What makes you think they are going to use these things at full power? They've probably already tested them several times on healthy volunteers, and now they want to see if it works in the field. I would assume one of the main reasons they're testing it here is because there's a far less chance of dying if you don't zap someone hard enough (giving you enough time to turn up the power). I'm guessing they're trying to figure out the lowest possible effective setting.
While I like the feature, I do have a rather small friend list and am careful with what I post. Before now there wasn't really any downside to making someone a friend on facebook. Sure, they could see your profile, but it was in its entirety and if they also have several hundred friends then they probably wouldn't notice if you made a change. Now, people are (subconsiously) realizing that it's generally a bad idea to post private information to a public website and to explicitly allow hundreds of people to have access to it. (But this feature will get all the blame and people will grow complacent again when it's "fixed".) I guess that this is the threshold where average joe starts to care about privacy...
Computing possible outcomes is only one of the strategies humans use to make a decision. For chess that may be enough for a decently human-acting AI. But for most games such an AI would either be terrible or impossible to beat (at which point it would be programmed to make mistakes thus go back to being terrible). Throwing more processing power at the problem won't fix it, though it might make the AI slightly better. What's really needed is a better approach to creating AIs.
A learning neural network would be an interesting solution, but has a few problems that make it undesirable IMHO. First, it would start out really bad (when it knew nothing). Now, that could be fixed by training it beforehand, but then you'd run into the second problem faster. The second problem would be that it would soon get to be too good to really be fun for most people. (Hardcore players would probably love it, since they seem to only really care about winning and not about playing, IMHO, but perhaps those are just the ones I've encountered.) You could artifically slow down its learning, but then it wouldn't have much of an advantage over a preprogrammed AI. Third, the AI could progress in an unexpected direction which would of course lead to unusual behaviors. (Like if it found a way to instantly kill the player on sight 100% of the time, or it started practicing "safety in numbers" and all the bots would just hoard together.) Fourth, I doubt a neural network would be very efficient, and I would hope the processor cycles could be better used elsewhere.
A good AI should follow an algorithm similar to what a human would do (plus a fair degree of randomness). It shouldn't be predictable, and it shouldn't be exempt from the rules the player has to follow (omniscient, infinite ammo/mana/stamina, being super strong to compensate for stupidity, etc.). It also should have the rough limitations of a human player (non-zero but realistic and variable reaction time, non-perfect aim, not knowing the exact stats in realtime of the player, not picking the perfect attack every time, etc.). A way to program this would be to have your beta testers explain how they play and program the various tatics into AI algorithms. The AI should use some overarching strategies (pincer attack, various unit combinations, etc.), but it shouldn't use them exclusively (like in Red Alert, the computer would send predictable teams, like 3 Heavy Tanks and a Mammoth Tank, which were ineffective in later game). Perhaps it could even adapt to the player (record and later use player strategies, more frequently use strategies that have worked in the past, do something random to see if it works). While an AI could never play exactly like a human, the difference shouldn't be painfully obvious like it is now. Of course, all of this is my fairly uninformed opinion, and perhaps the AI is better in newer games. I'm just ranting because I've spent litterally thousands of hours playing skirimish missions in various Command and Conquer games and have always wished that I didn't have to use kid's gloves or give it a huge advantage (>3 on 1) when I played it.
You rely on it about as much as you would a lock on your door. Easily defeated, but it keeps honest people honest. I, for one, live on a college campus. I prefer using WPA, but one of my devices doesn't support it. Rather than memorize long WEP keys (as I inevitably have to re-enter them from time to time), I just use MAC filtering. It keeps clueless people off my router, and since there's one "linksys" network in range, I doubt anyone is going to bother trying to get onto mine. The lack of encryption is moot because anything I care about is done over SSL or a VPN, and I expect it would be easier to just sniff my traffic after it leaves my router (college LAN). I would keep my router completely open, but I don't want spyware infested computers connecting at random and infecting my roommate's computer (or mine if I need to boot XP for something). Of course, IMHO, "fingerprinting" a wireless card is kinda pointless. The only valid use for it that I can think of would be for law enforcement as evidence in some computer crime case (since MAC addresses aren't terribly reliable).
That's interesting... By profiling you decrease the frequency of attempted attacks, but increase the probability of success. For example, say TSA agents profiled for Middle Eastern Islamic men. Terrorist groups would then try staging attacks with people that don't fit that description. But, since such people are more of a minority in terrorist groups there should be a far smaller pool of people that aren't going to be profiled and who are willing to die for the cause. So fewer terrorist attacks are attempted. But, for the ones that are attempted there is a greater chance of success. It would be interesting to figure out if profiling reduces the overall number of successful attacks... (I don't know and don't care to speculate.)
Hmm... I was using 150 MB of memory before loading that page and went up to 350 MB. This is on an XP system, 1 GB of RAM, no swap file, large system cache, and IE (between reformats so I got lazy). I'm at just over 1/3 usage, so I don't see the problem. I regularly view huge images and this hasn't been an issue for me except with older versions of Firefox (caused unstability in Windows and images just stopped loading in BeOS, but I only had 512 MB at that time). By "huge images" I mean about 20 tabs with about 15 x 1 MB images. I suppose my setup is worse for memory leaks, but I figure that if my RAM fills up then my pagefile wouldn't be far behind. But, everyone's needs are different, so if you need 20 apps in the tray and play some RAM-hungry games then 1 GB probably wouldn't cut it.
While I agree that one shouldn't do things that they wouldn't want others to know about, I still value privacy over transparancy. There are three main reasons I feel this way. First, information can be misunderstood or give someone false impressions. For example, someone used to be an alcoholic, overcame that years ago, yet doesn't get a job because some interviewer stops reading at the word "alcoholic". Second, information can be used maliciously. I'd prefer that stalking and identity theft be as hard as possible. Third, I may have nothing to hide, but why do you need to know everything about me? I'd rather not leave the possibility of something being misused rather than trust people that I don't know to not misuse it.
MAD is nice while it lasts, but IMHO it's certainly not sustainable. First of all, MAD relies on a massively disproportionate amount of offensive power compared to defensive power. But, any big technological breakthough in defensive technology would eliminate MAD. Secondly, WWII era technology is becoming easier and easier to obtain. So more and more countries will eventually have nukes. Are you sure we should be in MAD with some of them? Say you have some nuclear power with some psycho dictator that is delusional enough, or suicidal enough to launch despite MAD? Or what if some weird circumstances cause a nuke to be launched by mistake (wasn't there a movie about some missile sub almost doing this?). Besides, I'd prefer a military that could protect me, not just get revenge for me after I die.
In most games, gender doesn't affect your stats. On the other hand, it sometimes affects what classes you can be and armors you can wear. Typically, there are more things restricted to females than restricted to males. I suppose the game developers don't want to appear sexist, so they don't make females weaker (like putting a male-only restriction on super heavy armor), but they do tend to include dress-type armors and other female-only equipment (probably more for asthetic reasons and visual diversity). In Chrono Cross, for example, a "child-sized" female character (2/~40) can wear any armor in the game. In Shadowbane there is only one male restricted class, but two female restricted classes and a few female restricted armors. So by using a female character you usually have a slightly better selection of equipment. Plus they're more interesting to stare at for hours on end.
I don't encounter too many kids, so when I saw a ~10 year old get a call on an elevator in my hotel on vaction, I was a bit surprised so I thought about it a bit. Now, I think it's safe to assume that most kids would prefer visiting someone to play rather than just talking to them on a cell phone. So perhaps they are just imitating their parents because they think that having a cell phone glued their ear is normal (sad state of affairs IMHO), or perhaps they can't visit their friends very often. Now this could result from parents being "too busy" now to take their kids places (sleepovers and whatever), or maybe 24 hour news has scared them into preventing their kids from even going outside or using the internet (child predators and whatnot). I think the latter is becoming way too common and almost certainly not a good thing.
As for the former, just the other day I was at some lawyer's house (moving his wireless router to the middle of his house from the basement since that was "too complicated" for him). While I was there some kid stopped by and asked if the laywer's kid was there. The lawyer explained that he used to live in a community that was a little more spread out then his current one. Since no kids were within walking distance of his house his kid seemed to always be the odd-man-out, so they decided to move. Now his kid seems to be popular. But, I suppose that if some family was in that situation and couldn't move then a cell phone is the kid's only method of being sociable after school. Probably not too "healthy", but then again, I'm can't really talk. At ~10 I moved to a more rural area and computers replaced my after school social life.
While it is true that the numbers quoted are for an exhaustive search through the keyspace, the probability of the correct key being early enough in the keyspace for it to be recoverable in a useful amount of time is insignificant. I.e..001% of 100 trillion years is still 100 billion years. Not to mention, the encrypted data has a useful lifespan, I point to the recently cracked civil war messages as an example. Modern crypto algorithms are in use because the entire worldwide community of "smart crypto people" can't come up with a better way (or a practical one) to crack them them faster than bruteforce, which is impractical to the point of impossibility even if you had an "unlimitted budget" (or all the silicon in the universe). The NSA and people do crack crypto, but they know part of the plain text (like a file header or something) and use wordlists specific to the person they caught. If no part of the plaintext is known then bruteforce becomes nearly impossible as well (you don't know when you have the right key because it's all the same to a computer). In twenty or fourty years some crypto mathematician might make a breakthrough and find a vulnerability, but it'll probably take the computers of that time to crack it, and the information will be useless to someone with the resources to crack it.
Progress and bloat aren't the same. Take Azureus and uTorrent for example. Azureus can probably do more, but uTorrent is growing in popularity. I'm not trying to imply that Azureus is bloated necessarily, just that resource usage does matter, even if modern computers have power to spare.
I use BeOS myself even though I have a 3GHz P4 & 512 MB of RAM. It can only do 90% of what I need to do (Linux can probably do 92% though I use XP for the other 10%). The reason I prefer it to XP (or Linux)? It's faster. It's actually spoiled me. I don't have to leave my computer on all the time, or find some way to entertain myself while it boots. I've heard of people getting their cold boot time to under 5 seconds (plus what the BIOS takes). Opening folders takes about half the time as XP or Linux, and launching non-ported programs is also much faster. I also get a response immediately when I start hitting a sequence of hotkeys. None of this, Windows Button, U, H and then waiting another couple seconds before seeing a response. You really don't realize how slow XP is until you use something faster.
If I was paying for my own tuition, I would hate this even more (if possible). Why should a university spend money to help an unrelated commercial entity? I mean, students are paying for an education, not for people and equipment to aid the RIAA.
Re:One Point For Gmail
on
Gmail vs Pine
·
· Score: 3, Interesting
1. But your 200+ GB is on your own hardware. That may be good for privacy concerns (I agree with you completely there), but what happens if it goes down while you're away, or if the harddrive(s) fails? (Or however many it takes for your RAID array to loose data, either way I'm betting Google is more reliable.) That's also 200+ GB that you can't use for something else. Hardware might be cheap, but Gmail is free. (Not to mention that some people don't have a spare computer to run this on.)
2. You still need a telnet client. Since most people don't use them there's a fair chance that some locked-down PC you try to use will let you use a web browser, but not the command line/telnet. Also, if you care about privacy telnet isn't a very good idea (especially since Windows machines don't have anything capable of using the SSL version preinstalled AFAIK).
3 & 4. Same point as above. Also, with Gmail Google is paying for the bandwidth but with Pine you are (cheap as it might be). There's also the issue of your network going down, your ISP doing maintainance, or whatever else.
5. I don't own a cell phone and have never tried/wanted to do check my mail that way.
6. I'm guessing that Gmail does this without any effort on your part (including initial).
Just pointing out a few cons of your approach. On the other hand, I use BeMail myself, so I suppose I shouldn't critize the versitility of other approaches...
And there's an example of the two discreet categories the original poster argued don't exist. A virus, alive or not, would certainly be placed in a different category than, say, plastic. What about mitochondria? I doubt they can live on their own, so are they also in the "non-alive" category? How about our own cells, they (AFAIK) can't live without mitochondria, so wouldn't they be in the same category? (Or perhaps mitochondria are the actual life forms and eukaryotes are the ones that aren't alive.) The point is that the word "life" is binary in that it only has two possible states whereas in nature there seem to be multiple levels. Personally, I'd go even further and say that the word "life" describes an abstract concept where there isn't a natural distinction. I.e. the difference between metabolism and other chemical reactions is merely complexity, which is completely subjective, and therefore no there's not really any distinction.
There are actually quite a few free web hosts, many of which with far better offerings than Geocities or Angelfire. I haven't made a webpage in the last few years, but it looks like the site I used to use to find hosts still works. I never really got why Geocities and Angelfire are so popular when they just offer a small amount of space, ads, and static pages (last time I checked). I mean, if you look you can find free hosts that offer 5 GB of space, others that offer no ads, and still others that offer things like CGI or MySQL.
Hmm... I highly doubt an elementary school kid could immediately switch from Linux to Windows or OS X without stuggling with a bit of a learning curve. Although Linux and Windows are similar, they have enough differences to create difficulties for moderately skilled users. Novices might not notice the differences, and experts should be able to adapt rather quickly. But, children 12 years old and under can't realistically be expected to be at an expert level, but if the school does its job, they shouldn't be novices either. That's why I'm saying they should have experience with Windows.
Windows is, of course, the most likely OS to be used by these students later on, so they should have some experience with it. I avoid using Windows whenever possible anymore, but I certainly value my Windows-specific knowledge. Heck, the first time I tried using a Mac I could barely do anything due to the one button mouse. After probably a dozen failed installations over the last seven or so years I can now use Linux somewhat, but I rarely do anymore. So, primarily, I'm a BeOS user now, but that knowledge is almost useless to me whenever I want to play a game, use most programs, or use any computer but my own. If these children only learn how to use Linux (or some other flavor of unix) then much of that knowledge will be useless when they leave the school. They might be experiences with the open source applications for various tasks, but they would probably struggle with a Windows computer with a normal set of apps (IE, Outlook, MS Office, Windows Media Player, etc.).
As for playing games, well, remember these are 4 - 12 year olds. They aren't using the computers primarily to prepare research essays with graphs of their data or programming. (Well, perhaps the 12 year olds to an extent, but you get my point.) At the lower end of that age group I'd expect them to be playing games and drawing pictures. Maybe a little typing. For the upper end I'd expect them to be using the internet, digital encyclopedias, and typing papers. Playing games is an excellent way to arouse interest in computers. More interest usually means more learning, which I would expect is the point. They might be educational games, but they are games nonetheless. Additionally, the students should have enough supervision to prevent them from playing games when they should be doing something else. (Although, don't most OSes have games and the like?)
Although it almost pains me to say it, I'd think that some of the computers should be left running Windows. After all, it is a school situation, so my guess is that the point of the computers is to familiarize the students with them. Since the vast majority of computers they'll work with in the near-future will be running Windows, they should certainly be exposed to it. Of course, since children tend to learn new skills rather quickly, it would probably also be beneficial to introduce them to Linux. Later, if they need to use a Mac (for example), they'll probably figure out how to use it much quicker than someone who'd only used Windows. Perhaps Linux could be installed on the older computers and Windows left on the newer ones. If the Windows machines have the processor cycles to spare, run a local firewall (restricting traffic based on port and origin/destination) and perhaps something like System Safety Monitor to keep unauthorized programs from running. With these two safeguards, I have my doubts as to whether a virus scanner or spyware detector would be necessary (assumming IE and Outlook aren't used and the students don't have admin rights). Software might be a concern, but teaching the children to use a "Word Processor" rather than just "Microsoft Word 2000" would probably also be beneficial. As for other educational software, I'm not familiar enough with it to comment one way or the other.
That means, no flash and if you want to use javascript then make sure that it works without it. I, for one, middle click on any links that I want to visit, then close the current tab and look at each in turn. It's a lot more convienant than hitting "Back" every page. But with flash this doesn't work (and I care far less about the links sliding in from the side when I load the page than I do about actually using them). Also, if you solely rely on a plugin for navigation, what happens when people don't have that plugin? I use BeOS as my primary OS and guess which popular browser plugins are not availible for it? (BTW, a lot of people also disable those plugins or don't have them installed.)
With javascript use something like: href="blah.html" target="_blank" onclick="window.open(); return: false;". Don't use: href="#" onclick=... or href="javascript:window.open(). (My HTML/JS might be a bit rusty, but you get the idea.) Nothing is more annoying (or confusing the first time it happens) then middle clicking five links and opening the same page or blank pages five times.
Internet radio doesn't seem to be doing too badly to me. I just checked Shoutcast and there are well over 10,000 stations. There's aren't an excessive amount of ads either (at least on the stations I listen to). As for portability, what I do is usually just leave Streamripper on while I go to class, then fill my mp3 player up with the songs later. Probably isn't 100% legal, but it works and I personally don't find it unethical since it doesn't cut out the ads or anything.
Yet another feature that BitComet already has. Sadly, I expect Azureus and uTorrent to ignore this fact and implement their own standard. BitComet version 0.62 or 0.63 will probably conform to it. My point is, why doesn't anyone ever seem to know about BitComet's basic feature set? It's obviously a well known client. In fact, the last swarm I was in it was about equal in popularity to Azureus and BitTornado (only a couple people were using uTorrent, and someone was using the official client). If some feature has a possible exploit (like adding the DHT network as a backup in case the private tracker goes down) then everyone is up in arms about it. The useful features seem to go without notice, like UDP NAT bypass (great if you can't recieve incomming connections), an Intellegent Disk Cache (I WANT my torrent client to use more RAM so hard drive writing frequency is kept reasonable), Packet Header Encryption (the feature in question), the ability to share peer information even if the tracker goes down (implemented long before Azureus added DHT networks), sharing peer information between tracker updates (causes faster downloading), chatting with other BitComet users in the swarm, and others.
So if I were to download a file with the expectation that it was bogus then I'd be in the clear? Example, I see a file "Copyrighted.mp3". I decide to naively believe that nobody would put a copyrighted music file on the internet, so I download it to see what it really is. I realize that it is a copyrighted music file, so I "delete" it. I then seach for another file with the thought that the last download was just a fluke, and the cycle repeats. Since there are files on P2P networks that aren't what they claim to be this seems like it could be a valid defense... The tough part would be convincing people that you were really that naive.
I've been using them for a long time to redirect my e-mail. Anything @mysubdomain.cjb.net goes to me, so I can give everyone a different e-mail address and block any that spam me. Hopefully I won't have to change that now...
BitComet deliberately misreports upload and download amounts to trackers and seeds in order to get the "lion's share" of upload bandwidth from seeders.
I use a private tracker that tracks your share ratio (and my own tracker for transfering files on my LAN), if BitComet lied about the amount it uploads or downloads I'd know about it (and the private trackers would be up in arms about it). Perhaps it lies to seeders or other peers, but I find that unlikely and I am 99.5% sure it doesn't lie to the tracker.
BitComet disconnects and reconnects to download more than is fair via optimistic unchoke -- (which is meant to give new arrivals something to share. Sadly, Azereus is reported to do this too. Automatically droping working connections is hostile activity -- it creates lots of churn which costs extra bandwidth for trackers and peers alike.
I can't confirm or deny that, but if one client does it then the others almost have to follow suit to maintain a similar download speed (which is all most people care about). It may be hostile, but it looks like its becoming a de facto part of the standard, if BitComet and Azureus actually do this.
BitComet seems to favor uploading to other BitComet clients, even when getting faster download speeds from other clients. The most extreme case was a private tracker/torrent on a huge college lan with "100mbps" connections -- the person who did this could download at >5mbps if using BitComet but only ~5-15 KB/sec if using Torrent.
Well, perhaps that has something to do with the various extras BitComet implements. It can defeat both NATs and packetshaping. It also spreads peer information between tracker updates. So obviously it should download faster than clients that don't implement these features. The example you mention was probably a firewalled college student (like my connection). Without incoming connections you usually won't exceed 20 KB/s, but with BitComet's UDP NAT Bypass (only works with other BitComet users) your download proceeds almost as well as someone who wasn't restricted (in a college user's case it can jump from 10 KB/s to probably 2 MB/s or more). Of course, that's the main reason I support BitComet, if you are behind a firewall it can turn a three month download into a three day one, so it helps some people a lot if you run BitComet. Also, it only uses ~15 MB of RAM plus its disk cache, so it's not wasteful like the Java VM and your computer won't lag with all the harddisk activity if you have a decent connection. (Try downloading at ~100 KB/s on a 5 GB torrent onto a USB 1.1 external drive with Windows XP for an extreme example.)
Good think most people are right handed and S, E, and X are all next to each other on the left side of the (QWERTY) keyboard. Talk about good planning (although I'm hesitant to think of ways that might be useful with a typewriter).
Slashdot Mirror
What makes you think they are going to use these things at full power? They've probably already tested them several times on healthy volunteers, and now they want to see if it works in the field. I would assume one of the main reasons they're testing it here is because there's a far less chance of dying if you don't zap someone hard enough (giving you enough time to turn up the power). I'm guessing they're trying to figure out the lowest possible effective setting.
While I like the feature, I do have a rather small friend list and am careful with what I post. Before now there wasn't really any downside to making someone a friend on facebook. Sure, they could see your profile, but it was in its entirety and if they also have several hundred friends then they probably wouldn't notice if you made a change. Now, people are (subconsiously) realizing that it's generally a bad idea to post private information to a public website and to explicitly allow hundreds of people to have access to it. (But this feature will get all the blame and people will grow complacent again when it's "fixed".) I guess that this is the threshold where average joe starts to care about privacy...
Computing possible outcomes is only one of the strategies humans use to make a decision. For chess that may be enough for a decently human-acting AI. But for most games such an AI would either be terrible or impossible to beat (at which point it would be programmed to make mistakes thus go back to being terrible). Throwing more processing power at the problem won't fix it, though it might make the AI slightly better. What's really needed is a better approach to creating AIs.
A learning neural network would be an interesting solution, but has a few problems that make it undesirable IMHO. First, it would start out really bad (when it knew nothing). Now, that could be fixed by training it beforehand, but then you'd run into the second problem faster. The second problem would be that it would soon get to be too good to really be fun for most people. (Hardcore players would probably love it, since they seem to only really care about winning and not about playing, IMHO, but perhaps those are just the ones I've encountered.) You could artifically slow down its learning, but then it wouldn't have much of an advantage over a preprogrammed AI. Third, the AI could progress in an unexpected direction which would of course lead to unusual behaviors. (Like if it found a way to instantly kill the player on sight 100% of the time, or it started practicing "safety in numbers" and all the bots would just hoard together.) Fourth, I doubt a neural network would be very efficient, and I would hope the processor cycles could be better used elsewhere.
A good AI should follow an algorithm similar to what a human would do (plus a fair degree of randomness). It shouldn't be predictable, and it shouldn't be exempt from the rules the player has to follow (omniscient, infinite ammo/mana/stamina, being super strong to compensate for stupidity, etc.). It also should have the rough limitations of a human player (non-zero but realistic and variable reaction time, non-perfect aim, not knowing the exact stats in realtime of the player, not picking the perfect attack every time, etc.). A way to program this would be to have your beta testers explain how they play and program the various tatics into AI algorithms. The AI should use some overarching strategies (pincer attack, various unit combinations, etc.), but it shouldn't use them exclusively (like in Red Alert, the computer would send predictable teams, like 3 Heavy Tanks and a Mammoth Tank, which were ineffective in later game). Perhaps it could even adapt to the player (record and later use player strategies, more frequently use strategies that have worked in the past, do something random to see if it works). While an AI could never play exactly like a human, the difference shouldn't be painfully obvious like it is now. Of course, all of this is my fairly uninformed opinion, and perhaps the AI is better in newer games. I'm just ranting because I've spent litterally thousands of hours playing skirimish missions in various Command and Conquer games and have always wished that I didn't have to use kid's gloves or give it a huge advantage (>3 on 1) when I played it.
You rely on it about as much as you would a lock on your door. Easily defeated, but it keeps honest people honest. I, for one, live on a college campus. I prefer using WPA, but one of my devices doesn't support it. Rather than memorize long WEP keys (as I inevitably have to re-enter them from time to time), I just use MAC filtering. It keeps clueless people off my router, and since there's one "linksys" network in range, I doubt anyone is going to bother trying to get onto mine. The lack of encryption is moot because anything I care about is done over SSL or a VPN, and I expect it would be easier to just sniff my traffic after it leaves my router (college LAN). I would keep my router completely open, but I don't want spyware infested computers connecting at random and infecting my roommate's computer (or mine if I need to boot XP for something). Of course, IMHO, "fingerprinting" a wireless card is kinda pointless. The only valid use for it that I can think of would be for law enforcement as evidence in some computer crime case (since MAC addresses aren't terribly reliable).
That's interesting... By profiling you decrease the frequency of attempted attacks, but increase the probability of success. For example, say TSA agents profiled for Middle Eastern Islamic men. Terrorist groups would then try staging attacks with people that don't fit that description. But, since such people are more of a minority in terrorist groups there should be a far smaller pool of people that aren't going to be profiled and who are willing to die for the cause. So fewer terrorist attacks are attempted. But, for the ones that are attempted there is a greater chance of success. It would be interesting to figure out if profiling reduces the overall number of successful attacks... (I don't know and don't care to speculate.)
Hmm... I was using 150 MB of memory before loading that page and went up to 350 MB. This is on an XP system, 1 GB of RAM, no swap file, large system cache, and IE (between reformats so I got lazy). I'm at just over 1/3 usage, so I don't see the problem. I regularly view huge images and this hasn't been an issue for me except with older versions of Firefox (caused unstability in Windows and images just stopped loading in BeOS, but I only had 512 MB at that time). By "huge images" I mean about 20 tabs with about 15 x 1 MB images. I suppose my setup is worse for memory leaks, but I figure that if my RAM fills up then my pagefile wouldn't be far behind. But, everyone's needs are different, so if you need 20 apps in the tray and play some RAM-hungry games then 1 GB probably wouldn't cut it.
While I agree that one shouldn't do things that they wouldn't want others to know about, I still value privacy over transparancy. There are three main reasons I feel this way. First, information can be misunderstood or give someone false impressions. For example, someone used to be an alcoholic, overcame that years ago, yet doesn't get a job because some interviewer stops reading at the word "alcoholic". Second, information can be used maliciously. I'd prefer that stalking and identity theft be as hard as possible. Third, I may have nothing to hide, but why do you need to know everything about me? I'd rather not leave the possibility of something being misused rather than trust people that I don't know to not misuse it.
MAD is nice while it lasts, but IMHO it's certainly not sustainable. First of all, MAD relies on a massively disproportionate amount of offensive power compared to defensive power. But, any big technological breakthough in defensive technology would eliminate MAD. Secondly, WWII era technology is becoming easier and easier to obtain. So more and more countries will eventually have nukes. Are you sure we should be in MAD with some of them? Say you have some nuclear power with some psycho dictator that is delusional enough, or suicidal enough to launch despite MAD? Or what if some weird circumstances cause a nuke to be launched by mistake (wasn't there a movie about some missile sub almost doing this?). Besides, I'd prefer a military that could protect me, not just get revenge for me after I die.
In most games, gender doesn't affect your stats. On the other hand, it sometimes affects what classes you can be and armors you can wear. Typically, there are more things restricted to females than restricted to males. I suppose the game developers don't want to appear sexist, so they don't make females weaker (like putting a male-only restriction on super heavy armor), but they do tend to include dress-type armors and other female-only equipment (probably more for asthetic reasons and visual diversity). In Chrono Cross, for example, a "child-sized" female character (2/~40) can wear any armor in the game. In Shadowbane there is only one male restricted class, but two female restricted classes and a few female restricted armors. So by using a female character you usually have a slightly better selection of equipment. Plus they're more interesting to stare at for hours on end.
I don't encounter too many kids, so when I saw a ~10 year old get a call on an elevator in my hotel on vaction, I was a bit surprised so I thought about it a bit. Now, I think it's safe to assume that most kids would prefer visiting someone to play rather than just talking to them on a cell phone. So perhaps they are just imitating their parents because they think that having a cell phone glued their ear is normal (sad state of affairs IMHO), or perhaps they can't visit their friends very often. Now this could result from parents being "too busy" now to take their kids places (sleepovers and whatever), or maybe 24 hour news has scared them into preventing their kids from even going outside or using the internet (child predators and whatnot). I think the latter is becoming way too common and almost certainly not a good thing.
As for the former, just the other day I was at some lawyer's house (moving his wireless router to the middle of his house from the basement since that was "too complicated" for him). While I was there some kid stopped by and asked if the laywer's kid was there. The lawyer explained that he used to live in a community that was a little more spread out then his current one. Since no kids were within walking distance of his house his kid seemed to always be the odd-man-out, so they decided to move. Now his kid seems to be popular. But, I suppose that if some family was in that situation and couldn't move then a cell phone is the kid's only method of being sociable after school. Probably not too "healthy", but then again, I'm can't really talk. At ~10 I moved to a more rural area and computers replaced my after school social life.
While it is true that the numbers quoted are for an exhaustive search through the keyspace, the probability of the correct key being early enough in the keyspace for it to be recoverable in a useful amount of time is insignificant. I.e. .001% of 100 trillion years is still 100 billion years. Not to mention, the encrypted data has a useful lifespan, I point to the recently cracked civil war messages as an example. Modern crypto algorithms are in use because the entire worldwide community of "smart crypto people" can't come up with a better way (or a practical one) to crack them them faster than bruteforce, which is impractical to the point of impossibility even if you had an "unlimitted budget" (or all the silicon in the universe). The NSA and people do crack crypto, but they know part of the plain text (like a file header or something) and use wordlists specific to the person they caught. If no part of the plaintext is known then bruteforce becomes nearly impossible as well (you don't know when you have the right key because it's all the same to a computer). In twenty or fourty years some crypto mathematician might make a breakthrough and find a vulnerability, but it'll probably take the computers of that time to crack it, and the information will be useless to someone with the resources to crack it.
Progress and bloat aren't the same. Take Azureus and uTorrent for example. Azureus can probably do more, but uTorrent is growing in popularity. I'm not trying to imply that Azureus is bloated necessarily, just that resource usage does matter, even if modern computers have power to spare.
I use BeOS myself even though I have a 3GHz P4 & 512 MB of RAM. It can only do 90% of what I need to do (Linux can probably do 92% though I use XP for the other 10%). The reason I prefer it to XP (or Linux)? It's faster. It's actually spoiled me. I don't have to leave my computer on all the time, or find some way to entertain myself while it boots. I've heard of people getting their cold boot time to under 5 seconds (plus what the BIOS takes). Opening folders takes about half the time as XP or Linux, and launching non-ported programs is also much faster. I also get a response immediately when I start hitting a sequence of hotkeys. None of this, Windows Button, U, H and then waiting another couple seconds before seeing a response. You really don't realize how slow XP is until you use something faster.
If I was paying for my own tuition, I would hate this even more (if possible). Why should a university spend money to help an unrelated commercial entity? I mean, students are paying for an education, not for people and equipment to aid the RIAA.
1. But your 200+ GB is on your own hardware. That may be good for privacy concerns (I agree with you completely there), but what happens if it goes down while you're away, or if the harddrive(s) fails? (Or however many it takes for your RAID array to loose data, either way I'm betting Google is more reliable.) That's also 200+ GB that you can't use for something else. Hardware might be cheap, but Gmail is free. (Not to mention that some people don't have a spare computer to run this on.)
2. You still need a telnet client. Since most people don't use them there's a fair chance that some locked-down PC you try to use will let you use a web browser, but not the command line/telnet. Also, if you care about privacy telnet isn't a very good idea (especially since Windows machines don't have anything capable of using the SSL version preinstalled AFAIK).
3 & 4. Same point as above. Also, with Gmail Google is paying for the bandwidth but with Pine you are (cheap as it might be). There's also the issue of your network going down, your ISP doing maintainance, or whatever else.
5. I don't own a cell phone and have never tried/wanted to do check my mail that way.
6. I'm guessing that Gmail does this without any effort on your part (including initial).
Just pointing out a few cons of your approach. On the other hand, I use BeMail myself, so I suppose I shouldn't critize the versitility of other approaches...
And there's an example of the two discreet categories the original poster argued don't exist. A virus, alive or not, would certainly be placed in a different category than, say, plastic. What about mitochondria? I doubt they can live on their own, so are they also in the "non-alive" category? How about our own cells, they (AFAIK) can't live without mitochondria, so wouldn't they be in the same category? (Or perhaps mitochondria are the actual life forms and eukaryotes are the ones that aren't alive.) The point is that the word "life" is binary in that it only has two possible states whereas in nature there seem to be multiple levels. Personally, I'd go even further and say that the word "life" describes an abstract concept where there isn't a natural distinction. I.e. the difference between metabolism and other chemical reactions is merely complexity, which is completely subjective, and therefore no there's not really any distinction.
There are actually quite a few free web hosts, many of which with far better offerings than Geocities or Angelfire. I haven't made a webpage in the last few years, but it looks like the site I used to use to find hosts still works. I never really got why Geocities and Angelfire are so popular when they just offer a small amount of space, ads, and static pages (last time I checked). I mean, if you look you can find free hosts that offer 5 GB of space, others that offer no ads, and still others that offer things like CGI or MySQL.
Hmm... I highly doubt an elementary school kid could immediately switch from Linux to Windows or OS X without stuggling with a bit of a learning curve. Although Linux and Windows are similar, they have enough differences to create difficulties for moderately skilled users. Novices might not notice the differences, and experts should be able to adapt rather quickly. But, children 12 years old and under can't realistically be expected to be at an expert level, but if the school does its job, they shouldn't be novices either. That's why I'm saying they should have experience with Windows.
Windows is, of course, the most likely OS to be used by these students later on, so they should have some experience with it. I avoid using Windows whenever possible anymore, but I certainly value my Windows-specific knowledge. Heck, the first time I tried using a Mac I could barely do anything due to the one button mouse. After probably a dozen failed installations over the last seven or so years I can now use Linux somewhat, but I rarely do anymore. So, primarily, I'm a BeOS user now, but that knowledge is almost useless to me whenever I want to play a game, use most programs, or use any computer but my own. If these children only learn how to use Linux (or some other flavor of unix) then much of that knowledge will be useless when they leave the school. They might be experiences with the open source applications for various tasks, but they would probably struggle with a Windows computer with a normal set of apps (IE, Outlook, MS Office, Windows Media Player, etc.).
As for playing games, well, remember these are 4 - 12 year olds. They aren't using the computers primarily to prepare research essays with graphs of their data or programming. (Well, perhaps the 12 year olds to an extent, but you get my point.) At the lower end of that age group I'd expect them to be playing games and drawing pictures. Maybe a little typing. For the upper end I'd expect them to be using the internet, digital encyclopedias, and typing papers. Playing games is an excellent way to arouse interest in computers. More interest usually means more learning, which I would expect is the point. They might be educational games, but they are games nonetheless. Additionally, the students should have enough supervision to prevent them from playing games when they should be doing something else. (Although, don't most OSes have games and the like?)
Although it almost pains me to say it, I'd think that some of the computers should be left running Windows. After all, it is a school situation, so my guess is that the point of the computers is to familiarize the students with them. Since the vast majority of computers they'll work with in the near-future will be running Windows, they should certainly be exposed to it. Of course, since children tend to learn new skills rather quickly, it would probably also be beneficial to introduce them to Linux. Later, if they need to use a Mac (for example), they'll probably figure out how to use it much quicker than someone who'd only used Windows. Perhaps Linux could be installed on the older computers and Windows left on the newer ones. If the Windows machines have the processor cycles to spare, run a local firewall (restricting traffic based on port and origin/destination) and perhaps something like System Safety Monitor to keep unauthorized programs from running. With these two safeguards, I have my doubts as to whether a virus scanner or spyware detector would be necessary (assumming IE and Outlook aren't used and the students don't have admin rights). Software might be a concern, but teaching the children to use a "Word Processor" rather than just "Microsoft Word 2000" would probably also be beneficial. As for other educational software, I'm not familiar enough with it to comment one way or the other.
Corollary: Use basic HTML for navigation menus.
That means, no flash and if you want to use javascript then make sure that it works without it. I, for one, middle click on any links that I want to visit, then close the current tab and look at each in turn. It's a lot more convienant than hitting "Back" every page. But with flash this doesn't work (and I care far less about the links sliding in from the side when I load the page than I do about actually using them). Also, if you solely rely on a plugin for navigation, what happens when people don't have that plugin? I use BeOS as my primary OS and guess which popular browser plugins are not availible for it? (BTW, a lot of people also disable those plugins or don't have them installed.)
With javascript use something like: href="blah.html" target="_blank" onclick="window.open(); return: false;". Don't use: href="#" onclick=... or href="javascript:window.open(). (My HTML/JS might be a bit rusty, but you get the idea.) Nothing is more annoying (or confusing the first time it happens) then middle clicking five links and opening the same page or blank pages five times.
Internet radio doesn't seem to be doing too badly to me. I just checked Shoutcast and there are well over 10,000 stations. There's aren't an excessive amount of ads either (at least on the stations I listen to). As for portability, what I do is usually just leave Streamripper on while I go to class, then fill my mp3 player up with the songs later. Probably isn't 100% legal, but it works and I personally don't find it unethical since it doesn't cut out the ads or anything.
Yet another feature that BitComet already has. Sadly, I expect Azureus and uTorrent to ignore this fact and implement their own standard. BitComet version 0.62 or 0.63 will probably conform to it. My point is, why doesn't anyone ever seem to know about BitComet's basic feature set? It's obviously a well known client. In fact, the last swarm I was in it was about equal in popularity to Azureus and BitTornado (only a couple people were using uTorrent, and someone was using the official client). If some feature has a possible exploit (like adding the DHT network as a backup in case the private tracker goes down) then everyone is up in arms about it. The useful features seem to go without notice, like UDP NAT bypass (great if you can't recieve incomming connections), an Intellegent Disk Cache (I WANT my torrent client to use more RAM so hard drive writing frequency is kept reasonable), Packet Header Encryption (the feature in question), the ability to share peer information even if the tracker goes down (implemented long before Azureus added DHT networks), sharing peer information between tracker updates (causes faster downloading), chatting with other BitComet users in the swarm, and others.
So if I were to download a file with the expectation that it was bogus then I'd be in the clear? Example, I see a file "Copyrighted.mp3". I decide to naively believe that nobody would put a copyrighted music file on the internet, so I download it to see what it really is. I realize that it is a copyrighted music file, so I "delete" it. I then seach for another file with the thought that the last download was just a fluke, and the cycle repeats. Since there are files on P2P networks that aren't what they claim to be this seems like it could be a valid defense... The tough part would be convincing people that you were really that naive.
I've been using them for a long time to redirect my e-mail. Anything @mysubdomain.cjb.net goes to me, so I can give everyone a different e-mail address and block any that spam me. Hopefully I won't have to change that now...
BitComet deliberately misreports upload and download amounts to trackers and seeds in order to get the "lion's share" of upload bandwidth from seeders.
I use a private tracker that tracks your share ratio (and my own tracker for transfering files on my LAN), if BitComet lied about the amount it uploads or downloads I'd know about it (and the private trackers would be up in arms about it). Perhaps it lies to seeders or other peers, but I find that unlikely and I am 99.5% sure it doesn't lie to the tracker.
BitComet disconnects and reconnects to download more than is fair via optimistic unchoke -- (which is meant to give new arrivals something to share. Sadly, Azereus is reported to do this too. Automatically droping working connections is hostile activity -- it creates lots of churn which costs extra bandwidth for trackers and peers alike.
I can't confirm or deny that, but if one client does it then the others almost have to follow suit to maintain a similar download speed (which is all most people care about). It may be hostile, but it looks like its becoming a de facto part of the standard, if BitComet and Azureus actually do this.
BitComet seems to favor uploading to other BitComet clients, even when getting faster download speeds from other clients. The most extreme case was a private tracker/torrent on a huge college lan with "100mbps" connections -- the person who did this could download at >5mbps if using BitComet but only ~5-15 KB/sec if using Torrent.
Well, perhaps that has something to do with the various extras BitComet implements. It can defeat both NATs and packetshaping. It also spreads peer information between tracker updates. So obviously it should download faster than clients that don't implement these features. The example you mention was probably a firewalled college student (like my connection). Without incoming connections you usually won't exceed 20 KB/s, but with BitComet's UDP NAT Bypass (only works with other BitComet users) your download proceeds almost as well as someone who wasn't restricted (in a college user's case it can jump from 10 KB/s to probably 2 MB/s or more). Of course, that's the main reason I support BitComet, if you are behind a firewall it can turn a three month download into a three day one, so it helps some people a lot if you run BitComet. Also, it only uses ~15 MB of RAM plus its disk cache, so it's not wasteful like the Java VM and your computer won't lag with all the harddisk activity if you have a decent connection. (Try downloading at ~100 KB/s on a 5 GB torrent onto a USB 1.1 external drive with Windows XP for an extreme example.)
Good think most people are right handed and S, E, and X are all next to each other on the left side of the (QWERTY) keyboard. Talk about good planning (although I'm hesitant to think of ways that might be useful with a typewriter).