I believe one of the reasons for the drop in sales is that they are leaving the country.
There isn't a decent online retailer in Canada. Chapters? Took them 4 weeks to ship me a DVD (to ship it, it still hasn't arrived). This is after cancelling the order 3 times arbitrarily, and asking for a photocopy of both my card and statement. No, I didn't provide it. Sam The Record Man? Closed down.
So, I buy stuff from Amazon. They've never flubbed an order, or failed to ship. What does that mean? Yep, my sales don't show up on the counts for Canada. It doesn't mean that I'm not buying, I'm just not buying Canadian.
Motherboard - Advantech 5864L - video in/out, network and built in sound.
Riser card 2xPCI.
SBLive (so I can have digital audio!)
DVD decoder card (if you need that, I don't)
Add a small hard disk, a read-only linux image, and you are good to go. The hardest part will probably be getting the video out to work under linux. I haven't been able to find any references online.:)
They even make a small case and power supply,
here, but the case is a little ugly. The power supply doesn't appear to have a fan though.
I'm using an earlier version of the board as the basis for my car-based mp3 system. Works really well. Can't say anything about the PSU, it became available after I started with my system.
If ht://Dig does most of what you want, why not make it understand file:// URL's if it doesn't now? I wouldn't expect it to be _that_ difficult an addition, especially if you are concerned about the availability of source code.
Otherwise, why not just pay up the US$2k and buy a Glimpse site license? US$2k is probably worth about _two_ days of your time to the company (depending on the size of the company). That would cover everything as long as you kept it to one site.
Glimpse also appears to be available with source too, check out the bottom of the download page.
If the tool is to be used to create a commercial product that you are going to sell, then you won't be covered by that. Personally, if I was going to be using a GPL'ed (or other open source) tool as the core of a product I was going to sell, I would feel ethically bound to pay some money for it, regardless of it's availability for "free".
I realise that, but there isn't anything keeping them from saying that if you adhere to the standard, use this implementation, then you will interoperate. Other languages are able to work this way. Java does, C forces everyone to because they have to link in system libraries.
Different revisions of the standard could still change the name mangling to force link failure, but if you are both on the same revision, I would expect it to work.
If the system libraries were implemented in C++, don't you think we would have an implementation standard? Yep, me too.:)
Can you tell that this is currently a major beef with me?:) I'm having to write two hunks of C code, passing void *'s around because I can't link two libraries together, and I don't have the source code for the Forte one.
What I would like to see is an implementation standard for C++.
At least some way that I could link C++ code generated by GCC with others, such as Sun Forte. We use GCC for platform independence, but we are continually having to write stupid little C function wrappers around classes to get around all of the linking issues.
Then I might be able to do some code re-use at more than just the source level.
Jason Pollock
Re:Killing two birds with one stone
on
Fission in a Box
·
· Score: 2
Secondly, anything that uses water (heavy or light) as a moderator *can't* melt down. Without a moderator, the reactor stops dead (it needs the moderator to react). With water as a moderator, your moderator disappears as soon as it heats up enough to burst pipes. End of reaction.
I used to believe this too. Of course, like many things, the truth is a little more complicated. When I tried this line on a physicist friend of mine (who used to work for the AECL), she sat me down and smacked me around.:)
While the heavy water moderator is required to initiate the reaction (the rods are initially less than critical mass), once started, the rods end up enriching themselves, and may be able to sustain a reaction on their own.
Also, even if the reaction stops, the rods may generate enough heat to melt anyways. This is because a good fraction of the heat generated in the plant is by the decay of fission products. You can't stop the decay no matter what you do.
Of course, this is probably incredibly unlikely. But that just means you have to roll the dice more often.:)
Furthermore, it is officially illegal to route any network
traffic originating in Canada and destined for a location also in Canada, through any foreign country including the U.S.
This was done to prevent US carriers from simply putting POPs in Canada and routing traffic through the US, undercutting Canadian suppliers. This regulation is primarily for long distance (voice) carriers. It does appear to have achieved what it set out to do, which was force new entrants to the networking market to build their own networks, rather than route through their networks in the US. And yes, this rule _is_ enforced when it contravenes the _intent_ of the regulation. It isn't intended to beat up individuals.
No idea about the under-18 thing though. Where did that come from?
A couple of friends and I got hit by this exploit a couple of weeks ago. How did I find out? The sendmail failed to send out my passwd file (it kills off bind:) )! Imagine my surprise when it popped up in my inbox four days later. A quick halt, trot down to the shop to get RH7, burn a CD with all the updates, and re-install later, I'm up and back.
I think my machine was being used in a DOS attack at the time too, since it was sending out a lot of traffic. Maybe it was being used to target the other systems on the subnet. Who knows. I didn't do much in the way of investigation. I was remote at the time, and I pay for traffic.:)
Now, I've got a slightly more secure firewall, and I've turned off a bunch of services, and I've got tripwire running. I'm still seeing TCP connections on port 53 though, so lots of people are getting infected/running the exploit.
How can VMware be too expensive? US$79 for a version that will do what you want on linux. If a company can't afford to buy you that, how can it afford to pay your salary?
I mean, dreamweaver 4 costs US$299 per license, coldfusion US$495. If the company is complaining about another $80, when you are definitely billed out at more than that per _hour_, something is messed up.
The last time I read about this technology, they were talking about using it for cell-phone displays. The only problem was that the dye that displays blue wore out very quickly, on the order of 2-3 years. However, this is fine for a cell phone because the average life span of a phone is 1-2 years. For a TV/laptop though?
Why not just get a cheap prepaid cellular phone? You don't _have_ to give anyone your number.
It's like buying a PC. You may only want to play quake, but are you really going to turn down the free copy of Unreal Tournament that came with the network card?:)
The blackberry may be cool... For North America. Everywhere else in the world, you are already able to send and receive short messages (as well as email) on your cellular phone. It is also easy to write emails on them, especially with dictionary based word-prediction, like some of the Nokia's.
I don't understand why everyone thinks RIM is the best thing in the world. As soon as 3G or even GSM is taken up in North America, they are out of the market...
I don't think that you really get it. It _still_ has a _single_ point of failure, and it's a pretty large point, the whole box. Urbanjunkie is asking for a solution that will allow service to continue even if a whole box (any box) is destroyed.
Think earthquake. Then design a network layout and set of systems that would withstand the total destruction of any system in the network. That's what urbanjunkie is asking for.
Database replication can be nasty business. It all comes down to the failure modes. Things like distributed transactions can have some pretty nasty ones. No, I don't have a solution.:) If your period between updates is high, you can try having one authoritative system. You perform all your updates to it, and then periodically shut it down and copy the files to the remote servers.
I think you're right... But doesn't that just
continue the joke?
www.dictionary.com
affect1 (-fkt)
v. tr. affected, affecting, affects.
1.To have an influence on or effect a change in: Inflation affects the buying power of the dollar.
2.To act on the emotions of; touch or move.
3.To attack or infect, as a disease: Rheumatic fever can affect the heart.
O.k. putting all of the intelligence into the
endpoint is a _bad_ thing. By doing that, in order to offer a service, you have to require that they use a _specific_ platform, and a specific release. All of a sudden, the idea of interoperability goes out the window. We will be back to the situation where the only phones you will be able to use are ones approved by the phone company, complete with them being wired into the wall.:)
I think there is some confusion about the intent of the protocols discussed. SIP/H.323 are signalling protocols. They leave it up to the endpoint to make decisions. MGCP is a _control_ protocol (says so in the name!:) ). MGCP is great at controlling gateways. It is made to do things like control VoIP gateways, allowing thirdparty ISUP systems to manage the circuits and perform the H.323/SIP for them.
Moving intelligence away from the endpoint is _good_ when you are trying to get large port densities. Currently, port densities for VoIP gateways tend to max out at about ~1400 ports (eg Cisco 5800). 1400 ports only gets you four calls/second if they stay on the line for the average 5 minutes. Four calls is nothing, piddly, miniscule. Now imagine having to manage several hundred gateways, each one with it's own configuration just to get the call rate up to an acceptable level. Now you see the reason MGCP exists. Gateways are supposed to be stupid. They route speech. They shouldn't have to think about it, it's hard enough on the DSPs to just convert it to packets.:)
If you pull the intelligence required for basic services away from the endpoint (perhaps through proxies) you can then put the intelligence in a central location, allowing a lower cost solution (since the endpoints can be _very_ stupid), as well as allowing people to purchase more expensive endpoints for additional features. The basic features are still provided by the central server. Features such as call waiting, call forward on busy (voicemail), billing, number portability, voice prompts, announcements, etc. would all be implemented in a server/proxy. If you wanted fancier routing capabilities, you could buy something that gave you additional control, but you aren't required to buy it to get basic services.
Of course going in the other direction is also bad. You want the endpoint to be allowed to make decisions about what is going to happen. So, for communication with the final destination, MGCP is probably a poor choice, and that's why SIP/H.323 exist (my preference is H.323...SIP sucks.:P ).
Saying E2E is good and saying that protocols that exclude E2E (such as MGCP) are bad prevents a whole range of technological solutions from being used. A control protocol is just that, a control protocol. If you don't need (or want) intelligence at an endpoint, why require it?
Long Distance. But there is a difference. The upstream providers will want to make money on the call. People don't make money on software unless they have monopoly control on the market. You do make money by owning the infrastructure. The same as the gold rush. The miners didn't make the money, the guy selling the shovels did.
IM interop. Only because they don't have to message someone on a competing system. As soon as that happens, then they will start to complain.:) Already, I have people trying to convince me to get an AIM account to talk to them.
Dir Svcs. But someone needs to get paid to run the server. They aren't going to do it for free, and you can't sell ads on it. So, you charge for the service.
DNS. True. You do still need to be able to accept calls from the regular phone system, so you will still need a regular phone number.
But you do need 24x7, because it is usually during a power outage that people need to call emergency services.:)
I can see a place for a low-cost long distance charge avoidance. In fact, these already exist (net-2-phone?). Of course, that is already regulated (being an inter-LATA carrier), and billed.:) But in order to be a real replacement, it will need to match up the other guarantees. The power outage is a major reason why there aren't fiber-based phones out there. As soon as the power goes out, so does your phone. With regular phones, that doesn't happen.
But you are still free to implement your own codec for encryption purposes. You would need a real fast system (or large one-time-pad), but it is entirely doable. The provider just needs to provide the functionality or "The Man" will be on his ass faster than you can say "Show me the subpoena!".:) It's just that people tend to get protective of kids and harrassing calls. Other services like that are the whole "*69" thing.
I don't like having to pre-register my keys, but I do believe that wiretapping and call tracing is usefull.
Wonder how long an article goes before it gets to be summary only?:)
VoIP is different because of the regulation needed to keep monopolies from forming. If a call goes long-distance, it needs to be handled by another carrier. This prevents AT&T from owning everything.
This isn't interconnect at the IP level, this is interconnect at the "Where the hell is Joe" level. You still can't call the guy. Think ICQ, AIM, Yahoo, MSN IM.
Even with email addresses, you still need to find someone. Unless each system has it's own static IP and static DNS?
You don't understand how much of a barrier to entry having to change phone numbers is. Think business. They have spent $500k on marketing materials, and they would have to reprint it all in order to switch phone companies. Will they? Nope. Who cares about Joe at home? The phone companies don't. They don't make enough money from them.
Regulations do guarantee QOS. Since there are so few carriers (even in the US, there are really only 3 long distance carriers, the rest resell the first ones), that QOS will suffer. Or do you really think that your cable modem is up 24x7? I know mine isn't. It goes down whenever there's a power outage. Phones don't do that.
Not good enough. You have a phone, you have to be able to call 911... That they would agree to it means nothing if someones shooting people and someone picks up the office phone and can't dial 911. Watch the lawsuits fly.
This is _VERY_ important. You may not see it that way, but then how about tracing obscene/harrasing calls? Wire taps are a very important part of the investigative process. They have just the same level of proof to go through as if they wanted to search your house (higher actually). Any amount of you thinking that they shouldn't be allowed to do that won't make it go away. Wishing it were true doesn't count.
Jason Pollock
Regulations to prevent predatory pricing. This will stop companies from forcing others out of the market strictly because they have more cash in the bank. Do you think that the local upstart ISP-turned-CLEC could stand up to say AOL-Time Warner in a pricing battle?
Regulations to force interconnection. How would you like it if you were on AOL, but you couldn't call anyone on MSN? How about if the new competition wasn't able to acquire numbers? Don't laugh, that's a big problem here in New Zealand. The incumbent was refusing to give the competition 800 numbers...
Regulations to allow people to switch carriers easily. So, in order to change carriers, I need to give up my phone number? No way. (Number portability has solved this in most local areas).
Regulations to prevent prices from rising uncontrollably. If there is only one carrier (because you failed in 1,2 and 3, the resulting carrier will raise prices.
Regulations to guarantee QOS.
Regulations to guarantee emegency services. Do you really care that you can't dial 911?
Regulations to allow wire-tapping, obscene call tracing, etc. You need all of these before you can offer local services.
...
As you can see, there are many regulations out there that people _need_. Some are for safety, some are to level the playing field. But, all are needed.
In order for VoIP to be useful, it needs to interconnect with both the existing phone network, and be able to find other people. To do this, you need gatekeepers. It is reasonable to expect ISPs to put gatekeepers into the network, and then require all H.323/SIP traffic to traverse them. I know that my ISP already does this for HTTP/FTP/etc. H.323 and SIP both talk on standard ports, and need to perform codec negotiation, also on standard ports.
You need to talk to someone who isn't logged in? To do that you need to go through a gateway, which again can be logged, tracked and billed.
As for performance... The company I'm working for has a VoIP "switch" that is able to track and bill (in real-time, not post-processing) 500+ calls/second/system. For post-processing, you can crank that up to 1500 calls/second/system.
H.323 is not slow if you do it properly. Open H.323, while nicely architected and OO, is incredibly slow. The current limitation on H.323? The fact that each call needs 2-4 control sockets, and 2 UDP sockets.
As a comparison, Mobistar, a cell-phone operator in Belgium, generates <45 calls/second (regular phone traffic).
So, it is both possible, and probable to bill for VoIP traffic. In fact, I hope to get rich doing it.:)
Unregulated territory is not always the best. Most countries regulate Telcos to _prevent_ them from lowering prices. That way an incumbent is not able to sell services at a loss in order to kill off a competitor. It may result in higher prices in the short term as the competitor is established, but once they are established your cool. For an example of what happens in a (relatively) unregulated market, look at what Air Canada did to Canadian. After Canadian was dead, prices instantly went up by a considerable amount.
Interconnection fees are paid between carriers based on the number of minutes one person terminates a call for the other. In other words, if Telco A terminates(connects) a call for Telco B, B pays A x/minute.
Everyone figured out that hey, ISPs generate a lot of incoming traffic! That means that having lots of ISP ports generates a large amount of interconnect money.
The incumbent Telco's realized that they were forking over large amounts of cash to CLECs who were basically just clearing houses for ISPs (who were getting cheap modems because the cost was being offset by the interconnect fees). So, they went to the FCC and tried to get the ISPs relegated to the same class as long-distance. That would have effectively reversed the fee flow back towards the originator (since Long Distance companies pay to receive calls from the local loop).
The FCC refused.
VoIP carriers in the US are not exempt from tariffs, and they will charge that back. Note that this won't tariff all VoIP calls, but then is VoIP usefull if you can't call for a pizza or the ambulance with it? In order for it to be usefull, it has to interconnect with the regular old PSTN, so it can be billed appropriately at that point.
Yes I do blame AMD for it not working with RH6.2 (it isn't a bug, since it isn't Linux's fault). They are supposed to produce a drop in clone. It isn't. They could have easily avoided the problem by providing hardware to the OS developers, but that obviously hasn't happened. Therefore, yes it is their fault. In this case, it has cost them a customer.
It is perfectly valid for Linux to panic on a SIGILL. You can't blame the OS developers if your "clone" equipment doesn't work, it's the clone's fault. Do you blame Creative Labs if an SB clone doesn't work with the standard driver? Nope. You don't blame Linux either.
I bought an AMD Athlon 700. Then I tried to install RH 6.2 on it. Imagine my surprise when the kernal paniced when it tried to boot. Seems that RH 6.2 thought I had a PIII and was attempting to disable the CPUID.
The "solution" I was given? Build a new kernel (sorta difficult if it won't boot in the first place). So, I sent the Athlon back, and bought a PIII 800, which ended up being _cheaper_ than the Athlon (for _exactly_ the same system except MB and CPU) anyways.
AMD has neat CPUs, but the execution leaves much to be desired.
Yes, I do blame AMD for RH not working. At the very least, they should provide RH with example systems to use for testing. Software providers are frequently provided with free hardware for compatibility testing.
Slashdot Mirror
I believe one of the reasons for the drop in sales is that they are leaving the country.
There isn't a decent online retailer in Canada. Chapters? Took them 4 weeks to ship me a DVD (to ship it, it still hasn't arrived). This is after cancelling the order 3 times arbitrarily, and asking for a photocopy of both my card and statement. No, I didn't provide it. Sam The Record Man? Closed down.
So, I buy stuff from Amazon. They've never flubbed an order, or failed to ship. What does that mean? Yep, my sales don't show up on the counts for Canada. It doesn't mean that I'm not buying, I'm just not buying Canadian.
Jason PollockHere's what I'm looking at:
Add a small hard disk, a read-only linux image, and you are good to go. The hardest part will probably be getting the video out to work under linux. I haven't been able to find any references online. :)
They even make a small case and power supply, here, but the case is a little ugly. The power supply doesn't appear to have a fan though.
I'm using an earlier version of the board as the basis for my car-based mp3 system. Works really well. Can't say anything about the PSU, it became available after I started with my system.
Jason PollockIf ht://Dig does most of what you want, why not make it understand file:// URL's if it doesn't now? I wouldn't expect it to be _that_ difficult an addition, especially if you are concerned about the availability of source code.
Otherwise, why not just pay up the US$2k and buy a Glimpse site license? US$2k is probably worth about _two_ days of your time to the company (depending on the size of the company). That would cover everything as long as you kept it to one site.
Glimpse also appears to be available with source too, check out the bottom of the download page.
If the tool is to be used to create a commercial product that you are going to sell, then you won't be covered by that. Personally, if I was going to be using a GPL'ed (or other open source) tool as the core of a product I was going to sell, I would feel ethically bound to pay some money for it, regardless of it's availability for "free".
Jason PollockI realise that, but there isn't anything keeping them from saying that if you adhere to the standard, use this implementation, then you will interoperate. Other languages are able to work this way. Java does, C forces everyone to because they have to link in system libraries.
Different revisions of the standard could still change the name mangling to force link failure, but if you are both on the same revision, I would expect it to work.
If the system libraries were implemented in C++, don't you think we would have an implementation standard? Yep, me too. :)
Can you tell that this is currently a major beef with me? :) I'm having to write two hunks of C code, passing void *'s around because I can't link two libraries together, and I don't have the source code for the Forte one.
Jason PollockWhat I would like to see is an implementation standard for C++.
At least some way that I could link C++ code generated by GCC with others, such as Sun Forte. We use GCC for platform independence, but we are continually having to write stupid little C function wrappers around classes to get around all of the linking issues.
Then I might be able to do some code re-use at more than just the source level.
Jason PollockI used to believe this too. Of course, like many things, the truth is a little more complicated. When I tried this line on a physicist friend of mine (who used to work for the AECL), she sat me down and smacked me around. :)
While the heavy water moderator is required to initiate the reaction (the rods are initially less than critical mass), once started, the rods end up enriching themselves, and may be able to sustain a reaction on their own.
Also, even if the reaction stops, the rods may generate enough heat to melt anyways. This is because a good fraction of the heat generated in the plant is by the decay of fission products. You can't stop the decay no matter what you do.
Of course, this is probably incredibly unlikely. But that just means you have to roll the dice more often. :)
Cool eh?
Jason PollockThis was done to prevent US carriers from simply putting POPs in Canada and routing traffic through the US, undercutting Canadian suppliers. This regulation is primarily for long distance (voice) carriers. It does appear to have achieved what it set out to do, which was force new entrants to the networking market to build their own networks, rather than route through their networks in the US. And yes, this rule _is_ enforced when it contravenes the _intent_ of the regulation. It isn't intended to beat up individuals.
No idea about the under-18 thing though. Where did that come from?
Jason PollockA couple of friends and I got hit by this exploit a couple of weeks ago. How did I find out? The sendmail failed to send out my passwd file (it kills off bind :) )! Imagine my surprise when it popped up in my inbox four days later. A quick halt, trot down to the shop to get RH7, burn a CD with all the updates, and re-install later, I'm up and back.
I think my machine was being used in a DOS attack at the time too, since it was sending out a lot of traffic. Maybe it was being used to target the other systems on the subnet. Who knows. I didn't do much in the way of investigation. I was remote at the time, and I pay for traffic. :)
Now, I've got a slightly more secure firewall, and I've turned off a bunch of services, and I've got tripwire running. I'm still seeing TCP connections on port 53 though, so lots of people are getting infected/running the exploit.
Jason PollockVMWare express, US$79.00. You can only run Win9x, and it can't be from an existing partition.
As for the Windows license, he already has one, since he is already dual-booting to use the windows tools he needs to.
You only need the full workstation product if you need one of the additional features like WinNT/2k support, or booting from raw disks.
How can VMware be too expensive? US$79 for a version that will do what you want on linux. If a company can't afford to buy you that, how can it afford to pay your salary?
I mean, dreamweaver 4 costs US$299 per license, coldfusion US$495. If the company is complaining about another $80, when you are definitely billed out at more than that per _hour_, something is messed up.
Jason PollockI just found one of Jim Gray's replication papers, he's now at Microsoft. Is this the one you were talking about?
The Dangers of Replication and a Solution (1996)The last time I read about this technology, they were talking about using it for cell-phone displays. The only problem was that the dye that displays blue wore out very quickly, on the order of 2-3 years. However, this is fine for a cell phone because the average life span of a phone is 1-2 years. For a TV/laptop though?
So, anyone know if they've sorted out the blue?
Jason PollockWhy not just get a cheap prepaid cellular phone? You don't _have_ to give anyone your number.
It's like buying a PC. You may only want to play quake, but are you really going to turn down the free copy of Unreal Tournament that came with the network card? :)
Jason PollockThe blackberry may be cool... For North America. Everywhere else in the world, you are already able to send and receive short messages (as well as email) on your cellular phone. It is also easy to write emails on them, especially with dictionary based word-prediction, like some of the Nokia's.
I don't understand why everyone thinks RIM is the best thing in the world. As soon as 3G or even GSM is taken up in North America, they are out of the market...
Jason PollockI don't think that you really get it. It _still_ has a _single_ point of failure, and it's a pretty large point, the whole box. Urbanjunkie is asking for a solution that will allow service to continue even if a whole box (any box) is destroyed.
Think earthquake. Then design a network layout and set of systems that would withstand the total destruction of any system in the network. That's what urbanjunkie is asking for.
Database replication can be nasty business. It all comes down to the failure modes. Things like distributed transactions can have some pretty nasty ones. No, I don't have a solution. :) If your period between updates is high, you can try having one authoritative system. You perform all your updates to it, and then periodically shut it down and copy the files to the remote servers.
Jason PollockI think you're right... But doesn't that just
continue the joke?
www.dictionary.com
affect1 (-fkt)
v. tr. affected, affecting, affects.
1.To have an influence on or effect a change in: Inflation affects the buying power of the dollar.
2.To act on the emotions of; touch or move.
3.To attack or infect, as a disease: Rheumatic fever can affect the heart.
O.k. putting all of the intelligence into the endpoint is a _bad_ thing. By doing that, in order to offer a service, you have to require that they use a _specific_ platform, and a specific release. All of a sudden, the idea of interoperability goes out the window. We will be back to the situation where the only phones you will be able to use are ones approved by the phone company, complete with them being wired into the wall. :)
I think there is some confusion about the intent of the protocols discussed. SIP/H.323 are signalling protocols. They leave it up to the endpoint to make decisions. MGCP is a _control_ protocol (says so in the name! :) ). MGCP is great at controlling gateways. It is made to do things like control VoIP gateways, allowing thirdparty ISUP systems to manage the circuits and perform the H.323/SIP for them.
Moving intelligence away from the endpoint is _good_ when you are trying to get large port densities. Currently, port densities for VoIP gateways tend to max out at about ~1400 ports (eg Cisco 5800). 1400 ports only gets you four calls/second if they stay on the line for the average 5 minutes. Four calls is nothing, piddly, miniscule. Now imagine having to manage several hundred gateways, each one with it's own configuration just to get the call rate up to an acceptable level. Now you see the reason MGCP exists. Gateways are supposed to be stupid. They route speech. They shouldn't have to think about it, it's hard enough on the DSPs to just convert it to packets. :)
If you pull the intelligence required for basic services away from the endpoint (perhaps through proxies) you can then put the intelligence in a central location, allowing a lower cost solution (since the endpoints can be _very_ stupid), as well as allowing people to purchase more expensive endpoints for additional features. The basic features are still provided by the central server. Features such as call waiting, call forward on busy (voicemail), billing, number portability, voice prompts, announcements, etc. would all be implemented in a server/proxy. If you wanted fancier routing capabilities, you could buy something that gave you additional control, but you aren't required to buy it to get basic services.
Of course going in the other direction is also bad. You want the endpoint to be allowed to make decisions about what is going to happen. So, for communication with the final destination, MGCP is probably a poor choice, and that's why SIP/H.323 exist (my preference is H.323...SIP sucks. :P ).
Saying E2E is good and saying that protocols that exclude E2E (such as MGCP) are bad prevents a whole range of technological solutions from being used. A control protocol is just that, a control protocol. If you don't need (or want) intelligence at an endpoint, why require it?
Jason PollockAs you can see, there are many regulations out there that people _need_. Some are for safety, some are to level the playing field. But, all are needed.
In order for VoIP to be useful, it needs to interconnect with both the existing phone network, and be able to find other people. To do this, you need gatekeepers. It is reasonable to expect ISPs to put gatekeepers into the network, and then require all H.323/SIP traffic to traverse them. I know that my ISP already does this for HTTP/FTP/etc. H.323 and SIP both talk on standard ports, and need to perform codec negotiation, also on standard ports.
You need to talk to someone who isn't logged in? To do that you need to go through a gateway, which again can be logged, tracked and billed.
As for performance... The company I'm working for has a VoIP "switch" that is able to track and bill (in real-time, not post-processing) 500+ calls/second/system. For post-processing, you can crank that up to 1500 calls/second/system. H.323 is not slow if you do it properly. Open H.323, while nicely architected and OO, is incredibly slow. The current limitation on H.323? The fact that each call needs 2-4 control sockets, and 2 UDP sockets.
As a comparison, Mobistar, a cell-phone operator in Belgium, generates <45 calls/second (regular phone traffic).
So, it is both possible, and probable to bill for VoIP traffic. In fact, I hope to get rich doing it. :)
Jason PollockUnregulated territory is not always the best. Most countries regulate Telcos to _prevent_ them from lowering prices. That way an incumbent is not able to sell services at a loss in order to kill off a competitor. It may result in higher prices in the short term as the competitor is established, but once they are established your cool. For an example of what happens in a (relatively) unregulated market, look at what Air Canada did to Canadian. After Canadian was dead, prices instantly went up by a considerable amount.
Jason PollockVoIP carriers in the US are not exempt from tariffs, and they will charge that back. Note that this won't tariff all VoIP calls, but then is VoIP usefull if you can't call for a pizza or the ambulance with it? In order for it to be usefull, it has to interconnect with the regular old PSTN, so it can be billed appropriately at that point.
Jason PollockYes I do blame AMD for it not working with RH6.2 (it isn't a bug, since it isn't Linux's fault). They are supposed to produce a drop in clone. It isn't. They could have easily avoided the problem by providing hardware to the OS developers, but that obviously hasn't happened. Therefore, yes it is their fault. In this case, it has cost them a customer.
It is perfectly valid for Linux to panic on a SIGILL. You can't blame the OS developers if your "clone" equipment doesn't work, it's the clone's fault. Do you blame Creative Labs if an SB clone doesn't work with the standard driver? Nope. You don't blame Linux either.
I bought an AMD Athlon 700. Then I tried to install RH 6.2 on it. Imagine my surprise when the kernal paniced when it tried to boot. Seems that RH 6.2 thought I had a PIII and was attempting to disable the CPUID.
The "solution" I was given? Build a new kernel (sorta difficult if it won't boot in the first place). So, I sent the Athlon back, and bought a PIII 800, which ended up being _cheaper_ than the Athlon (for _exactly_ the same system except MB and CPU) anyways.
AMD has neat CPUs, but the execution leaves much to be desired.
Yes, I do blame AMD for RH not working. At the very least, they should provide RH with example systems to use for testing. Software providers are frequently provided with free hardware for compatibility testing.
Jason Pollock