with regard to distribution reserved by copyright ownership is legitimate. But he's asking for all P2P to be blocked, not just that distributing content that is a copyright infringement. All he will end up doing is driving P2P to be more hidden (via IPsec, for example) and indistiguishable from other traffic.
Long ago, people used to play their own music on musical instruments or just sing to entertain themselves at home (the rich could attend a concert). Then along came the record player and people could sit back and rest a few minutes before having to get up and change the record. Then along came the automatic changer that would lift the needle arm and move it aside so the next record could drop and be played automatically. When the CD came along, people had to change CDs manually, but very soon there were automatic CD changers that would let you have somewhere from 5 to 100's of CDs inside. Still, it was a bit of a hassle to manage the CDs in there and the time gap between them was annoying. Not to mention, carrying a portable multi-disk player around was also not practical.
Listening to music was work. It has become less work through the ages, so it is clear that people strive to avoid the work.
Computers and derived products like the iPod changed all that. By copying the music to a larger collective storage such as a hard drive or flash memory, it was possible to not only play the music back with very little work, in many forms it became practical to also carry music around with you. So people started copying their favorite tracks from CDs to their devices. At the same time, music was becoming available online (whether legal or illegal) and people found that to be even more convenient than having to handle and store a bunch of CDs somewhere.
The work associated with listening to music has been reduced significantly by computer technology and the related devices. CD sales are down for a wide range of reasons. The work of copying them to the computer, and the issues in DRM trying to prevent that, are a couple of those factors. CD sales never will recover, even if DRM is removed, although they could go back up some, as EMI has found (people can again easily copy the CD into their computer).
Selling CDs in a store is not the business to be in. The CD is for so many people no longer a playback medium. It is a transfer medium... as is the internet.
If they want extra security, maybe it's because they aren't confident they have enough security now. Is there such a thing as "enough security"? But, I'd click "yes" just to see what the site thinks is "extra security", while distrusting it more just for asking (e.g. why would it have not just used it by default if it was indeed more secure).
Direct connections from Comcast customer addresses, and those of many, many other ISPs, are gracefully refused[1] by thousands, maybe tens of thousands, of mail servers all over the world. The big reason for this are the hundreds of thousands, possibly millions, of zombie computers running some malware that is making connections directly to port 25 to deliver spam in massive quantities that Comcast and others would more readily stop if it were forwarded through their own mail servers. If you send your email through your ISP's mail server, you are distinguishing your mail above that of most spam. If you send it via an outside mail server that is properly configured through some hosting with your own domain name, it will be even more distinguished (because many networks even refuse email from Comcast's mail servers). It is your responsibility to make sure your email stands above the spam (while spammers are trying to make their email look as important as yours).
[1] The TCP connection is actually accepted, the SMTP protocol is carried out, and a message is sent via an SMTP 5XX error code to indicate the mail was refused, and usually why.
It's not so much about what the technology is. Email is NOT dead. It certainly has issues (spam). Instant messaging serves a different purpose than Email.
My grandparents used to send mail (hand written on paper) via the postman, to each other and their other friends, way back long ago. My parents didn't do that... they had telephones!
The important thing to understand here is the difference between communicating and socializing. Email and IM serve their respective needs. Email is lousy at socializing in real time because, being designed as a store and forward system, there can be time delays. And in social circles, a few seconds delay can ruin the whole exchange.
Some day in the future we'll figure out how to do brain implants and exchange thoughts in groups. Kids will be sitting around wherever while their brains will be having mental intercourse. Eventually the whole world will be a single collective.
We should never have needed services like TinyURL. But certain insane webmasters went nuts and started creating URLs that were just way too long. All web sites should use only short and reasonable URLs with the path name part limited to no more than 12 characters. Shorter domain names and shorter email addresses would help, too.
... you have to keep track of everything you allocate and be sure you clean it all up when you are done with it. That need does not go away with C# or other languages like Java. The danger with C is not that you might forget to free some object, but rather, that you do free some object and forget to remove a reference to it.
Either the Boston police are totally incompetent, or the police in all the other cities where these things showed up are totally incompetent (depending on whether there really is a threat or not). My bet is on the former.
Sure, DC makes a lot of sense for combining power because you don't have a phasing issue. OTOH, AC is not that hard to combine in phase. If the source is a motor driven generator, once initially connected in phase, any variation tends to hold it in phase. If the source is electronically inverted AC, the electronics can match the phase quite easily. You still need voltage conversion and control because the solar arrays will vary in voltage by a great amount. The controller would adjust the DC fed to the utility to maintain the current limit your can handle. Uncontrolled attachment of the solar array would more likely just end up with toasted silicon.
But the biggest problem to this idea is that while DC makes some sense for long distance transmission lines between states, it does not (yet) make sense for local area distribution (the wires going around a small town and to your home). Virtually every appliance needs AC just because it is designed that way (DC could be done but everyone would have to swap out so much stuff). Converting DC to AC at each street is not practical at all (the facility needed to house the conversion units would be much larger than the AC transformers in place now). So while DC can reduce the complexity of local power generation attaching to "the grid" somewhat, the complexity of establishing a local distribution in DC is way more costly.
If you build a solar power farm so massively large that you are producing as much as a small generator plant, then setting up your own connection to the transmission grid would make more sense. But it will still be a long, long time before the existing transmission grid network ever converts entirely to DC. The power from your solar farm would either need to find a DC point to connect to or be converted to AC somewhere (closer to the attachment point is better).
AC and DC are affected by the loss due to resistance about equally for the same current. The problem is, until recently, voltage conversion was not practical for DC. AC could do it easily with a transformer. Thus AC could be stepped up to very high voltages for transmission over long distances with much less current. Not only would the current be less, but that lesser voltage loss would be even less significant relative to the high voltage (double the voltage and the effect of the loss is only 1/4 as much).
AC still has some issues with long distance power transmission. Today, we have the technology to do voltage conversion with DC, and to convert between AC and DC readily at these high voltage and power levels. This is making DC transmission lines more practical. Without the voltage conversion issue, DC actually has advantages over AC for long distance transmission, so we'll see more and more of these. OTOH, that's a bad thing, too, as we become more and more dependent, we are also more and more vulnerable to attacks on the power grid.
... there's no way in hell a court would ever hold you liable...
Of course no court would find them liable. But that's not the issue. The issue is that AT&T wants to enter the entertainment business, just like all the other big facility based providers. They want to become your "one stop source" for everything in audio, video, gaming, and reading (and bill you for everything on one big bill). But they face TWO obstacles to that. The content industry has probably already made it clear to them what they must do in order to acquire the right to "broadcast" their content. And they also worry their own revenues might be impacted by the same material coming over from other sources.
Well, OK, it really is a key... it's just not normally a text string in the way it is used. The idea here is that the "key" is a series of selections on a range of images being displayed (in randomized arrangement). Certain images are to be clicked on in a specific sequence that the user has memorized. In a technical sense this really isn't much different than keystrokes in a certain sequence. The images could be of typefaces and the sequence thus be a string of characters. But by having real images of various things, especially like things that are hard to describe the differences in words, then the only way to decrypt the data (if some internal spyware was not involved in the first place) is for the original person to be involved. So they can't just ask you to hand over something like a key... they have to have you actually make the selections for them.
In the USA, the standard voltage is 120 volts. Virtually every computer power supply and these days the power supplies of most auxiliary equipment can operate on 240 volts, either with a flip of a switch, or through autoranging which usually supports 100 volts (as in Japan) through 240 volts (as in Australia, UK, etc). 240 volts (or in some cases 208 volts) is usually available for special circuits using opposite alternating polarities. Most equipment will operate slightly more efficiently on 240 volts. There is also less power loss in the wiring due to the lower average current. While this might only amount to a 2% to 4% overall savings, it is something to consider. Note that the use of a transformer to step the voltage up to 240 from 120 will lose some or all of that savings, so this only works if you and wire the 240 volts up directly.
There are a few problems with this that won't be readily solved unless there is a big market demand for this. One of them is that the way 240 volts arrives to your home in the USA is different than in places like the UK. The USA gets 2 opposite polarities at 120 volts relative to ground that combined have a difference of 240 volts relative to each other. The UK gets 1 polarity of power and a grounded neutral wire. That means things like surge protectors have to be wired differently. So just buying surge protectors from UK will leave you less protected, or in some cases not even work at all. Equipment to handle power distribution and protection with the USA style of 240 volts is hard to find and expensive when you do (usually because it is designed for very large appliances like a 50 amp electric stove, or the whole house). A UPS for this voltage (usually listed as 208 volts because that is what many business offices get since that's the voltage between 2 phases of 3 phase power in the USA) is usually a very large device (3 kVA and up).
It's a chicken and egg problem. The right equipment won't be made unless there is a market. The market won't come about without the right equipment.
There is one advantage of the USA style of 240 volts that neither the 120 volt circuits in the USA nor the 240 volt circuits in Australia an the UK have. This is "balanced power" (an equal but opposite voltage on each wire, relative to ground, and no current on the grounded wire) which greatly reduces the level of hum that can get into audio equipment, especially sensitive sound studio equipment. This can only solve the hum problem for equipment that can use 240 volts (otherwise the hum problem would have to solved using a special balanced 60 volt system that delivers 120 volts).
But we can at least get started despite some of these difficulties. I just wish they would make a 240 volt version of that "Kill A Watt" meter mentioned in the article.
Hart is seeking... a change in the company's advertising to reflect its traffic-shaping practices...
So, we might see the next ad spin this like: "Comcast protects web users from those nasty file sharers that are so frequently overloading the internet".
What means are they using to ensure that no one can forge the IP address (by stuffing the ARP table... which may or may not be involved depending on the way they network this), or forge the MAC address (which might be known to any other computer nearby or maybe even on campus that communicates directly)? I do know ways ensure this, but they are not trivial to deploy (for example, one of them requires special software on the client, and another requires some very specialized hardware on a per client basis). Send some traffic to another host in the same room, or same floor, or same building or same campus. What MAC address is on the result? Sniff on the wire going into the router from outside, not just the ethernet port on your computer. Unless everything going over the wires entering each room are encrypted, it's probably a hole.
I've watched people sniff and collect MAC addresses, wait for these machines to go offline (not all do, but the ones that do become the victims), forge the MAC, and do whatever they wanted to do with no tracking to their MAC. Given the right equipment (not cheap), it could be tracked down to a specific wire, but in a college environment, that's not quite good enough.
This does not mean they have not really secured this tight. But don't believe it just because they say so. Find out how. If they are confident, they won't have a problem telling you how they did it.
A friend of mine had a transmitter that operated on the same frequency as police radar. It wasn't a real radar, so it was a lot smaller and easier to conceal. When driving down the road (and I went with him a couple times), he would turn the transmitter on. It was funny to see somewhere between half and two-thirds of the vehicles ahead of us hit the breaks, as seen by all the break lights suddenly coming on.
I found that a transmitter on a certain frequency would shut off cruise control systems from many years back (a poor design). 50 watts would take them out as much as 100 feet away. It was funny to see a car very very slowly inching its way past just to the left, then suddenly fall way back.
So have they found another even more dangerous toy for us to play with on the highway? Or is this simply a nice solution to install in the back of my car to deal with tailgaters?
Of course you will have these machine printed ballots print all the vote selections in clearly readable text. But in addition to that, also include a copy of all the votes in bar code, along with a secure checksum. Before putting the ballot into the box, scan it on a verification machine. This machine performs optical character reading (OCR) of the text. It compares that to the bar code, and generates a checksum to compare as well. If anything is inconsistent, it reports an error so that vote can be done over. It should also put a red stamp on it. If the ballot is OK, it records an UN-official tally, sending that to a central site over a secure channel, for a master UN-official tally. The UN-official tally can be given to the media for public release right after all the polls close. They would be able to report 100% within seconds of closing (and get back to regular TV programming). In the mean time, the process to count the ballots officially begins. The ballots are shipped in their locked boxes to the central facility under armed guard, where they are counted again by machine scan. The ballots must be kept for the duration of the longest term of office voted in that election. They can be hand counted if ever needed.
Voters will also receive a receipt that prints the time and location of their vote, which ballot printing machine they used, and which vote scanning machine they used. That information plus the vote itself is then securely checksumed and that is printed numerically and in bar code. Every receipt is totally unique. The same info is on each ballot and is to be recorded during the official vote. The list of counted votes (using the same checksum as the receipt) shall be copied to a central computer that can be queried by receipt number to confirm that a vote was counted. The receipt shall NOT contain the actual votes.
Slashdot Mirror
Well, I need tools that work on the computer I use. I can assure you that what Microsoft makes can do no better than "works sometimes". Hint: I don't run Windows. Something web based might help, but some people are saying even that is bad.
Why don't you just write up a complete requirements document that specifies everything you want in a calendar scheduling system. Be sure to explain things as if they were the first time ever done. In other words, absolutely do not reference any other software or system. Don't say "do such-and-such the same way as foo-bar does it". Don't make it necessary for the developer to look at any existing implementation or even ever have experienced using one.
Thanks for posting a detailed coverage of the different forms of SSD. I knew about the Flash based disks, but I don't call them as "SSD". If their marketing people do, that's certainly going to be an area of confusion. I can see why they would. I do know that at least some RAM based SSDs will retain everything stored in them across soft reboots, and a few I've heard of have short term capacitor backup that can hold the content for a minute or two powered off (for when you need to do a hard reset).
But in what I was posting, SSD meant to me a RAM based device. I certainly wouldn't consider a Flash based device for swapping purposes. But I have looked at Compact Flash and SD (if someone comes up with an SD-to-SATA interface, I'm probably good to go on that plan) for system images. Given the slow speeds involved, such a system may need to store compressed images there and run special initialization to uncompress and load up an SSD or RAMdisk with the /usr and /opt portions of the root file system.
I have used past swap space formulas based on some percentage of RAM size. I remember when the recommended ratio was at least 20 times RAM. But more recently I've decided it is actually more important to figure it based on factors like extreme peak memory demands (to avoid the OOM killer) and how much swap time one is willing to tolerate to shift what programs are swapped in and out. It can vary from person to person. I still consider that a peak is going to be potentially at least double what one usually ends up running. The ritual formulas are certainly not optimal, but they do keep most people out of trouble, even if wasteful in many cases.
I do agree that once you have more RAM than the running processes need, more gains nothing significant. But I do see periodic spikes in RAM demand, like when I open several Slashdot stories at the same time and follow each through their referenced articles and beyond. Firefox can get bloated, and ironically is more bloated as a single process than if I start multiple instances separately (which doesn't happen if I'm just following the links). With 100 windows open, Firefox is not just running with a large VM, it walks all over that large VM all the time, too. Then when most of those windows get closed down, pieces of memory still appear to be allocated in scattered ways that preclude shrinking that VM footprint (I've never seen FF shrink below 85% of what it is at any time I've made A-B measurements on it).
My ultimate point is that even if RAM based SSD reaches the point where it can run at the best SATA or SCSI speed, and at the same price as the RAM you put in the RAM slots on the mainboard, it's still better to just put the RAM itself in, if you have the capacity for it. It will change the dynamics of the system and the way the needs are calculated. For example real I/O won't have to compete with swapping (since there won't be any swapping). FYI, a couple decades ago when I administered IBM mainframes, we'd often dedicate not just a whole drive or two for swapping, even though it was far more space than ever needed, we'd frequently dedicate a whole channel or two for the purpose, just to manage the I/O contentions (mainframes specialized in lots of I/O overlapping).
Back to a Linux PC. Having no swap at all means you must have enough RAM to handle all the situations where you want to avoid OOM. But you won't readily see just how much you will need because the system should operate rather smoothly right up to the brick wall (as opposed the current level of degraded performance due to heavier and heavier swap work). But you can't just put in "sufficient RAM" and enable swap space you don't expect to need because the system has a propensity to go ahead and do the swapping when I/O output activity gets heavy (trying to use the "enough RAM for process needs" for the potentially unlimited I/O buffering it could do).
One thing I find my computer quite often busy doing is swapping. With only 512MB of RAM, and many bloated programs running, it can't hold everything in RAM all at once. But worse, I find, is when a program is doing a lot of I/O output, which gets buffered in RAM more than it should. If the data being copied is a 40GB HD video file, the assumption that I might be reading the file back in soon (so it should be cached in RAM) just doesn't cut it. An SSD dedicated just for swapping might be faster (eliminates the seeks, but still uses I/O bus bandwidth). And it won't prevent existing pages from being swapped out, requiring them to be swapped back in again (usually a lot sooner than I would be reading those large files back in, which obviously cannot be read in whole).
But is SSD the answer for this (swapping)? If it were significantly cheaper than regular RAM, I might think so. For other uses (live copies of /usr, and such) it certainly could help. What I think is the answer for my case is to go overboard on RAM. My current estimate of normal RAM usage I need for my next computer build (in progress ... 1/3 of the parts already purchased) is 2GB. But what I plan to do in this case, however, is go with 8GB of RAM ... and not enable any swap space at all. Normally, the amount of swap space I would allocate is the lesser of 1: 2x the RAM ... and 2: the amount of data that can be transferred in one direction in 30 seconds. I'm switching to SATA so the latter figure will be larger. Still, the 8GB figure well exceeds the 2GB I expect to need for a while.
Suppose with that 2GB of RAM I deploy 6GB of swap space. That gives me a total of 8GB of space for dirty pages (not counting I/O output buffers which have a destination elsewhere). But during the course of normal use, dirty pages often get forced out to swap because of things like I/O output buffering, which also in turn slows down that I/O (more so if it's in the same disk as the swap space, due to head seek times). Now compare that to 8GB of RAM with no swap space at all. The capacity for keeping dirty pages is the same. But when heavy I/O starts to get pushy, there's no where else for those dirty pages to go (to make room to needlessly overbuffer the I/O). The end result should simply be that the I/O can do nothing more than be written where it belongs as fast as it can (and it can be faster since swapping isn't using up any I/O bus bandwidth nor tying up the disk heads into other locations in the case of non-SSD).
So what else is SSD good for? Maybe for /usr if the price is right. But if SSD is just RAM, bottled up through a SATA/SCSI/IDE/etc, how is that any better than RAM? Is 16GB (high end of what /usr needs for nearly everyone) of SSD cheaper than 16GB of RAM by enough to make it worthwhile? I suspect not, unless the SSD is just using cheap RAM.
For every bit of traffic sent over P2P, that's actually TWO traffic handling bits of cost in the last mile space. Someone somewhere was sending a bit, and someone else somewhere was was receiving that bit. That would not affect things when technology with separate-but-equal lanes exist for each direction (such as a T1, OC-3, etc). Broadband, however, does not have this. While there are separate lanes in most parts, it is not equal. The upstream lanes are quite constrained. Imagine a highway with 3 well paved lanes heading west and a dirt road heading east. That's what we have now.
And we have that because the cable companies, telcos (because they want a piece of the action the cable companies have), and people like Mark Cuban, think strictly in terms of producer/provider vs. consumer/sheep in the model of communications. They don't want to see the community model ever be allowed to work because that means they are nothing more than equal to us when that happens.
Good broadband is faster than T1. Mark Cuban most certainly can afford his own personal T3 or even better than that. If I had the money he has, I would definitely have a couple OC-3's at home. He probably does.
But it is not about that. His complaint is that he doesn't like the "unfair" competition from video distributors that make use of P2P so they don't have to invest in a massive concentration of bandwidth that is capable of delivering video to all who want it, all from the several distribution points they would use. That, and he doesn't like the fact that his own videos would get distributed on P2P without him getting a cut of the action through fees charged and/or ads presented.
His complaints with regard to distribution reserved by copyright ownership is legitimate. But he's asking for all P2P to be blocked, not just that distributing content that is a copyright infringement. All he will end up doing is driving P2P to be more hidden (via IPsec, for example) and indistiguishable from other traffic.
Long ago, people used to play their own music on musical instruments or just sing to entertain themselves at home (the rich could attend a concert). Then along came the record player and people could sit back and rest a few minutes before having to get up and change the record. Then along came the automatic changer that would lift the needle arm and move it aside so the next record could drop and be played automatically. When the CD came along, people had to change CDs manually, but very soon there were automatic CD changers that would let you have somewhere from 5 to 100's of CDs inside. Still, it was a bit of a hassle to manage the CDs in there and the time gap between them was annoying. Not to mention, carrying a portable multi-disk player around was also not practical.
Listening to music was work. It has become less work through the ages, so it is clear that people strive to avoid the work.
Computers and derived products like the iPod changed all that. By copying the music to a larger collective storage such as a hard drive or flash memory, it was possible to not only play the music back with very little work, in many forms it became practical to also carry music around with you. So people started copying their favorite tracks from CDs to their devices. At the same time, music was becoming available online (whether legal or illegal) and people found that to be even more convenient than having to handle and store a bunch of CDs somewhere.
The work associated with listening to music has been reduced significantly by computer technology and the related devices. CD sales are down for a wide range of reasons. The work of copying them to the computer, and the issues in DRM trying to prevent that, are a couple of those factors. CD sales never will recover, even if DRM is removed, although they could go back up some, as EMI has found (people can again easily copy the CD into their computer).
Selling CDs in a store is not the business to be in. The CD is for so many people no longer a playback medium. It is a transfer medium ... as is the internet.
And the DVD is headed in the same direction.
If they want extra security, maybe it's because they aren't confident they have enough security now. Is there such a thing as "enough security"? But, I'd click "yes" just to see what the site thinks is "extra security", while distrusting it more just for asking (e.g. why would it have not just used it by default if it was indeed more secure).
... "Senator Hunt 2". Your weapon is a big fat briefcase of cash.
... the New York Times drop their stupid login requirement?
Direct connections from Comcast customer addresses, and those of many, many other ISPs, are gracefully refused[1] by thousands, maybe tens of thousands, of mail servers all over the world. The big reason for this are the hundreds of thousands, possibly millions, of zombie computers running some malware that is making connections directly to port 25 to deliver spam in massive quantities that Comcast and others would more readily stop if it were forwarded through their own mail servers. If you send your email through your ISP's mail server, you are distinguishing your mail above that of most spam. If you send it via an outside mail server that is properly configured through some hosting with your own domain name, it will be even more distinguished (because many networks even refuse email from Comcast's mail servers). It is your responsibility to make sure your email stands above the spam (while spammers are trying to make their email look as important as yours).
[1] The TCP connection is actually accepted, the SMTP protocol is carried out, and a message is sent via an SMTP 5XX error code to indicate the mail was refused, and usually why.
It's not so much about what the technology is. Email is NOT dead. It certainly has issues (spam). Instant messaging serves a different purpose than Email.
My grandparents used to send mail (hand written on paper) via the postman, to each other and their other friends, way back long ago. My parents didn't do that ... they had telephones!
The important thing to understand here is the difference between communicating and socializing. Email and IM serve their respective needs. Email is lousy at socializing in real time because, being designed as a store and forward system, there can be time delays. And in social circles, a few seconds delay can ruin the whole exchange.
Some day in the future we'll figure out how to do brain implants and exchange thoughts in groups. Kids will be sitting around wherever while their brains will be having mental intercourse. Eventually the whole world will be a single collective.
We should never have needed services like TinyURL. But certain insane webmasters went nuts and started creating URLs that were just way too long. All web sites should use only short and reasonable URLs with the path name part limited to no more than 12 characters. Shorter domain names and shorter email addresses would help, too.
... you have to keep track of everything you allocate and be sure you clean it all up when you are done with it. That need does not go away with C# or other languages like Java. The danger with C is not that you might forget to free some object, but rather, that you do free some object and forget to remove a reference to it.
Either the Boston police are totally incompetent, or the police in all the other cities where these things showed up are totally incompetent (depending on whether there really is a threat or not). My bet is on the former.
Not exactly.
Sure, DC makes a lot of sense for combining power because you don't have a phasing issue. OTOH, AC is not that hard to combine in phase. If the source is a motor driven generator, once initially connected in phase, any variation tends to hold it in phase. If the source is electronically inverted AC, the electronics can match the phase quite easily. You still need voltage conversion and control because the solar arrays will vary in voltage by a great amount. The controller would adjust the DC fed to the utility to maintain the current limit your can handle. Uncontrolled attachment of the solar array would more likely just end up with toasted silicon.
But the biggest problem to this idea is that while DC makes some sense for long distance transmission lines between states, it does not (yet) make sense for local area distribution (the wires going around a small town and to your home). Virtually every appliance needs AC just because it is designed that way (DC could be done but everyone would have to swap out so much stuff). Converting DC to AC at each street is not practical at all (the facility needed to house the conversion units would be much larger than the AC transformers in place now). So while DC can reduce the complexity of local power generation attaching to "the grid" somewhat, the complexity of establishing a local distribution in DC is way more costly.
If you build a solar power farm so massively large that you are producing as much as a small generator plant, then setting up your own connection to the transmission grid would make more sense. But it will still be a long, long time before the existing transmission grid network ever converts entirely to DC. The power from your solar farm would either need to find a DC point to connect to or be converted to AC somewhere (closer to the attachment point is better).
AC and DC are affected by the loss due to resistance about equally for the same current. The problem is, until recently, voltage conversion was not practical for DC. AC could do it easily with a transformer. Thus AC could be stepped up to very high voltages for transmission over long distances with much less current. Not only would the current be less, but that lesser voltage loss would be even less significant relative to the high voltage (double the voltage and the effect of the loss is only 1/4 as much).
AC still has some issues with long distance power transmission. Today, we have the technology to do voltage conversion with DC, and to convert between AC and DC readily at these high voltage and power levels. This is making DC transmission lines more practical. Without the voltage conversion issue, DC actually has advantages over AC for long distance transmission, so we'll see more and more of these. OTOH, that's a bad thing, too, as we become more and more dependent, we are also more and more vulnerable to attacks on the power grid.
I'm hoping for 1920 x 1200 progressive at 75 fps ... because my monitor can handle it :-)
I guess that's not going to be 3 months away :-(
Of course no court would find them liable. But that's not the issue. The issue is that AT&T wants to enter the entertainment business, just like all the other big facility based providers. They want to become your "one stop source" for everything in audio, video, gaming, and reading (and bill you for everything on one big bill). But they face TWO obstacles to that. The content industry has probably already made it clear to them what they must do in order to acquire the right to "broadcast" their content. And they also worry their own revenues might be impacted by the same material coming over from other sources.
Well, OK, it really is a key ... it's just not normally a text string in the way it is used. The idea here is that the "key" is a series of selections on a range of images being displayed (in randomized arrangement). Certain images are to be clicked on in a specific sequence that the user has memorized. In a technical sense this really isn't much different than keystrokes in a certain sequence. The images could be of typefaces and the sequence thus be a string of characters. But by having real images of various things, especially like things that are hard to describe the differences in words, then the only way to decrypt the data (if some internal spyware was not involved in the first place) is for the original person to be involved. So they can't just ask you to hand over something like a key ... they have to have you actually make the selections for them.
In the USA, the standard voltage is 120 volts. Virtually every computer power supply and these days the power supplies of most auxiliary equipment can operate on 240 volts, either with a flip of a switch, or through autoranging which usually supports 100 volts (as in Japan) through 240 volts (as in Australia, UK, etc). 240 volts (or in some cases 208 volts) is usually available for special circuits using opposite alternating polarities. Most equipment will operate slightly more efficiently on 240 volts. There is also less power loss in the wiring due to the lower average current. While this might only amount to a 2% to 4% overall savings, it is something to consider. Note that the use of a transformer to step the voltage up to 240 from 120 will lose some or all of that savings, so this only works if you and wire the 240 volts up directly.
There are a few problems with this that won't be readily solved unless there is a big market demand for this. One of them is that the way 240 volts arrives to your home in the USA is different than in places like the UK. The USA gets 2 opposite polarities at 120 volts relative to ground that combined have a difference of 240 volts relative to each other. The UK gets 1 polarity of power and a grounded neutral wire. That means things like surge protectors have to be wired differently. So just buying surge protectors from UK will leave you less protected, or in some cases not even work at all. Equipment to handle power distribution and protection with the USA style of 240 volts is hard to find and expensive when you do (usually because it is designed for very large appliances like a 50 amp electric stove, or the whole house). A UPS for this voltage (usually listed as 208 volts because that is what many business offices get since that's the voltage between 2 phases of 3 phase power in the USA) is usually a very large device (3 kVA and up).
It's a chicken and egg problem. The right equipment won't be made unless there is a market. The market won't come about without the right equipment.
There is one advantage of the USA style of 240 volts that neither the 120 volt circuits in the USA nor the 240 volt circuits in Australia an the UK have. This is "balanced power" (an equal but opposite voltage on each wire, relative to ground, and no current on the grounded wire) which greatly reduces the level of hum that can get into audio equipment, especially sensitive sound studio equipment. This can only solve the hum problem for equipment that can use 240 volts (otherwise the hum problem would have to solved using a special balanced 60 volt system that delivers 120 volts).
But we can at least get started despite some of these difficulties. I just wish they would make a 240 volt version of that "Kill A Watt" meter mentioned in the article.
So, we might see the next ad spin this like: "Comcast protects web users from those nasty file sharers that are so frequently overloading the internet".
What means are they using to ensure that no one can forge the IP address (by stuffing the ARP table ... which may or may not be involved depending on the way they network this), or forge the MAC address (which might be known to any other computer nearby or maybe even on campus that communicates directly)? I do know ways ensure this, but they are not trivial to deploy (for example, one of them requires special software on the client, and another requires some very specialized hardware on a per client basis). Send some traffic to another host in the same room, or same floor, or same building or same campus. What MAC address is on the result? Sniff on the wire going into the router from outside, not just the ethernet port on your computer. Unless everything going over the wires entering each room are encrypted, it's probably a hole.
I've watched people sniff and collect MAC addresses, wait for these machines to go offline (not all do, but the ones that do become the victims), forge the MAC, and do whatever they wanted to do with no tracking to their MAC. Given the right equipment (not cheap), it could be tracked down to a specific wire, but in a college environment, that's not quite good enough.
This does not mean they have not really secured this tight. But don't believe it just because they say so. Find out how. If they are confident, they won't have a problem telling you how they did it.
A friend of mine had a transmitter that operated on the same frequency as police radar. It wasn't a real radar, so it was a lot smaller and easier to conceal. When driving down the road (and I went with him a couple times), he would turn the transmitter on. It was funny to see somewhere between half and two-thirds of the vehicles ahead of us hit the breaks, as seen by all the break lights suddenly coming on.
I found that a transmitter on a certain frequency would shut off cruise control systems from many years back (a poor design). 50 watts would take them out as much as 100 feet away. It was funny to see a car very very slowly inching its way past just to the left, then suddenly fall way back.
So have they found another even more dangerous toy for us to play with on the highway? Or is this simply a nice solution to install in the back of my car to deal with tailgaters?
Of course you will have these machine printed ballots print all the vote selections in clearly readable text. But in addition to that, also include a copy of all the votes in bar code, along with a secure checksum. Before putting the ballot into the box, scan it on a verification machine. This machine performs optical character reading (OCR) of the text. It compares that to the bar code, and generates a checksum to compare as well. If anything is inconsistent, it reports an error so that vote can be done over. It should also put a red stamp on it. If the ballot is OK, it records an UN-official tally, sending that to a central site over a secure channel, for a master UN-official tally. The UN-official tally can be given to the media for public release right after all the polls close. They would be able to report 100% within seconds of closing (and get back to regular TV programming). In the mean time, the process to count the ballots officially begins. The ballots are shipped in their locked boxes to the central facility under armed guard, where they are counted again by machine scan. The ballots must be kept for the duration of the longest term of office voted in that election. They can be hand counted if ever needed.
Voters will also receive a receipt that prints the time and location of their vote, which ballot printing machine they used, and which vote scanning machine they used. That information plus the vote itself is then securely checksumed and that is printed numerically and in bar code. Every receipt is totally unique. The same info is on each ballot and is to be recorded during the official vote. The list of counted votes (using the same checksum as the receipt) shall be copied to a central computer that can be queried by receipt number to confirm that a vote was counted. The receipt shall NOT contain the actual votes.