Of course you could do this. But as everyone else has found out the infrastructure costs you money. Your high-speed internet connection incurs a monthly charge as does the local PSTN connection. Then there are the costs of your server equipment and the electricity to run it and I haven't even mentioned your labor costs to maintain it all.
You mean the broadband that I'm already paying for... not a big gamer, not a big music downloader. Except for the occassional linux iso, I don't use it so much. And of course, you mean the POTS line I already pay for, too. Oh, and the electricity I spend on a server that runs 24/7 hosting a few weird projects of mine already... and the labor to keep it running.
Seems like it wouldn't cost me any more than it already does. Like I care if my line is busy, no one calls me anyway (besides, I'd still have priority to kick them off the line if I needed to use it in a hurry).
Here's a clue, you can't "stick it to the phone companies" unless you own all of the wires between the communicating nodes.
The cell phone revolution has put a dent in it, as have other services. But in the US, long distance revenue is still a major cash earner. With local unmetered access, a free service that let's anyone make long distance calls in the US would hurt them.
Dunno if it does decent crypto on its own or not, but even if not... wrap the connection in a openvpn or ipsec tunnel, and be done with it. Of course, the "phone over broadband" commercial service is out of the question at that point.
What I want to know, is if we all hook up our asterisk servers together, and allow them to place local unmetered calls, could we get decent coverage of the US (or even the world) ?
Could we even allow people without computers/internet to call us locally, and route their calls around for them, for free? Would be a neat way to stick it to the phone companies.
(Even better yet, a cell phone plan that allows unlimited calls within the same service...)
Not a whole lot, actually. General Instruments (bought by Motorola) was the dominant player. Scientific Atlanta might have had 15-20% marketshare, if that. And with an incompatible digital cable protocol, it doesn't make much sense to mix up your systems nationwide, so they were getting their asses kicked.
Hell, they're probably worth more for satellite recievers, than their cablebox business, at this point.
Atheros(B and G) here. On a 2.6.x kernel, it was a straightforward compile. It's not bad at all. If you use pansy-assed package managers, I would expect them to work too, no troubles whatsoever. And kismet seems to have at least _some_ support for it, even if I don't know what I'm doing.
However, there doesn't seem to be a way to scan for other networks that doesn't kill its link. This sucks, because I have a very pretty dockapp that can show 4 or 5 APs and their signal strength all simultaneously (god I hate wmwifi and its 1992 cellphone lcd display). Is this just an atheros thing, is it a linux thing, or is it a wirelesstools thing?
I've been playing with X10. Wonder how much power an appliance module draws? For things like printers at least, turning it off completely and allowing it to still respond seems possible.
For a TV though? Even with my littel packard bell IR reciever in the living room to let the server know I'm trying to turn it on, it's just not possible (even allowing for reduced response time). TV's generally don't have a serial port through which they can be turned on, after all.
The profit-maximizing price for a single product, the only one you sell, may be different than the profit-maximizing price for that same product, if its one of 1000s that you sell.
The record labels have 1000s of products. Making an extra half a million on one album/band/act/song is a bad deal, if it undermines your hundreds of other albums/bands/acts/songs. So yes, selling a good song at the profit-maximizing price is a bad idea, if it means you lose control of all your other serf performers. Only having one price to sell any song at, means you can't even manipulate this!
Oh. You haven't seen the bucket of bolts that OO.o is. But I'll forgive that. Let's just make an even harder business case than the one where we farm it out to India.
"Yes Sir. Whenever we have a problem, we'll go around searching for someone who is free to work on this short term to fix the problem. And capable. And willing to accept sums of money as small as $600."
Yes, that's it! Imagine having to stay til 9pm for the conference call to them, only to have them not understand a thing you are trying to relate, because A) the users aren't articulate enough to explain it concisely B) the problem is non-trivial to replicate C) their fluency in english is imperfect and often misses slang/jargon/intonation D) they're programmers, not UI experts.
I'm all for this approach myself, but it probably doesn't make sense this one time. Do they really want to roll out new patches every 2 weeks? Do they want to wait longer and never really see the benefits of having someone fix it for them?
And then there's the whole cost thing. He has to sell this to someone, and they're unlikely to have enough enlightened self-interest to realize anything other than "this is immediately cheaper than M$". For it to be cheaper, they'd have to spend no more than half the savings or so... and $32,000 a year for a OO.o developer wouldn't really buy the kind of guy who could kick ass and make the changes they'd need.
You solution works if they have 20,000 users worldwide and are saving $1,000,000 a year on licenses or something. But not for anything much smaller than that.
So he sees something scary on the box that hints he won't be able to play it. Googling for it tells him very little, and doesn't assuage any fears that it won't play. All it turns up is references to people using the stupid windows software, and how awful it is.
Find me a single howto that says "if you see interblahblah on the dvd box, ignore it, it's still a standard DVD", then yes, he is an idiot. But having looked through the mplayer docs myself just the other night (my old laptop didn't have a DVD drive, so I never bothered to learn the command line option to play one), I can say that such warnings (anti-warnings) didn't exactly jump out at me.
I mean, for christ's sake. If you people aren't bitching about something that's simple that should have been googlable, then you're complaining that the question is too technical, and that if he's asking it he shouldn't be paid to be doing the job. You can't have it both ways, or there are no "ask slashdot" questions left.
And to think I missed downmodding you by less than 24 hours.
Read this again, the things he says indicate that he did search himself, quite possible even on usenet. Definitely on forums somewhere. Sometimes google doesn't come up with anything. Institutionalized bitching from people who hate "ask slashdot" on principle is easily twice as lame as running to someone else whenever you have a technical problem you can't solve yourself.
Why upgrade to vista? Even if new machines will make it faster, won't vista just use some more cpu, starving your machine of what it could have had from it, had you used XP?
It does? Sony did make a few games that only worked on the PS2.
They didn't just rewrite the old games to not work on the PSX, and then slap an "XP" suffix on the title.
Oh, well, that explains the rest of your post, doesn't it.
Funny you should put it like that. I mean, you haven't noticed that word processors were pretty well modern sometime in the middle of the 80s? Or that spreadsheets were not long after? File and print sharing too, for that matter. So just exactly where is the killer app that will take advantage of your 3ghz machine? Game consoles basically have the killer app built into them, the personal computer doesn't.
Ok, that's a little better. Still, when you're adding an extra 4 bytes, there's not much difference (in modern equipment) between that and an extra 12 bytes. Remember, we want more than just unique addresses, we want no fragmentation of space if at all possible. There are lots of other issues to consider too, not to mention that a healthy address range makes things possible we may not be aware of.
For instance, an experimental (IPv4) network of my own embedded geometric locations into the subnets we allocated. Since we were using 10.x.x.x, there was a limit to just what we could embed, we ended up giving out/26s, with six 3bit dimensions. With IPv6, we would have basically been able to choose whatever geometry made the most sense, and not even had to consider extent limitations.
I do have issues with IPv6, but the size of the address isn't one of them.
Yes. Let's shoot for a number that isn't an exponent of 2, and if it is, is definitely not on a byte boundary.
(4 billion, with an extra bit becomes 8 billion, with a second extra, becomes 16 billion, with a third becomes 32 billion... see a pattern here? But even if we give it a whole 'nother byte, a 5 byte address is extremely awkward. I hope you don't program for a living. Dumbass.)
Slashdot Mirror
We're pretty good at those too, actually.
You must be a VoIP newb.
Guilty as charged.
Of course you could do this. But as everyone else has found out the infrastructure costs you money. Your high-speed internet connection incurs a monthly charge as does the local PSTN connection. Then there are the costs of your server equipment and the electricity to run it and I haven't even mentioned your labor costs to maintain it all.
You mean the broadband that I'm already paying for... not a big gamer, not a big music downloader. Except for the occassional linux iso, I don't use it so much. And of course, you mean the POTS line I already pay for, too. Oh, and the electricity I spend on a server that runs 24/7 hosting a few weird projects of mine already... and the labor to keep it running.
Seems like it wouldn't cost me any more than it already does. Like I care if my line is busy, no one calls me anyway (besides, I'd still have priority to kick them off the line if I needed to use it in a hurry).
Here's a clue, you can't "stick it to the phone companies" unless you own all of the wires between the communicating nodes.
The cell phone revolution has put a dent in it, as have other services. But in the US, long distance revenue is still a major cash earner. With local unmetered access, a free service that let's anyone make long distance calls in the US would hurt them.
Dunno if it does decent crypto on its own or not, but even if not... wrap the connection in a openvpn or ipsec tunnel, and be done with it. Of course, the "phone over broadband" commercial service is out of the question at that point.
What I want to know, is if we all hook up our asterisk servers together, and allow them to place local unmetered calls, could we get decent coverage of the US (or even the world) ?
Could we even allow people without computers/internet to call us locally, and route their calls around for them, for free? Would be a neat way to stick it to the phone companies.
(Even better yet, a cell phone plan that allows unlimited calls within the same service...)
I can't get a job outside of helpdesk or desktop support.
Yet, your questions are so dumb as to be insulting. No one would give me a chance at a linux admin position, junior or otherwise.
But just what are they opening? The new XML formats? Or the binary sludge formats?
That says it all. It's shit. I actually liked some of TNG as a series, their best movie was as good as a bad Voyager though...
That's distorted somehow.
Not a whole lot, actually. General Instruments (bought by Motorola) was the dominant player. Scientific Atlanta might have had 15-20% marketshare, if that. And with an incompatible digital cable protocol, it doesn't make much sense to mix up your systems nationwide, so they were getting their asses kicked.
Hell, they're probably worth more for satellite recievers, than their cablebox business, at this point.
Atheros(B and G) here. On a 2.6.x kernel, it was a straightforward compile. It's not bad at all.
If you use pansy-assed package managers, I would expect them to work too, no troubles whatsoever. And kismet seems to have at least _some_ support for it, even if I don't know what I'm doing.
However, there doesn't seem to be a way to scan for other networks that doesn't kill its link. This sucks, because I have a very pretty dockapp that can show 4 or 5 APs and their signal strength all simultaneously (god I hate wmwifi and its 1992 cellphone lcd display). Is this just an atheros thing, is it a linux thing, or is it a wirelesstools thing?
I've been playing with X10. Wonder how much power an appliance module draws? For things like printers at least, turning it off completely and allowing it to still respond seems possible.
For a TV though? Even with my littel packard bell IR reciever in the living room to let the server know I'm trying to turn it on, it's just not possible (even allowing for reduced response time). TV's generally don't have a serial port through which they can be turned on, after all.
The greasemonkey script is supposed to do this for slashdot. Haven't upgraded it yet myself.
The profit-maximizing price for a single product, the only one you sell, may be different than the profit-maximizing price for that same product, if its one of 1000s that you sell.
The record labels have 1000s of products. Making an extra half a million on one album/band/act/song is a bad deal, if it undermines your hundreds of other albums/bands/acts/songs. So yes, selling a good song at the profit-maximizing price is a bad idea, if it means you lose control of all your other serf performers. Only having one price to sell any song at, means you can't even manipulate this!
Oh. You haven't seen the bucket of bolts that OO.o is. But I'll forgive that. Let's just make an even harder business case than the one where we farm it out to India.
"Yes Sir. Whenever we have a problem, we'll go around searching for someone who is free to work on this short term to fix the problem. And capable. And willing to accept sums of money as small as $600."
Yes, that's it! Imagine having to stay til 9pm for the conference call to them, only to have them not understand a thing you are trying to relate, because A) the users aren't articulate enough to explain it concisely B) the problem is non-trivial to replicate C) their fluency in english is imperfect and often misses slang/jargon/intonation D) they're programmers, not UI experts.
That's an easy sell to a boss type.
I'm all for this approach myself, but it probably doesn't make sense this one time. Do they really want to roll out new patches every 2 weeks? Do they want to wait longer and never really see the benefits of having someone fix it for them?
And then there's the whole cost thing. He has to sell this to someone, and they're unlikely to have enough enlightened self-interest to realize anything other than "this is immediately cheaper than M$". For it to be cheaper, they'd have to spend no more than half the savings or so... and $32,000 a year for a OO.o developer wouldn't really buy the kind of guy who could kick ass and make the changes they'd need.
You solution works if they have 20,000 users worldwide and are saving $1,000,000 a year on licenses or something. But not for anything much smaller than that.
Thank God.
Everyone knows that fiercely competitive ecosystems never get slaughtered down to the last species. The internet will survive!
I was thinking the same thing. Is it so hard to glance through Makefile before running 'make install' ?
So he sees something scary on the box that hints he won't be able to play it. Googling for it tells him very little, and doesn't assuage any fears that it won't play. All it turns up is references to people using the stupid windows software, and how awful it is.
Find me a single howto that says "if you see interblahblah on the dvd box, ignore it, it's still a standard DVD", then yes, he is an idiot. But having looked through the mplayer docs myself just the other night (my old laptop didn't have a DVD drive, so I never bothered to learn the command line option to play one), I can say that such warnings (anti-warnings) didn't exactly jump out at me.
I mean, for christ's sake. If you people aren't bitching about something that's simple that should have been googlable, then you're complaining that the question is too technical, and that if he's asking it he shouldn't be paid to be doing the job. You can't have it both ways, or there are no "ask slashdot" questions left.
And to think I missed downmodding you by less than 24 hours.
Read this again, the things he says indicate that he did search himself, quite possible even on usenet. Definitely on forums somewhere. Sometimes google doesn't come up with anything. Institutionalized bitching from people who hate "ask slashdot" on principle is easily twice as lame as running to someone else whenever you have a technical problem you can't solve yourself.
Cool. Running it on 2000 or XP?
Why upgrade to vista? Even if new machines will make it faster, won't vista just use some more cpu, starving your machine of what it could have had from it, had you used XP?
It does? Sony did make a few games that only worked on the PS2.
They didn't just rewrite the old games to not work on the PSX, and then slap an "XP" suffix on the title.
Oh, well, that explains the rest of your post, doesn't it.
Funny you should put it like that. I mean, you haven't noticed that word processors were pretty well modern sometime in the middle of the 80s? Or that spreadsheets were not long after? File and print sharing too, for that matter. So just exactly where is the killer app that will take advantage of your 3ghz machine? Game consoles basically have the killer app built into them, the personal computer doesn't.
Yes. But would you have gotten an Xbox if it still only ran the same games the NES had? Because that's a more accurate analogy than what you used.
I would say that Microsoft ran out of ideas years ago, but in their case, it's more like they couldn't find any more to steal.
Ok, that's a little better. Still, when you're adding an extra 4 bytes, there's not much difference (in modern equipment) between that and an extra 12 bytes. Remember, we want more than just unique addresses, we want no fragmentation of space if at all possible. There are lots of other issues to consider too, not to mention that a healthy address range makes things possible we may not be aware of.
/26s, with six 3bit dimensions. With IPv6, we would have basically been able to choose whatever geometry made the most sense, and not even had to consider extent limitations.
For instance, an experimental (IPv4) network of my own embedded geometric locations into the subnets we allocated. Since we were using 10.x.x.x, there was a limit to just what we could embed, we ended up giving out
I do have issues with IPv6, but the size of the address isn't one of them.
Yes. Let's shoot for a number that isn't an exponent of 2, and if it is, is definitely not on a byte boundary.
(4 billion, with an extra bit becomes 8 billion, with a second extra, becomes 16 billion, with a third becomes 32 billion... see a pattern here? But even if we give it a whole 'nother byte, a 5 byte address is extremely awkward. I hope you don't program for a living. Dumbass.)
False.
There's only 1 electron. The universe just has a very efficient multi-threading library.