The DRM is in there so that I can play my legitimately purchased content at full resolution, which is something you can't do with your box. That's an interesting take on the situation. Of course, there's no technical reason that legitimately purchased content cannot play at full resolution. There's just an artificial reason--DRM.
Microsoft didn't have to play their game. They chose to. Whenever the ICT flag gets set in a movie, I won't be able to use my monitor to view high definition video through Vista, should I desire to. Despite the fact that it would decode just fine, if not for DRM. Despite the fact that it would display just fine, if not for DRM.
Even then, there isn't much in the way of media center software, and what there is isn't really that great. I can't stand MythTV's interface. LinuxMCE is even worse. To me, the only one that shows any promise whatsoever is the Linux port of XBMC, but it's pretty far from having a release.
The product lines (not from the beginning, obviously):
Windows 3.1 -> Windows 95 -> Windows 98 -> Windows ME -> Windows XP Windows NT 3.51 -> Windows NT 4.0 -> Windows 2000 -> Windows XP
Although Windows 2000 was really quite usable for most home users (compared to NT4, especially), it was not considered a home-user OS. That niche was filled by Windows ME. The OSs were even released within the same year (about 6 months apart in 2000, if I remember correctly.) XP came just a year later.
In the general case, it makes sense for companies to charge what the market will bear. This means setting your prices so that your customer base (n) times your charge per customer (m) minus your operating costs (c) is maximized. That's what these companies are doing--they believe that the market will bear an increased bandwidth cost for some people (and presumably the same cost for everyone else), so they're increasing prices.
Of course, the reason that people aren't absolutely gouged in every aspect of financial life is because most markets have competition. That means that by lowering costs just a little, a given company will attract more customers, getting higher profits than if they were charging the same rate as their competitors. Most cable internet providers, however, either have monopolies or are offering such a highly differentiated service that they effectively have monopolies (most DSL doesn't get nearly to the raw speeds advertised by cable providers, plus they tend to get much more tightly integrated services, which customers like for some reason.) There is no business incentive for a monopoly to charge less than what will maximize their profits.
This is why laissez-faire economics is good in theory, but in reality, at least some government regulation is necessary if you believe that price gouging should not be allowed, and that a certain standard of living should be available to everyone. Society benefits when its members are educated and can communicate freely. The Internet helps this. Price gouging Internet access hinders it.
Man, I thought everyone understood this issue by now.
The problem is that if you're saturating your uplink with lots of data, you have a harder time sending your ACKs. ACKs get thrown in with all of the rest of your upload, but because your uplink is so small compared to your downlink, and you're probably sending very large packets upstream, to boot, you can't ACK as many downstream packets, and TCP/IP congestion control throttles your connection.
Read up on ack prioritization for more information. Many 3rd-party firmware home routers will let you prioritize ACK packets on your uplink, so that if it has the choice between sending an ACK and sending anything else, it will send the ACK first. Because they're so small, you'll probably never even notice the difference in your upload, but your downloads will be much faster when you're otherwise saturated.
If you don't stream, you can get a pretty decent estimate. Streaming services make it extremely hard to estimate how much you're actually using, and you'd really need a meter of some sort to figure it out.
Look at the recent history of Internet providers--they want high-profile sites to pay them for giving their customers access. The cable companies want Google to pay so that people can get to GMail. They want Apple to pay so that people can get to the iTunes Music Store.
Now, apparently, they also want their own customers to pay for accessing these services. They're trying to get paid on both ends. At a bare minimum, it's avarice. When you notice that these companies are also offering competing services, it is essential that we ask the question of whether or not they are abusing their monopoly.
I think that it's unlikely that my provider will go to this method. I'm on DSL, and we have a fairly low throughput anyway. But if they do, I'll be finding an alternative solution, simply on principle, and I'm not even a very high bandwidth user.
The bonus for a la carte pricing is that perhaps viruses will suddenly start hurting people where they feel it the most--their wallets. A single node in a highly active Storm Worm partition can use more than 4 gigabytes over the course of a month.
You might have one DSL company too (if you can really call that broadband). Man, that's spoiled. Broadband on the Internet used to be anything over ISDN.
Of course, the real definition means signaling in a wider frequency spectrum than other signaling over the same medium. DSL is certainly broadband, because the other signaling is POTS. Using this definition of the term means that comparing cable Internet access to DSL is meaningless.
Regardless, it's pretty common knowledge that monopolies are bad for the consumer, and that's why we have governments to manage monopolies. All of those capital-L Libertarians need to realize this.
I don't know of any good books, but I do know that Postgresql has been changing quite a bit lately (or rather, adding useful and interesting features.) What are you looking for in a book? Is the online documentation not good enough?
Yeah, I used to be one of those idiot users. The database was nothing more than a datastore, and WHERE was nothing more than a handy way of filtering results. It's a really common thing, as learning SQL syntax is easier than writing your own data-handling routines.
But man, once you get the relational DB bug, it bites hard. I wouldn't touch MySQL with a 10 foot pole these days. Even with InnoDB, it's just not as good as Postgresql.
Value additions for using my specific service? Inertia? (AIM didn't start out by serving ads, you know.) Inertia specifically counts for a lot, as once you have your JID, you won't want to give it up (just as I wouldn't want to give up my e-mail address, even if my provider started doing things that I didn't like.)
I've messed with the xmpp extension to irssi. I know that it's a work-in-progress, but it seems very incomplete. There's virtually no documentation, either, and I was pretty disconcerted with the apparent lack of encryption (I don't know if there really wasn't encryption, or if the messages just aren't clear (No certificate found? Certificate not trusted on a server with a thawte cert?)
Likewise, this protocol seems like it was designed for humans XML is intended to be human readable, presumably for debugging purposes (though possibly also for other reasons.) Making your protocol human-readable doesn't mean that it's a bad protocol--it means that it had goals which apparently differs from yours.
There are problems with doing this. First, it means that you're running an additional application on your phone. Second is that to receive notifications, the Jabber client must have an open connection to the server full-time. That's going to be a drain on the battery.
I actually have a smartphone with a Jabber client, and since I primarily use IRC, I wrote some scripts to glue messaging between xmpp-irssi and irc (mostly a gateway between the two so that I could bridge my IRC session to my phone.) The drain was immediately noticeable, and I've switched from having it always-on to only turning it on when I specifically want to "IRC" from my phone. Of course, this means that I don't get realtime MSG notifications, but them's the breaks.
Of course, for my case, it would be trivial to send an SMS notification to my phone when I receive a message, thus allowing me to hop on IRC. But then I'm still using SMS (albeit fewer.)
XMPP is a simple, if deep protocol. Most libraries implement it at a fairly low level--they don't give you a function which sends a message to a specific jid; instead, they give you functions to create a message packet and send that on. They provide functions to send the IQ and Presence packets, but require that you explicitly do so rather than taking care of that upon connect, and offering functions to let you change your status.
Learning to use the libraries, then, requires reading the protocol specifications.
There's not usually a good financial incentive for companies to provide IM, unless they're providing it in a proprietary way so that they can serve advertising (which is basically the only currently known way to "monetize" otherwise free Internet services.) As such, interoperability is rarely desirable, as you want anyone who talks to people on your network to have to sign up on your network, allowing you to serve ads.
In fact, the only reason I can see for a company to move to XMPP (or to include an XMPP gateway) is because they're losing users. If AOL started losing users to Google because more of those people's contacts were using Google, it would make sense to implement XMPP to try to stem the tide. Keep them on AOL clients, they can still talk to Google, everyone's (more or less) happy (and it looks like this is the direction that AOL is going.)
From there, it could go two ways--people from other networks could start defecting to an XMPP-enabled network, thus causing all networks to go XMPP, or the largest of the networks might balk, dividing the userbase. If Microsoft succeeds in buying Yahoo, then they will have a massive network of IM users compared to AOL/Google. They'll integrate Yahoo IM into MSN, and AOL/Google will become a toy network. You'll need an account on MSN/Yahoo to talk to the vast majority of users that use that network, and slowly, MSN will end up the winner.
Well, happy IMing on unencrypted, stone-age, propertiary networks that force-feed you with ads and censor your messages, if that's what you want. XML doesn't solve any of these problems (and they're not all problems.) There's no technical reason that any given messaging service couldn't use SSL, and XMPP is extensible, and an implementation of it can be made proprietary enough to require a client that will force-feed you ads. Any network can censor messages, assuming they can read them.
Your post is overrated.
Yeah, I know, this is Slashdot, where people like to spew completely uninformed pseudo-opinions, but this one is just too obvious. Oh, sorry, I guess you covered all of that.
The total bandwidth used by a phone is pretty small, even when it's active. When the phone is inactive, the beacons it sends and receives are trivial.
Compare to trying to stream live TV at "prime time" (after work when everyone finally gets home.) Compare to bittorrent. The amount of data that personal computers send is pretty high, and increasing all the time. Technical innovation can help somewhat, but you eventually hit a saturation point. Reducing range would certainly go a long way, but I wonder how feasible it is.
I'd like to see wireless devices that reduce their power based upon transmission needs. If it's sitting on the desk with the AP, it obviously doesn't need to be transmitting at full power. If the AP polled long-range periodically, and nothing responded, it obviously doesn't need to use full power. It's possible that some devices do this, but they all don't yet.
Hey, VGA and USB are also common accross the industry, so why not the power as well. It's partially inertia, I'm sure, but also the fact that each computer maker doesn't want to compete in every peripheral's market. Dell doesn't want to compete on external hard drives, external TV cards, external fingerprint readers, external mice, external keyboards, external keyboard lights, etc, particularly when doing so would preclude providing industry standard ports on their devices. However it's pretty easy to keep proprietary batteries and cables simply because there is no industry standard that a large percentage of the industry has agreed upon.
Peripheral support with standard connectors has been a staple of PC computing for a very long time. Before laptops even came out, computers had serial ports, and it made sense (when creating laptops) to provide one. However batteries are somewhat unique to portable computers, and it makes sense to use a proprietary one to discourage third-party competition.
When you are on the bottom, it means, "This is less relevant than the higher results" (or more accurately, our algorithm considers this to be less relevant.) If it was irrelevant, it shouldn't be returned at all.
Regardless, neither result is fact. One result is more deterministic and less prone to human error, but that doesn't mean that it isn't still subjective.
No one will want to risk that, if nothing else, due to the sometimes capricious nature of the courts. Under this plan, expect patent examiners to be in short supply.
It probably keeps the whole redirect logic on the server. They could redirect dynamically ala Google Ads, based upon the URI. I don't understand why it couldn't be done on the server itself, though, as it should have access to all of the header information (and more) that the client has.
As far as software patents go, it's not all that bad. But that's like saying that as far as serial killers go, Charles Manson wasn't all that bad.
Slashdot Mirror
Microsoft didn't have to play their game. They chose to. Whenever the ICT flag gets set in a movie, I won't be able to use my monitor to view high definition video through Vista, should I desire to. Despite the fact that it would decode just fine, if not for DRM. Despite the fact that it would display just fine, if not for DRM.
Not the original poster.
Even then, there isn't much in the way of media center software, and what there is isn't really that great. I can't stand MythTV's interface. LinuxMCE is even worse. To me, the only one that shows any promise whatsoever is the Linux port of XBMC, but it's pretty far from having a release.
The product lines (not from the beginning, obviously):
Windows 3.1 -> Windows 95 -> Windows 98 -> Windows ME -> Windows XP
Windows NT 3.51 -> Windows NT 4.0 -> Windows 2000 -> Windows XP
Although Windows 2000 was really quite usable for most home users (compared to NT4, especially), it was not considered a home-user OS. That niche was filled by Windows ME. The OSs were even released within the same year (about 6 months apart in 2000, if I remember correctly.) XP came just a year later.
In the general case, it makes sense for companies to charge what the market will bear. This means setting your prices so that your customer base (n) times your charge per customer (m) minus your operating costs (c) is maximized. That's what these companies are doing--they believe that the market will bear an increased bandwidth cost for some people (and presumably the same cost for everyone else), so they're increasing prices.
Of course, the reason that people aren't absolutely gouged in every aspect of financial life is because most markets have competition. That means that by lowering costs just a little, a given company will attract more customers, getting higher profits than if they were charging the same rate as their competitors. Most cable internet providers, however, either have monopolies or are offering such a highly differentiated service that they effectively have monopolies (most DSL doesn't get nearly to the raw speeds advertised by cable providers, plus they tend to get much more tightly integrated services, which customers like for some reason.) There is no business incentive for a monopoly to charge less than what will maximize their profits.
This is why laissez-faire economics is good in theory, but in reality, at least some government regulation is necessary if you believe that price gouging should not be allowed, and that a certain standard of living should be available to everyone. Society benefits when its members are educated and can communicate freely. The Internet helps this. Price gouging Internet access hinders it.
Man, I thought everyone understood this issue by now.
The problem is that if you're saturating your uplink with lots of data, you have a harder time sending your ACKs. ACKs get thrown in with all of the rest of your upload, but because your uplink is so small compared to your downlink, and you're probably sending very large packets upstream, to boot, you can't ACK as many downstream packets, and TCP/IP congestion control throttles your connection.
Read up on ack prioritization for more information. Many 3rd-party firmware home routers will let you prioritize ACK packets on your uplink, so that if it has the choice between sending an ACK and sending anything else, it will send the ACK first. Because they're so small, you'll probably never even notice the difference in your upload, but your downloads will be much faster when you're otherwise saturated.
If you don't stream, you can get a pretty decent estimate. Streaming services make it extremely hard to estimate how much you're actually using, and you'd really need a meter of some sort to figure it out.
Look at the recent history of Internet providers--they want high-profile sites to pay them for giving their customers access. The cable companies want Google to pay so that people can get to GMail. They want Apple to pay so that people can get to the iTunes Music Store.
Now, apparently, they also want their own customers to pay for accessing these services. They're trying to get paid on both ends. At a bare minimum, it's avarice. When you notice that these companies are also offering competing services, it is essential that we ask the question of whether or not they are abusing their monopoly.
I think that it's unlikely that my provider will go to this method. I'm on DSL, and we have a fairly low throughput anyway. But if they do, I'll be finding an alternative solution, simply on principle, and I'm not even a very high bandwidth user.
The bonus for a la carte pricing is that perhaps viruses will suddenly start hurting people where they feel it the most--their wallets. A single node in a highly active Storm Worm partition can use more than 4 gigabytes over the course of a month.
Of course, the real definition means signaling in a wider frequency spectrum than other signaling over the same medium. DSL is certainly broadband, because the other signaling is POTS. Using this definition of the term means that comparing cable Internet access to DSL is meaningless.
Regardless, it's pretty common knowledge that monopolies are bad for the consumer, and that's why we have governments to manage monopolies. All of those capital-L Libertarians need to realize this.
I don't know of any good books, but I do know that Postgresql has been changing quite a bit lately (or rather, adding useful and interesting features.) What are you looking for in a book? Is the online documentation not good enough?
Yeah, I used to be one of those idiot users. The database was nothing more than a datastore, and WHERE was nothing more than a handy way of filtering results. It's a really common thing, as learning SQL syntax is easier than writing your own data-handling routines.
But man, once you get the relational DB bug, it bites hard. I wouldn't touch MySQL with a 10 foot pole these days. Even with InnoDB, it's just not as good as Postgresql.
Value additions for using my specific service? Inertia? (AIM didn't start out by serving ads, you know.) Inertia specifically counts for a lot, as once you have your JID, you won't want to give it up (just as I wouldn't want to give up my e-mail address, even if my provider started doing things that I didn't like.)
I've messed with the xmpp extension to irssi. I know that it's a work-in-progress, but it seems very incomplete. There's virtually no documentation, either, and I was pretty disconcerted with the apparent lack of encryption (I don't know if there really wasn't encryption, or if the messages just aren't clear (No certificate found? Certificate not trusted on a server with a thawte cert?)
There are problems with doing this. First, it means that you're running an additional application on your phone. Second is that to receive notifications, the Jabber client must have an open connection to the server full-time. That's going to be a drain on the battery.
I actually have a smartphone with a Jabber client, and since I primarily use IRC, I wrote some scripts to glue messaging between xmpp-irssi and irc (mostly a gateway between the two so that I could bridge my IRC session to my phone.) The drain was immediately noticeable, and I've switched from having it always-on to only turning it on when I specifically want to "IRC" from my phone. Of course, this means that I don't get realtime MSG notifications, but them's the breaks.
Of course, for my case, it would be trivial to send an SMS notification to my phone when I receive a message, thus allowing me to hop on IRC. But then I'm still using SMS (albeit fewer.)
XMPP is a simple, if deep protocol. Most libraries implement it at a fairly low level--they don't give you a function which sends a message to a specific jid; instead, they give you functions to create a message packet and send that on. They provide functions to send the IQ and Presence packets, but require that you explicitly do so rather than taking care of that upon connect, and offering functions to let you change your status.
Learning to use the libraries, then, requires reading the protocol specifications.
There's not usually a good financial incentive for companies to provide IM, unless they're providing it in a proprietary way so that they can serve advertising (which is basically the only currently known way to "monetize" otherwise free Internet services.) As such, interoperability is rarely desirable, as you want anyone who talks to people on your network to have to sign up on your network, allowing you to serve ads.
In fact, the only reason I can see for a company to move to XMPP (or to include an XMPP gateway) is because they're losing users. If AOL started losing users to Google because more of those people's contacts were using Google, it would make sense to implement XMPP to try to stem the tide. Keep them on AOL clients, they can still talk to Google, everyone's (more or less) happy (and it looks like this is the direction that AOL is going.)
From there, it could go two ways--people from other networks could start defecting to an XMPP-enabled network, thus causing all networks to go XMPP, or the largest of the networks might balk, dividing the userbase. If Microsoft succeeds in buying Yahoo, then they will have a massive network of IM users compared to AOL/Google. They'll integrate Yahoo IM into MSN, and AOL/Google will become a toy network. You'll need an account on MSN/Yahoo to talk to the vast majority of users that use that network, and slowly, MSN will end up the winner.
What do you recommend for a good XMPP client?
Your post is overrated. Yeah, I know, this is Slashdot, where people like to spew completely uninformed pseudo-opinions, but this one is just too obvious. Oh, sorry, I guess you covered all of that.
The total bandwidth used by a phone is pretty small, even when it's active. When the phone is inactive, the beacons it sends and receives are trivial.
Compare to trying to stream live TV at "prime time" (after work when everyone finally gets home.) Compare to bittorrent. The amount of data that personal computers send is pretty high, and increasing all the time. Technical innovation can help somewhat, but you eventually hit a saturation point. Reducing range would certainly go a long way, but I wonder how feasible it is.
I'd like to see wireless devices that reduce their power based upon transmission needs. If it's sitting on the desk with the AP, it obviously doesn't need to be transmitting at full power. If the AP polled long-range periodically, and nothing responded, it obviously doesn't need to use full power. It's possible that some devices do this, but they all don't yet.
Peripheral support with standard connectors has been a staple of PC computing for a very long time. Before laptops even came out, computers had serial ports, and it made sense (when creating laptops) to provide one. However batteries are somewhat unique to portable computers, and it makes sense to use a proprietary one to discourage third-party competition.
When you are on the bottom, it means, "This is less relevant than the higher results" (or more accurately, our algorithm considers this to be less relevant.) If it was irrelevant, it shouldn't be returned at all.
Regardless, neither result is fact. One result is more deterministic and less prone to human error, but that doesn't mean that it isn't still subjective.
No one will want to risk that, if nothing else, due to the sometimes capricious nature of the courts. Under this plan, expect patent examiners to be in short supply.
It probably keeps the whole redirect logic on the server. They could redirect dynamically ala Google Ads, based upon the URI. I don't understand why it couldn't be done on the server itself, though, as it should have access to all of the header information (and more) that the client has.
As far as software patents go, it's not all that bad. But that's like saying that as far as serial killers go, Charles Manson wasn't all that bad.
You know, the "I'm going to file for X-OBVIOUS-PATENT" comments are really, really tired.