That doesn't sound right to me, but I'm not going to test it because I'd rather to go to bed.
The realm is not a trusted string in any way, shape, or form, and if a browser did automatically hand out your username and password to any site claiming the same "Realm" it should cause quite a stir in the security community. Reasonably, I'd expect browsers to follow the specs you linked to in the Apache docs but only within the same domain.
On the other hand, Basic authentication isn't widely used, so I guess most people wouldn't encounter ill effects of such a "feature", and most browsers only remember passwords based on the domain name anyway. The chances of anyone accessing a legitimate site that uses Basic authentication and then accessing an illegitimate site that happens to use the exact same realm name in the same browser session are pretty remote. Still, it seems a bit too simplistic for the modern web.
I've no idea how old that entry is, but I really do suspect it dates from earlier, simpler times. The server doesn't provide a Last-Modified header and I couldn't see a datestamp anywhere in the file.
No, the children are there by force of law and are not public employees. Why is this so hard for you?
Children are there by the force of their parents. Parents have the option to send their kids to a private school or to home-school them. The only valid distinction you're really making is that they're not public employees. Besides, many of the legal rights of children rest with their parents or guardians.
Regardless, it seems unfair and abusive to subject some people to surveillance while other people interacting with them aren't also subject to the surveillance. It's not that hard to stage a situation and film it from an angle that gives the appearance of particular things happening, or which simply omits important contextual information. What about doctored videos?
You also need to consider the ramifications of making it perfectly legitimate to secretly (or even not so secretly) surveille all public employees. While I do see the point from a "they work for us, so we should be able to see what they're up to" perspective, most people are not comfortable with the idea of being recorded every moment. If private sector employees are exempt from such rules, then it would be incredibly difficulty for the public sector to attract and retain staff.
Most government employees already get paid less for what they do than they would in an equivalent position in the private sector. Most developed nations have trouble getting enough teachers, doctors, nurses, etc. to fill crucial public sector positions. Adding another incentive to stay out of government jobs seems like a really bad idea.
Further, what are you trying to protect with this surveillance? I'm a network administrator for a government department. I work in an office, and rarely interact with people outside of our organisation. Should I be subject to surveillance throughout my workday, or does it only apply to people in a public-facing role (teachers, police, etc.) or to prevent abuse of power?
If it's the latter, all it's going to do is move the bad apples to the private sector where they aren't subject to surveillance. Most public sector jobs with any kind of power or authority over others have a private sector equivalent. If you're really trying to give people the power to protect themselves from abuse by way of being able to record it, it seems strange to have a class of people who have the same power but are immune to that protection.
Seems strange they'd actually bother to sell separate "Hyper-V" and "non Hyper-V" products, given how little they intend to charge for it and that they intend to sell it separately anyway.
Microsoft, however, also plans to sell Hyper-V directly to corporate users who could wipe a server clean and install Hyper-V Server, which is priced at $28 and allows an unlimited number of virtual machines on a single box.
With v6, you've unnecessarily exposed something that you don't exactly want open, and would have to firewall every single device connected to your "wonderful" new world of everything connected to v6.
You seem to be arguing that a NAT router in front of your network is easier than having a firewall with a default deny policy in front of your network. Then you go and point out that pretty much all NAT routers also function as firewalls. So... you're complaining that IPv6 requires you to have a device in front of your network to manage access, which you already have and need for IPv4 NAT anyway.
With a firewall that drops all incoming traffic to addresses you don't want exposed, then nobody has any way of knowing if there's anything active on that address, anyway. You can also argue that it's harder to find interesting things on your network if everything has a different address. Most home users have a single address, and to find interesting things on it you need to portscan that single address. Under IPv6, it's feasible for home users to have hundreds of addresses, of which maybe half a dozen will actually be used, and most of them won't respond to incoming connections anyway.
the solution that is the simplest for the most needs should be the one that is implemented, and that is IPv4, not the insane complexity of v6.
Most people find IPv4 plenty complicated already. People use it because they have to in order to use the internet. When v4 address space is too scarce to be affordable, people will use IPv6 because it's what you need to use. Consumer devices will simplify it enough for normal people to be able to use even if they don't understand what the hell it's about, just like they do now for IPv4.
You can basically turn your argument around and say that what might be necessary for YOUR needs isn't necessary at all for MOST needs, and the solution that is the simplest for the most needs should be the one that is implemented, and that is IPv4, not the insane complexity of v6.
You seem to be arguing that pro-IPv6 people are "elitists" trying to force their own preferences on other people as if they're arrogant and insensitive folk who don't care about others needs or opinions. But at the same time you're espousing your own opinions as if they're the One Ultimate Truth. You don't find cell phones or HD media at all useful, therefore they're simply flawed and anyone that thinks they're useful despite the flaws is somehow "brainwashed". You're also saying that because IPv4 NAT is okay for your needs, it should be enough for anyone.
I think you were modded flamebait for being arrogant and close-minded, not because people simply disagree. Also your point was kind of ignorant, because there's nothing about IPv6 that automatically implies everything will be "exposed", anything more than IPv4 does.
To return to the topic with a point which I think you've missed but might help you understand why some people think the IPv6 promise of "lots of publically routeable addresses for everyone" is useful, consider that most people only have a single v4 address; getting more costs a fair bit, because they're kind of scarce. This will only get worse over time.
Suppose you have something on your home network with a web interface you want to be able to access remotely. So you set up port forwarding on your NAT gateway from port 80 to the device, and you're set. Nice and easy.
Now imagine you get another device which also has a web interface, and which you want to be able to access remotely. So you set up port forwarding from port 80... oops... er... 81 to your new device's port 80, and you're set. Nice and easy, so long as you remember that you need to use www.my.home for device 1's HTTP service and www.my.home:81 for device 2's. Also your work firewall might not allow you to connect to port 81, so maybe you need try a few ports until you find one that's allowed.
I'm fairly sure the IPv4 space is mapped into a part of the IPv6 space, and I think that IPv6-only hosts would be able to convert an IPv4 address (i.e. from an IN A DNS response) into the IPv6 equivalent. This could then be routed to a NAT gateway which would rewrite the packet as an IPv4 one. So it should be doable. But note the "fairly sure" and "I think" and "should".
In practice, everyone that's interested in interoperability (i.e. most people) will use both stacks simultaneously. Most systems that support IPv6 now have both a v4 and v6 address. They typically look for a v6 address first but failing that will fall back to v4 communication.
The phrase "begs the question" has an intuitive meaning which is understood by any English speaker even if they've never encountered the phrase before. The fact that the intuitive meaning is "wrong" is irrelevant. The "correct" meaning is only going to be understood by those who study philosophy or logical fallacies, i.e. a specialised field. It's further complicated by the fact that there are other terms such as "circular argument" which mean much the same thing (and are typically used when people are trying to explain what begging the question "actually" means) and which are understood by most people.
It's similar to the debate over "kilobyte" meaning 1,000 or 1,024 bytes. The sanest thing to do is to treat it as a term which has a special meaning within a particular industry, i.e. when used in the context of computing "kilo" means 1,024, while in other uses it continues to mean 1,000; and make allowances for the fact that people who don't know about the fact it has a different meaning in computing will sometimes use it incorrectly.
It's perfectly reasonable to correct people who use the term incorrectly in a discussion about logic. Correcting laypeople serves no purpose, as the only people who give a crap are those who get a feeling of superiority when they know some technicality that 99% of the population doesn't give a crap about. Most people will continue to use the phrase "incorrectly", because they know that everybody else will understand what they mean, which is the purpose of language.
Just because we don't currently have an algorithm that can defeat spammers, it doesn't mean it's impossible to ever create it. The fact that "even Google" are constantly tweaking their search methods suggests that they don't believe it's as good as it can possibly get, and theirs is just one possible way of indexing the internet. The answer may well in a completely different approach.
My hunch is that the "end game" will be a massively distributed network where individual users index some subset of the internet. Possibly even website operators would index their own sites; which is a nice way of distributing the cost of crawling trillions of internet sites. There's obvious problems here that are even harder to solve than the ones Wikia's approach needs to solve, and I haven't the faintest idea of how to do it. Mind you, I wouldn't have the faintest idea of how to set up a secure transmission method over something as inherently insecure as the internet, but that didn't stop people smarter than me from solving that problem.
The short history of computing suggests that declaring things "impossible" is a sure-fire way to look a fool in the near future when someone does it.
having the code public will make it very easy for the bot herders to control it
This assumes that it's impossible to devise a web search algorithm that can automatically and reliably avoid poisoning. I don't think there's any particular reason to believe that this is true. Humans are quite capable of identifying with a fairly high amount of accuracy what sites are just linkfarms, or useful sites which are the target of a de-ranking campaign, so it should be possible to have a computer automatically do it.
I'm not saying they will succeed, just that it isn't necessarily flawed at a fundamental level, as you seem to imply. It is a difficult problem to solve, but that doesn't mean it can't be solved.
Regardless of that disagreement, it seems the big thing they're relying on is users ranking search results, which is easily gamed if there's money to be made from doing so. So ultimately I do agree with you, but not because the code will be "open".
I don't think games lend themselves to evolution very well. Consider the amount of money and time the commercial studios spend building new game engines virtually from scratch. Game engines tend to be engineered around specifics of gameplay, and while you can modify an engine to suit your particular game, it becomes increasingly difficult to integrate updates to the core engine. And an engine that's even just two or three years old results in a game that looks and feels decidedly dated.
The real problem though is probably more one of leadership. It takes a lot of people working together for one goal to make a game like Crysis. Consider how long Crytek spent developing that, and Far Cry before it. This is a team of very talented people who know exactly what they're doing, but it still took them years of full-time work to create the game.
It's very difficult to get this to happen without paying people, because if you're doing it for free then you want to work on the game that you want, not what someone else wants. Using Crysis single-player as an example (I don't generally like multiplayer shooters): for me, the only really good bit of it was the first time you fight enemies with nano-armour, in the graveyard during extraction. Everything leading up to that was pretty much training for that fight, so I didn't mind it. The rest of the game was waiting for a similarly engaging situation, which never happened. After plodding through the rest of it, your reward is a freaking "boss fight". Double-plus lame-o.
If I'd been working on the game, I'd want to focus more on those particular types of fights. Others would want to focus on fights with the ordinary soldiers. Others might even want more boss fights. If everyone gets to add everything they want, it ends up as a complete mess; not to mention the fact that doing that would require infinite resources. If you tell people no, they'll go sulk and probably stop contributing.
This, combined with the tremendous amount of people-hours that need to be invested into a game before it even becomes playable, makes it very hard for the open source community to develop sophisticated games. Even when they can, they tend to be clones of existing games because that way everyone has an agreed-upon goal to work towards. If it's a new game, then it's likely to be a watered-down, consensus approach to a genre; again, so that people working on it have a common goal. The further you stray from the beaten track and into new territory, the harder it will be to find sufficient people who even understand your goal, much less who share it.
Put another way, how many times have you played a game and thought it was pretty good, but wished they'd focussed more on some particular aspect? Pretty much every game you've ever played, right? Now imagine that every single developer working on the game wanted to focus on their favourite aspect, and they were given leeway to do so. What would the result be like? Would the game ever actually reach a state where it could be released? Probably not.
Contrary to some popular depictions, the pyramid builders were not slaves or foreigners. Excavated skeletons show that they were Egyptians who lived in villages developed and overseen by the pharaoh's supervisors.
The builders' villages boasted bakers, butchers, brewers, granaries, houses, cemeteries, and probably even some sorts of health-care facilities--there is evidence of laborers surviving crushed or amputated limbs. Bakeries excavated near the Great Pyramids could have produced thousands of loaves of bread every week.
One of the more popular theories is that farmers would work on them while the Nile was flooded, as they couldn't do anything with their fields while they were underwater.
I imagine the trailing zeppelins were staggered at ever increasing heights, forcing the pirates to display their acrobatic prowess to ascend to the highest zeppelin of all: the Led Zeppelin.
Needless to say, the ninjas were not amused by the pirates' display of athleticism, and challenged them to a showdown. The ultimate test of wits, skill and bravery, from which only the toughest would emerge alive... Dodgeball!
You do realize that once an area of study has been 'discovered' or 'invented', that other people are allowed to continue to do research in that area, right?
Does it encrypt the data, or just set the folder ACLs so it can't be accessed?
If it's just ACLs, then you can read it from anywhere. Linux's NTFS support ignores ACLs for example, because it's going to have a very hard time trying to make them map to anything sensible. On another Windows box the SUIDs will be unknown but respected, but you should be able to take ownership of the folder and reset the permissions.
As a California taxpayer, is it too unreasonable of me to expect research funded by my tax money to be available freely?
What are you, some kind of communist?
Seriously though, I find it fascinating that they can be so sure the age of the solar system is within such a small (relatively speaking) margin of error. But I'm still a bit sceptical that at some point the theories they've based this on will be disproven. OTOH, IANAA and have no idea how they came up with this age, but even if it seems sound now every so often we discover we didn't actually know something we were sure we knew.
Well, we're not all with Telstra. Most ISPs have their own content mirrors which are unmetered traffic for their customers. While all the ISPs do have to pay Telstra for bandwidth out to the exchanges at typical monopolistic rates, for end-users at other ISPs OOo will continue to be a free download from their ISP's mirror. And Telstra's customers are probably too stupid to know any better anyway (I say probably because there are some cases where using BigPond (Telstra's retail arm) makes sense, but they're few and far between).
But pretty much all you said is right. The liberal government really fucked up in selling Telstra off the way they did.
I expect you're right, but it's interesting to note they're referring to this as Fibre Channel over Ethernet, and not over IP. The reduction in overhead there (not just packet size, but avoiding the whole IP stack) might be enough to really help; and if you're running separate 10 Gigabit Ethernet for the storage subsystem (i.e. not piggy backing on an existing IP network) it might be really nice. Or at least, comparable in performance and a heck of a lot cheaper.
On the other hand, really decent switches that can cope with heavy usage of 10-GigE without delaying packets at all aren't going to be massively cheap, and you'd need very high quality NICs in all the servers as well. Even then, fibre's still probably going to be faster than copper... but that's just something I made up. Maybe someone who knows more about the intricacies of transmitting data over each can enlighten us all?
There was recently an article about "storing" data within fibre as sound rather than converting it to for storage in electrical components, since the latter is kind of slow; how does this compare to transmission via current over copper?
But the hesitation is a pain [...] what's wrong with underlining the damn menu item letter
Agreed; I guess there's a lot of people who don't like having the keys underlined, since they made the default in XP to hiding the accelerator keys in menus until you press Alt. I guess that's the problem with asking Average Joe's for their opinion; you wind up with a product optimised for Average Joe's.
In Office 2007's defence, if you memorize the accelerator key for each tab, it will display the keys for that tab immediately. Some of them are a bit strange I'll admit; why is "View" Alt+W? One interesting thing is the accelerator for the Insert tab is Alt+N. Why not Alt+I? If you press Alt+I, you get a message saying "Office 2003 Access Key: Alt, I - Continue typing the Office 2003 menu key sequence, or press Escape to cancel". So I guess you can use some of the sequences you've memorised from 2003 still. Alt+T (the old Tools menu) has the same behaviour. Might be worth investigating more.
why should I have to ***relearn*** keystrokes?? The whole point of keystrokes is that they're automatic. I will gladly do if it's beneficial, but I have yet to see that this rearrangement benefits *me*, so it's pure cost.
Okay, fair enough, but I guess the "accept all changes" thing is an example of a possible benefit to you: this functionality wasn't previously accessible via keystrokes, but now it is. Maybe it's one you won't use, but there probably are other functions that are now accessible via keystrokes that were previously only on the button toolbar. Now yes, they could have crammed these functions into the menus somewhere, but sometimes it is better to start fresh.
Anyway, next time you get a chance try out some of the Office 2003 shortcut keys you've memorised and see how many are still supported. With any luck you might find yourself pleasantly surprised.
Yep, and not only that but also most of the practical applications of this are already solved using load balancing devices. They can keep a pool of available servers, and allocate new connections to them either by dumb round-robin type behaviour, but also by monitoring how long each takes to respond and using that to gauge their respective load.
There might be some classes of work where the server is actually best placed to determine its ability to perform the task, but for most things it doesn't matter if the server thinks everything's peachy, it's more important what external forces think. The server might be running at low load because its network port is stuffed and is dropping every second packet; an external load balancer will notice the slow response rates and stop sending it requests, while the server itself would just think it's not got much work and try to get more.
You are aware that all the items in the ribbon have Alt+XY shortcuts, aren't you? If not, just hold down your Alt key for a moment, and it'll pop up labels of which key activates which tab. Press one of them (while holding down Alt) and it'll switch the labels to show which key to press to activate the individual functions. If there's stuff you use regularly, it shouldn't be any harder to memorize than menu shortcut keys.
where is the menu item "Accept all changes in document" when you're tracking changes?
As best as I can tell, it's under "Review", "Accept" (in the Changes section), then the last menu item: Accept All Changes in Document. Or, Alt+R, A, D if you'd prefer (Alt+R opens the Review ribbon tab, A opens "Accept", and "D" accepts all changes in the Document.
Btw I don't do much word processing and have never used the track changes feature, it's just not exactly well hidden.
Slashdot Mirror
That doesn't sound right to me, but I'm not going to test it because I'd rather to go to bed.
The realm is not a trusted string in any way, shape, or form, and if a browser did automatically hand out your username and password to any site claiming the same "Realm" it should cause quite a stir in the security community. Reasonably, I'd expect browsers to follow the specs you linked to in the Apache docs but only within the same domain.
On the other hand, Basic authentication isn't widely used, so I guess most people wouldn't encounter ill effects of such a "feature", and most browsers only remember passwords based on the domain name anyway. The chances of anyone accessing a legitimate site that uses Basic authentication and then accessing an illegitimate site that happens to use the exact same realm name in the same browser session are pretty remote. Still, it seems a bit too simplistic for the modern web.
I've no idea how old that entry is, but I really do suspect it dates from earlier, simpler times. The server doesn't provide a Last-Modified header and I couldn't see a datestamp anywhere in the file.
Children are there by the force of their parents. Parents have the option to send their kids to a private school or to home-school them. The only valid distinction you're really making is that they're not public employees. Besides, many of the legal rights of children rest with their parents or guardians.
Regardless, it seems unfair and abusive to subject some people to surveillance while other people interacting with them aren't also subject to the surveillance. It's not that hard to stage a situation and film it from an angle that gives the appearance of particular things happening, or which simply omits important contextual information. What about doctored videos?
You also need to consider the ramifications of making it perfectly legitimate to secretly (or even not so secretly) surveille all public employees. While I do see the point from a "they work for us, so we should be able to see what they're up to" perspective, most people are not comfortable with the idea of being recorded every moment. If private sector employees are exempt from such rules, then it would be incredibly difficulty for the public sector to attract and retain staff.
Most government employees already get paid less for what they do than they would in an equivalent position in the private sector. Most developed nations have trouble getting enough teachers, doctors, nurses, etc. to fill crucial public sector positions. Adding another incentive to stay out of government jobs seems like a really bad idea.
Further, what are you trying to protect with this surveillance? I'm a network administrator for a government department. I work in an office, and rarely interact with people outside of our organisation. Should I be subject to surveillance throughout my workday, or does it only apply to people in a public-facing role (teachers, police, etc.) or to prevent abuse of power?
If it's the latter, all it's going to do is move the bad apples to the private sector where they aren't subject to surveillance. Most public sector jobs with any kind of power or authority over others have a private sector equivalent. If you're really trying to give people the power to protect themselves from abuse by way of being able to record it, it seems strange to have a class of people who have the same power but are immune to that protection.
Seems strange they'd actually bother to sell separate "Hyper-V" and "non Hyper-V" products, given how little they intend to charge for it and that they intend to sell it separately anyway.
Microsoft, however, also plans to sell Hyper-V directly to corporate users who could wipe a server clean and install Hyper-V Server, which is priced at $28 and allows an unlimited number of virtual machines on a single box.http://www.networkworld.com/news/2007/111207-microsoft-virtualization-server.html
I hear they've got the sewing market all stitched up, though.
In an earlier post you said:
With v6, you've unnecessarily exposed something that you don't exactly want open, and would have to firewall every single device connected to your "wonderful" new world of everything connected to v6.
You seem to be arguing that a NAT router in front of your network is easier than having a firewall with a default deny policy in front of your network. Then you go and point out that pretty much all NAT routers also function as firewalls. So... you're complaining that IPv6 requires you to have a device in front of your network to manage access, which you already have and need for IPv4 NAT anyway.
With a firewall that drops all incoming traffic to addresses you don't want exposed, then nobody has any way of knowing if there's anything active on that address, anyway. You can also argue that it's harder to find interesting things on your network if everything has a different address. Most home users have a single address, and to find interesting things on it you need to portscan that single address. Under IPv6, it's feasible for home users to have hundreds of addresses, of which maybe half a dozen will actually be used, and most of them won't respond to incoming connections anyway.
the solution that is the simplest for the most needs should be the one that is implemented, and that is IPv4, not the insane complexity of v6.
Most people find IPv4 plenty complicated already. People use it because they have to in order to use the internet. When v4 address space is too scarce to be affordable, people will use IPv6 because it's what you need to use. Consumer devices will simplify it enough for normal people to be able to use even if they don't understand what the hell it's about, just like they do now for IPv4.
You can basically turn your argument around and say that what might be necessary for YOUR needs isn't necessary at all for MOST needs, and the solution that is the simplest for the most needs should be the one that is implemented, and that is IPv4, not the insane complexity of v6.
You seem to be arguing that pro-IPv6 people are "elitists" trying to force their own preferences on other people as if they're arrogant and insensitive folk who don't care about others needs or opinions. But at the same time you're espousing your own opinions as if they're the One Ultimate Truth. You don't find cell phones or HD media at all useful, therefore they're simply flawed and anyone that thinks they're useful despite the flaws is somehow "brainwashed". You're also saying that because IPv4 NAT is okay for your needs, it should be enough for anyone.
I think you were modded flamebait for being arrogant and close-minded, not because people simply disagree. Also your point was kind of ignorant, because there's nothing about IPv6 that automatically implies everything will be "exposed", anything more than IPv4 does.
To return to the topic with a point which I think you've missed but might help you understand why some people think the IPv6 promise of "lots of publically routeable addresses for everyone" is useful, consider that most people only have a single v4 address; getting more costs a fair bit, because they're kind of scarce. This will only get worse over time.
Suppose you have something on your home network with a web interface you want to be able to access remotely. So you set up port forwarding on your NAT gateway from port 80 to the device, and you're set. Nice and easy.
Now imagine you get another device which also has a web interface, and which you want to be able to access remotely. So you set up port forwarding from port 80... oops... er... 81 to your new device's port 80, and you're set. Nice and easy, so long as you remember that you need to use www.my.home for device 1's HTTP service and www.my.home:81 for device 2's. Also your work firewall might not allow you to connect to port 81, so maybe you need try a few ports until you find one that's allowed.
I'm fairly sure the IPv4 space is mapped into a part of the IPv6 space, and I think that IPv6-only hosts would be able to convert an IPv4 address (i.e. from an IN A DNS response) into the IPv6 equivalent. This could then be routed to a NAT gateway which would rewrite the packet as an IPv4 one. So it should be doable. But note the "fairly sure" and "I think" and "should".
In practice, everyone that's interested in interoperability (i.e. most people) will use both stacks simultaneously. Most systems that support IPv6 now have both a v4 and v6 address. They typically look for a v6 address first but failing that will fall back to v4 communication.
The phrase "begs the question" has an intuitive meaning which is understood by any English speaker even if they've never encountered the phrase before. The fact that the intuitive meaning is "wrong" is irrelevant. The "correct" meaning is only going to be understood by those who study philosophy or logical fallacies, i.e. a specialised field. It's further complicated by the fact that there are other terms such as "circular argument" which mean much the same thing (and are typically used when people are trying to explain what begging the question "actually" means) and which are understood by most people.
It's similar to the debate over "kilobyte" meaning 1,000 or 1,024 bytes. The sanest thing to do is to treat it as a term which has a special meaning within a particular industry, i.e. when used in the context of computing "kilo" means 1,024, while in other uses it continues to mean 1,000; and make allowances for the fact that people who don't know about the fact it has a different meaning in computing will sometimes use it incorrectly.
It's perfectly reasonable to correct people who use the term incorrectly in a discussion about logic. Correcting laypeople serves no purpose, as the only people who give a crap are those who get a feeling of superiority when they know some technicality that 99% of the population doesn't give a crap about. Most people will continue to use the phrase "incorrectly", because they know that everybody else will understand what they mean, which is the purpose of language.
Your sig is out of date!
Sounds reasonable to me. I mean, do you respect stupid people, even if they give you their money?
Just because we don't currently have an algorithm that can defeat spammers, it doesn't mean it's impossible to ever create it. The fact that "even Google" are constantly tweaking their search methods suggests that they don't believe it's as good as it can possibly get, and theirs is just one possible way of indexing the internet. The answer may well in a completely different approach.
My hunch is that the "end game" will be a massively distributed network where individual users index some subset of the internet. Possibly even website operators would index their own sites; which is a nice way of distributing the cost of crawling trillions of internet sites. There's obvious problems here that are even harder to solve than the ones Wikia's approach needs to solve, and I haven't the faintest idea of how to do it. Mind you, I wouldn't have the faintest idea of how to set up a secure transmission method over something as inherently insecure as the internet, but that didn't stop people smarter than me from solving that problem.
The short history of computing suggests that declaring things "impossible" is a sure-fire way to look a fool in the near future when someone does it.
This assumes that it's impossible to devise a web search algorithm that can automatically and reliably avoid poisoning. I don't think there's any particular reason to believe that this is true. Humans are quite capable of identifying with a fairly high amount of accuracy what sites are just linkfarms, or useful sites which are the target of a de-ranking campaign, so it should be possible to have a computer automatically do it.
I'm not saying they will succeed, just that it isn't necessarily flawed at a fundamental level, as you seem to imply. It is a difficult problem to solve, but that doesn't mean it can't be solved.
Regardless of that disagreement, it seems the big thing they're relying on is users ranking search results, which is easily gamed if there's money to be made from doing so. So ultimately I do agree with you, but not because the code will be "open".
Mod parent +1 Funny, but then -1 Didn't Provide A Link
I don't think games lend themselves to evolution very well. Consider the amount of money and time the commercial studios spend building new game engines virtually from scratch. Game engines tend to be engineered around specifics of gameplay, and while you can modify an engine to suit your particular game, it becomes increasingly difficult to integrate updates to the core engine. And an engine that's even just two or three years old results in a game that looks and feels decidedly dated.
The real problem though is probably more one of leadership. It takes a lot of people working together for one goal to make a game like Crysis. Consider how long Crytek spent developing that, and Far Cry before it. This is a team of very talented people who know exactly what they're doing, but it still took them years of full-time work to create the game.
It's very difficult to get this to happen without paying people, because if you're doing it for free then you want to work on the game that you want, not what someone else wants. Using Crysis single-player as an example (I don't generally like multiplayer shooters): for me, the only really good bit of it was the first time you fight enemies with nano-armour, in the graveyard during extraction. Everything leading up to that was pretty much training for that fight, so I didn't mind it. The rest of the game was waiting for a similarly engaging situation, which never happened. After plodding through the rest of it, your reward is a freaking "boss fight". Double-plus lame-o.
If I'd been working on the game, I'd want to focus more on those particular types of fights. Others would want to focus on fights with the ordinary soldiers. Others might even want more boss fights. If everyone gets to add everything they want, it ends up as a complete mess; not to mention the fact that doing that would require infinite resources. If you tell people no, they'll go sulk and probably stop contributing.
This, combined with the tremendous amount of people-hours that need to be invested into a game before it even becomes playable, makes it very hard for the open source community to develop sophisticated games. Even when they can, they tend to be clones of existing games because that way everyone has an agreed-upon goal to work towards. If it's a new game, then it's likely to be a watered-down, consensus approach to a genre; again, so that people working on it have a common goal. The further you stray from the beaten track and into new territory, the harder it will be to find sufficient people who even understand your goal, much less who share it.
Put another way, how many times have you played a game and thought it was pretty good, but wished they'd focussed more on some particular aspect? Pretty much every game you've ever played, right? Now imagine that every single developer working on the game wanted to focus on their favourite aspect, and they were given leeway to do so. What would the result be like? Would the game ever actually reach a state where it could be released? Probably not.
Modern thinking is that the pyramids weren't actually built by slaves.
http://www.nationalgeographic.com/pyramids/pyramids.html#who
Contrary to some popular depictions, the pyramid builders were not slaves or foreigners. Excavated skeletons show that they were Egyptians who lived in villages developed and overseen by the pharaoh's supervisors.The builders' villages boasted bakers, butchers, brewers, granaries, houses, cemeteries, and probably even some sorts of health-care facilities--there is evidence of laborers surviving crushed or amputated limbs. Bakeries excavated near the Great Pyramids could have produced thousands of loaves of bread every week.
One of the more popular theories is that farmers would work on them while the Nile was flooded, as they couldn't do anything with their fields while they were underwater.
I imagine the trailing zeppelins were staggered at ever increasing heights, forcing the pirates to display their acrobatic prowess to ascend to the highest zeppelin of all: the Led Zeppelin.
Needless to say, the ninjas were not amused by the pirates' display of athleticism, and challenged them to a showdown. The ultimate test of wits, skill and bravery, from which only the toughest would emerge alive... Dodgeball!
Not if I patent it they aren't!
Does it encrypt the data, or just set the folder ACLs so it can't be accessed?
If it's just ACLs, then you can read it from anywhere. Linux's NTFS support ignores ACLs for example, because it's going to have a very hard time trying to make them map to anything sensible. On another Windows box the SUIDs will be unknown but respected, but you should be able to take ownership of the folder and reset the permissions.
If it IS encrypted, that's another matter.
What are you, some kind of communist?
Seriously though, I find it fascinating that they can be so sure the age of the solar system is within such a small (relatively speaking) margin of error. But I'm still a bit sceptical that at some point the theories they've based this on will be disproven. OTOH, IANAA and have no idea how they came up with this age, but even if it seems sound now every so often we discover we didn't actually know something we were sure we knew.
Well, we're not all with Telstra. Most ISPs have their own content mirrors which are unmetered traffic for their customers. While all the ISPs do have to pay Telstra for bandwidth out to the exchanges at typical monopolistic rates, for end-users at other ISPs OOo will continue to be a free download from their ISP's mirror. And Telstra's customers are probably too stupid to know any better anyway (I say probably because there are some cases where using BigPond (Telstra's retail arm) makes sense, but they're few and far between).
But pretty much all you said is right. The liberal government really fucked up in selling Telstra off the way they did.
I expect you're right, but it's interesting to note they're referring to this as Fibre Channel over Ethernet, and not over IP. The reduction in overhead there (not just packet size, but avoiding the whole IP stack) might be enough to really help; and if you're running separate 10 Gigabit Ethernet for the storage subsystem (i.e. not piggy backing on an existing IP network) it might be really nice. Or at least, comparable in performance and a heck of a lot cheaper.
On the other hand, really decent switches that can cope with heavy usage of 10-GigE without delaying packets at all aren't going to be massively cheap, and you'd need very high quality NICs in all the servers as well. Even then, fibre's still probably going to be faster than copper... but that's just something I made up. Maybe someone who knows more about the intricacies of transmitting data over each can enlighten us all?
There was recently an article about "storing" data within fibre as sound rather than converting it to for storage in electrical components, since the latter is kind of slow; how does this compare to transmission via current over copper?
Agreed; I guess there's a lot of people who don't like having the keys underlined, since they made the default in XP to hiding the accelerator keys in menus until you press Alt. I guess that's the problem with asking Average Joe's for their opinion; you wind up with a product optimised for Average Joe's.
In Office 2007's defence, if you memorize the accelerator key for each tab, it will display the keys for that tab immediately. Some of them are a bit strange I'll admit; why is "View" Alt+W? One interesting thing is the accelerator for the Insert tab is Alt+N. Why not Alt+I? If you press Alt+I, you get a message saying "Office 2003 Access Key: Alt, I - Continue typing the Office 2003 menu key sequence, or press Escape to cancel". So I guess you can use some of the sequences you've memorised from 2003 still. Alt+T (the old Tools menu) has the same behaviour. Might be worth investigating more.
why should I have to ***relearn*** keystrokes?? The whole point of keystrokes is that they're automatic. I will gladly do if it's beneficial, but I have yet to see that this rearrangement benefits *me*, so it's pure cost.Okay, fair enough, but I guess the "accept all changes" thing is an example of a possible benefit to you: this functionality wasn't previously accessible via keystrokes, but now it is. Maybe it's one you won't use, but there probably are other functions that are now accessible via keystrokes that were previously only on the button toolbar. Now yes, they could have crammed these functions into the menus somewhere, but sometimes it is better to start fresh.
Anyway, next time you get a chance try out some of the Office 2003 shortcut keys you've memorised and see how many are still supported. With any luck you might find yourself pleasantly surprised.
Yep, and not only that but also most of the practical applications of this are already solved using load balancing devices. They can keep a pool of available servers, and allocate new connections to them either by dumb round-robin type behaviour, but also by monitoring how long each takes to respond and using that to gauge their respective load.
There might be some classes of work where the server is actually best placed to determine its ability to perform the task, but for most things it doesn't matter if the server thinks everything's peachy, it's more important what external forces think. The server might be running at low load because its network port is stuffed and is dropping every second packet; an external load balancer will notice the slow response rates and stop sending it requests, while the server itself would just think it's not got much work and try to get more.
But if you put Sony batteries in your suicide robot, you get twice the explosive damage*!
* when compared to leading brands of conventional suicidal followers.
You are aware that all the items in the ribbon have Alt+XY shortcuts, aren't you? If not, just hold down your Alt key for a moment, and it'll pop up labels of which key activates which tab. Press one of them (while holding down Alt) and it'll switch the labels to show which key to press to activate the individual functions. If there's stuff you use regularly, it shouldn't be any harder to memorize than menu shortcut keys.
where is the menu item "Accept all changes in document" when you're tracking changes?As best as I can tell, it's under "Review", "Accept" (in the Changes section), then the last menu item: Accept All Changes in Document. Or, Alt+R, A, D if you'd prefer (Alt+R opens the Review ribbon tab, A opens "Accept", and "D" accepts all changes in the Document.
Btw I don't do much word processing and have never used the track changes feature, it's just not exactly well hidden.
I just realised it's probably Leopard that's stalling during the boot process with a screen that happens to be a shade of blue... sorry.
(OTOH if it leaves the machine unusable, what IS the "big difference"?)