First, there is no such thing as "the Asterisk Network". Asterisk is a device.
This means, service providers are not in direct competition with Asterisk as you suggest. To the contrary. Due to its multi-protocol support, Asterisk is probably the most service agnostic device there is in this space. Service providers can easily gain by embracing Asterisk, at least those providers who support open standards.
In a world with closed services, it is very difficult for service providers to win business from customers of their competitors or to win lost customers back, simply because customers have to switch equipment to change providers.
In a world with service agnostic devices like Asterisk, it is very easy for service providers to get a chance to win the business of other providers' customers and if only as a secondary service. Not only does Asterisk make it easy to change providers, but it can handle multiple concurrent providers. As a result, a provider can more easily win back a lost customer's business. In fact, a provider may not lose the customer's business in the first place. Properly configured, Asterisk will use another provider's service at times when a given provider has problems to complete calls. This means, the user will not feel the outage and is less likely to change providers as long as the mix of providers used delivers an overall positive experience.
With a single closed service using proprietary standards or locked devices, once the provider loses a customer as a result of quality problems, may they be temporary or not, that customer is usually gone forever. Quite often such a customer may become a foe advocating strongly against the provider they have left behind.
As far as measures to make competing services perform worse are concerned, again, because of its flexibility and multi-protocol, multi-codec support, Asterisk is the device that's probably better suited to cope than most others.
This can be as simple as changing port numbers, which locked devices and service will not allow you to do. It can be more sophisticated such as tuning the QoS parameters, also something which many devices won't allow you to do. Or it can be as neat as using Asterisk's secret weapon: IAX.
Not only is IAX a NAT friendly protocol, but it also is an extremely efficient and robust protocol. Quite a few people on the Asterisk mailing list have reported experiences using Asterisk in extremely bad environments where IAX did cope astonishingly well while other protocols went from severely degraded service to no service at all.
So, no matter how you look at it, you will find that Asterisk has a pretty good value proposition. There is a reason why it has become so popular.
... and that is ILBC, which you can have with any open source telephony solution, too.
as codecs go, there is no such thing as a good codec per se, it depends on the circumstances. Some codecs are good in some circumstances, some other codecs are good in some other circumstances. That's why most telephony solutions negotiate codecs depending on the quality of the connection when they connect with each other.
Skype is inferior in this respect because it can't adjust to a more suitable codec in the event that ILBC isn't a good match for the connection.
And since Skype is closed, there is nothing you can do about it. You just have to put up with the limitations imposed upon you by Skype.
Well, someone seems to think this is funny, I don't. Whoever believes this is a joke, please check out the site below...
http://tuxmobil.org/mobilix_asterix.html
It has details on the case I described with several links to other sources and it also lists other cases where this French publishing company has sued users of far fetched similar names as the characters in their comic books.
Those guys don't shy away from any frivolous lawsuit they can bring and they have the dollars to follow it through.
It would be very sad if Asterisk did show up on their radar screen.
Judging by the amount of comments this article has received so far, it would seem that there was enough interest to justify it showing up even if it wasn't all that "new" as you say.
Also, most of the publicity Skype is getting is more or less repetition, not always, but many times it is. So, shouldn't we be happy that an open source telephony & VOIP project like Asterisk gets it's share of free publicity every now and then?
depending on what stacks we are talking about, your kernel may end up containing 99.99% telco stuff and 0.01% Linux. Of course, I am deliberately exaggerating but the point is that you probably don't want to turn Linux kernel development into a telco R&D lab.
It may seem a little paranoid, but I really believe we should take care not to use the incorrect name "Asterix" when referring to Asterisk.
Evidently, the publishing company of the popular French comic books with the famous character Asterix are extremely litigious. They have sued a German Linux company over the use of the name Mobilix for a Linux distro that was aimed specifically at mobile computing, hence Mobilix.
They don't have any character by the name Mobilix in their comic books, but they claimed that Mobilix was passing off anyway just because it was similar in appearance as all their characters have a name ending in "ix". Worse still, they won the lawsuit and those poor Linux developers had to hand over their domain. AFAIK, the case is now awaiting appeal at the German high court, so not all is lost, but it's pretty scary nevertheless.
Now, if there is enough noise on the net where people use "Asterix" instead of Asterisk, those comic book people may get ideas and sue Digium. They may claim that there is proof of confusion with their Asterix trademark by merely pointing out how many times somebody used "Asterix" instead of Asterisk. And they may just find a judge crazy enough to go along with that.
So, I say, let's be careful, let those sleeping dogs lie.
So you actually meant to say you had a bad experience with Broadvoice full stop. At least I can't see anything in your post that would even hint at something going wrong with Asterisk and the technology behind it.
To the contrary. With Asterisk you can connect with any of a few hundred service providers with next to zero effort. Thus, you can change providers in a heart beat or even use multiple providers concurrently and have Asterisk automatically fail over to another provider when your call doesn't go through. How do you think your local telco achieves quality of service? They do the same thing: using redundant routes.
As far as charging disputes go, Asterisk generates what telephone folks call CDRs, call detail records, one for every single call. This can help you to backup your claim that you have been incorrectly charged.
Yes, Asterisk is an excellent tool for what you want to do.
It has quite a few neat tricks for telemarketer avoidance. First, there is a thing called Zapateller, which if enabled, sends a so called SIT sequence on the line when a call comes in and telemarketer's equipment hangs up on that.
Those telemarketers who don't have equipment that hangs up on the SIT sequence, eg. telemarketers located overseas, can usually be blocked just as easily by sending all calls without caller IDs or with unknown caller IDs to a voice menu that asks the caller to press a touch tone key. This is because telemarketers use so called predicitve diallers, systems that dial and only connect the call to their staff when there is somebody human on the other end of the line. If a predictive dialler hits your voice menu, it will just hang up and call some other number.
Likewise you can do all kinds of smart things with calls from callers you do know. For example, some you may want to forward to your mobile, to some others you may want to announce an alternative number to call, yet others you may want to forward to someone else or to voicemail. Asterisk can send voicemail to you by email as an attachement and it can send you an SMS to your mobile phone with the number of the caller and the time of the call.
You say you aren't keen on using the VOIP features, but VOIP isn't only about making long distance calls over the net. It is also about extending the reach of your home phone line. For example, you may be out of the house but as long as you have internet access, you could still be picking up your phone at home when a call comes in. Or you could make a call using your home phone while you are some place else.
It's pretty addictive. Once you've started using something like Asterisk, you keep using it in more and more interesting and innovative ways.
With the help of Cygwin, a Unix compatibility API layer for Windows, you can now run Asterisk on Windows. Keep in mind though that this has only been a very recent development, so it is probably less mature than Asterisk on other platforms such as Linux, BSD, MacOS X and Solaris.
There is also OpenPBX from Voicetronix, a Perl based PBX which runs on Windows, but it doesn't have the IP capabilities Asterisk has.
The sharing of landlines between Asterisk users is already happening on a community basis. It's an initiative by Jeff Pulver's Free World Dialup, aka FWD, a free VOIP network, and it's called FWDout.
And, yes, most of the open protocols Asterisk supports can be used for both audio and video, namely SIP, IAX and H.323.
"What I would like to know is how I may build a free (as in speech) Skype-like network with my friends, using Asterisk (or something else)?"
That's rather easy to do with Asterisk.
The first thing to do is - surprise - to set up an Asterisk server. Next, you configure a user account for yourself and one for each of your friends.
Then you tell your friends your server's address or DNS name, username and password and ask them to download a software phone that supports any of the open standards Asterisk supports, eg. SIP, IAX and H.323 to name the most important ones.
For Windows, your preferred choice would probably be the Firefly softphone, which supports both SIP and IAX, another one is called X-lite which is SIP only. For Linux there are quite a few open source softphones supporting various protocols, SJphone, Kphone, GnomeMeeting and more. For MacOS X there is X-Lite and the cross platform iaxComm (Win/Lin/Mac). All those are free.
Then all that remains to do is to tell those softphones how to find your Asterisk server and what their username and password is. In some cases a little fine tuning may be needed. For example, if someone is behind NAT, you may have to work around NAT traversal problems.
The easiest way to avoid NAT problems is to use the IAX protocol and a softphone that support IAX, eg. Firefly or iaxComm. IAX doesn't have NAT issues, so no work arounds are needed.
Note, that Asterisk supports multiple protocols concurrently. So, some of your friends might come in using SIP while others use IAX and yet others use H.323. The overhead for Asterisk to translate between protocols is negligible.
Everybody can now call everybody else by their username, which could be a nickname or an internal phone number. If a user isn't logged in, calls will go to voicemail. You can also set up chat rooms for multi-party voice conferencing.
In addition, you can set up so called SIP URIs, which is akin to an email address. In fact, your email address may well be identical to your SIP URI. Using that SIP URI, anybody with a SIP device can now call anyone on your DIY VOIP network, if you want to allow that.
Your friends can also register their ordinary phone numbers with a directory service like E164.org and if somebody with an appropriately configured IP-PBX calls that number, the call would not pass over the PSTN but over the internet via your Asterisk server to the owner of that number.
All this is not very difficult to do and you don't need a very powerful box either. So, all I can say is: Go for it!
Asterisk also runs on *BSD, MacOS X and Solaris. With the help of Cygwin it even runs on Windows now.
In fact, talking about an easy to set up home PBX, you might actually find MacOS X to be far more likely to suit your needs.
There is an Asterisk installer for the Mac, so you don't have to built it yourself and there are GUI based setup wizards, or assistants as they're called in the Mac world, which allow non-geeks without tech skills to set up a basic home PBX in just a few minutes.
A driver for using the Mac's built-in modem as a voice port to connect to a POTS line is on its way.
But even if you don't have a Mac nor want to buy one, I assume that similar tools will eventually show up for Windows now that Asterisk runs under Cygwin.
Asterisk on Linux will probably remain a "mostly for geeks" affair. Then again, there are some promising efforts under way to package Asterisk and Linux in a "works out of the box" fashion, for example Asterisk@Home.
Anyway, you shouldn't compare Asterisk with Skype because Asterisk is a _server_ application that can be linked to just about _any_ service and Skype is a _client_ application that is _locked_ to one single service.
The beauty of Zaptel is that it is a philosophy and a convention more than anything else.
Consequently, there is no such thing as "the zaptel people" or "zaptel management".
Anybody who follows the convention to write a driver automatically becomes part of what you call "the zaptel people". Anybody who participates in maintaining any zaptel code automatically becomes part of what you call "zaptel management".
In this regard, Zaptel is about as anarchist as it gets. It started live as a free BSD driver released under BSD license by Jim Dixon for a T1 card he designed and also released "open source".
Later on, Mark Spencer of Digium ported Jim's zaptel driver to Linux. He also wrote and released more drivers following the same convention to support other hardware. Although Digium deserve a lion share of the credit to make Zaptel successful, they didn't invent it, they don't own it and they don't necessarily control it.
More about Zaptel is at http://www.zapatatelephony.org
With all due respect, your advice seems rather inappropriate for the kind of drivers we are talking about here. This is not about storage controllers or video cards.
If every vendor of telephony gear was to insert their APIs and drivers into the Linux kernel, you will have to rename it to something like "kernel mode telephony library with some minor operating system features attached".
I certainly know who Gerry works for. Yet, considering the wording he chose, "a division, specializing in telecommunications equipment, of a very large hardware manufacturer", I didn't feel it was appropriate for me to spell it out.
But indeed, I agree with you that there is an incentive for Intel to release open source dialogic drivers for Asterisk if only to make sure potential customers don't have to go through Digium if they want to use Intel/Dialogic hardware.
But there is also the factor of increasing competition. Only a year ago, PRI telephony hardware for Asterisk was available only from Digium and Intel/Dialogic. Digium was the default and Intel/Dialogic was the fallback you would choose if you couldn't use Digium. For example, Digium is very weak on international support, they have type approval for their gear only in a very few places while Intel/Dialogic have local presence and type approval just about anywhere.
However, over the last year things have been changing dramatically. Today, pretty much any telephony interface vendor is aware of and interested in Asterisk. Voicetronix have entered the PRI market, Acculab, Eicon, Sangoma also support Asterisk now, others are in the process of development or at least evaluation, eg. Brooktrout. More vendors are likely to follow.
Unlike Digium, those vendors have presence in and type approval for most international markets. This means Intel/Dialogic are no longer the default fallback when Digium doesn't fit. It is therefore in Intel's interest to make it as easy as possible for integrators to be able to use Intel/Dialogic gear. Open source drivers can help to achieve this.
Intel have been very active in the GNU Bayonne community, but they have left Asterisk support entirely to Digium. Of course this has a lot to do with Bayonne's initial focus on IVR applications which is the traditional Dialogic domain. However, one might take the view that Intel have bet on the wrong horse. It certainly wouldn't be to their disadvantage to engage directly in the Asterisk community and release open source drivers. Zaptel compatibility would be the icing on the sugar.
You may want to check out the open Zaptel interface driver suite. [Google for Zapata Telephony]
It was originally developed by Jim Dixon for his Tormenta T1 card (open source GPLed hardware BTW) but has since been used with open source telephony projects such as Asterisk.
Asterisk is an interesting example to study in respect of open versus closed telephony drivers.
Some vendors have closed driver support for Asterisk, eg. Intel/Dialogic which means their drivers can only be sold through a non-GPL Asterisk License. This however means that they rely on sales through Digium, who hold rights in Asterisk. The irony is that Digium are also a telephony interface card vendor and thus a competitor.
Voicetronix have open source driver support for Asterisk through their own GPLed drivers. Yet, the action on open source drivers is with Zaptel and so Voicetronix have to do the work on their open source drivers all by themselves and their drivers lack features that Zaptel drivers have.
Sangoma Technologies support Zaptel in addition to their own drivers. To Asterisk and other telephony packages using Zaptel, a Sangoma device is just another Zaptel device. A significant benefit for end users, open source projects and the vendor.
Zaptel is now supported on Linux (x86 and PPC) and BSD. In addition, work is under way for Zaptel on Solaris and MacOS X.
Hydrogen powered engines or fuel cells exhaust water in form of steam. And if you release hydrogen into the atmosphere, it will react with the oxygen in the air and burn to water, again in form of steam.
Water steam in the air tends to eventually come down as rain. No greehhouse gas. That's the beauty of hydrogen.
A problem arises only when the hydrogen is produced through burning fossil fuels. However, it can be produced cleanly, eg. solar powered hydrogen plants.
But even if the hydrogen is produced by burning fossil fuels, you still get a benefit because power plants are typically more efficient than otto engines and also because the emissions are kept out of the cities.
All you have to do is let your telephone portal know all the phone numbers of people you know.
Then, you let your phone portal pick up the line and play a greeting to anybody who is not identified as somebody you know. The majority of telemarketer's systems will hang up on that already. However, after the greeting you let your portal ask for an action to be taken by the caller "press 1". Even the most persistent telemarketer systems will hang up on that.
The reason is that the folks who work for telemarketers don't usually call you themselves. The dialling is automated and calls are screened by automated systems. Only if a real human picks up will the call be connected through to the call center.
No, reason to get paranoid. Blocking telemarketers is far easier than blocking spam.
Forget all you know about SPAM blocking because blocking telemarketers is totally different and by no means anywhere near as difficult as blocking spammers.
If you do already have a problem experiencing too many calls from telemarketers, you should consider using Asterisk as your telephone portal. Asterisk can block those telemarketers efficiently for you no matter if they come in over POTS (plain old telephone system) or VOIP.
The simplest anti-telemarketer tool in Asterisk's arsenal is called Zapateller, a feature that if enabled will play a so called SIT sequence to callers which forces telemarketers' predictive diallers to hang up instantly. You won't even know somebody was calling.
Beyond Zapateller there are various ways to make sure you only get legitimate calls. An easy way to get telemarketers to hang up is to let Asterisk's autoattendant pick up the call and play a welcome message. Most telemarketers will hang up on that recording.
This is because telemarketer call centers are volume businesses where time is money. Every second their marketeers are not talking to potential customers are a waste of their time and money. Consequently they have invested heavily in equipment that makes sure that they only get connected when the system has determined that there is a real human on the line. Such systems are called predictive diallers.
A predictive dialler will pick numbers from a database to call and screen those calls when the line is being picked up. The marketeers will only be connected if the predictive dialler has determined that there is a human at the other end of the line. These systems are quite sophisticated.
This can be very easily used against them. In most cases it takes as little as playing a welcome message first. If the predicitive dialler is programmed to listen and wait for some time to see if the call gets put through to a human, then it takes as little as asking for some interaction, like "press 1 if you are not a telemarketer". Predictive diallers will simply hang up on this.
Asterisk has scripting abilities for building such voice menus which make it very easy to set up an autoattendant that will make telemarketers give up on trying to reach you. Further, you can teach Asterisk to put through calls from people you know immediatly and let them bypass the autoattendant. To do that, you teach Asterisk the phone numbers and names of people you know, and then ask those people to make sure they call you with their caller ID.
Really, I don't think that SPIT is going to be a problem similar to that of SPAM. Filtering out spam is fairly difficult. Blocking SPIT is rather straight forward. No reason to get paranoid IMO.
Actually, Vicimarketing have a predicitive dialler module for Asterisk. If call centres are asking for predictive diallers for Asterisk then it is only a matter of time that they discover IAX.
So, sooner or later you'll get telemarketing calls over IAX, rest assured of that.
Of course, if you are using Asterisk as your phone portal, then it is fairly easy to filter them out. Asterisk has a whole arsenal of tools to block telemarketers very efficiently.
that's fine with me. As I said, you had already disqualified yourself by your inappropriate choice of language and my response was meant not for you but for other readers.
Those Sipuras are -shall we say- well featured. They have a lot of tricks up their sleve that are not immediately apparent. One of those is the ability to define your own dial plan.
This means you can teach it to dial out via SIP network A when dialing a certain prefix, eg. *9 and dial out via another SIP network B when dialling some other prefix, eg. *8.
You can use this to your advantage such that you can participate both in the Free World Dialup network as well as use some commercial SIP provider for PSTN gateway service all from a single device.
Slashdot Mirror
the point is that the choice for the telcos is this:
either they embrace the VOIP revolution and the resulting empowering of customers even though this may hurt a little here and there,
or they can fight it and in the process of doing so get wiped out and replaced by something like Skype.
First, there is no such thing as "the Asterisk Network". Asterisk is a device.
This means, service providers are not in direct competition with Asterisk as you suggest. To the contrary. Due to its multi-protocol support, Asterisk is probably the most service agnostic device there is in this space. Service providers can easily gain by embracing Asterisk, at least those providers who support open standards.
In a world with closed services, it is very difficult for service providers to win business from customers of their competitors or to win lost customers back, simply because customers have to switch equipment to change providers.
In a world with service agnostic devices like Asterisk, it is very easy for service providers to get a chance to win the business of other providers' customers and if only as a secondary service. Not only does Asterisk make it easy to change providers, but it can handle multiple concurrent providers. As a result, a provider can more easily win back a lost customer's business. In fact, a provider may not lose the customer's business in the first place. Properly configured, Asterisk will use another provider's service at times when a given provider has problems to complete calls. This means, the user will not feel the outage and is less likely to change providers as long as the mix of providers used delivers an overall positive experience.
With a single closed service using proprietary standards or locked devices, once the provider loses a customer as a result of quality problems, may they be temporary or not, that customer is usually gone forever. Quite often such a customer may become a foe advocating strongly against the provider they have left behind.
As far as measures to make competing services perform worse are concerned, again, because of its flexibility and multi-protocol, multi-codec support, Asterisk is the device that's probably better suited to cope than most others.
This can be as simple as changing port numbers, which locked devices and service will not allow you to do. It can be more sophisticated such as tuning the QoS parameters, also something which many devices won't allow you to do. Or it can be as neat as using Asterisk's secret weapon: IAX.
Not only is IAX a NAT friendly protocol, but it also is an extremely efficient and robust protocol. Quite a few people on the Asterisk mailing list have reported experiences using Asterisk in extremely bad environments where IAX did cope astonishingly well while other protocols went from severely degraded service to no service at all.
So, no matter how you look at it, you will find that Asterisk has a pretty good value proposition. There is a reason why it has become so popular.
... and that is ILBC, which you can have with any open source telephony solution, too.
as codecs go, there is no such thing as a good codec per se, it depends on the circumstances. Some codecs are good in some circumstances, some other codecs are good in some other circumstances. That's why most telephony solutions negotiate codecs depending on the quality of the connection when they connect with each other.
Skype is inferior in this respect because it can't adjust to a more suitable codec in the event that ILBC isn't a good match for the connection.
And since Skype is closed, there is nothing you can do about it. You just have to put up with the limitations imposed upon you by Skype.
You are confusing Voicepulse (residential) with Voicepulse Connect.
The former has monthly plans, the latter is specifically for people/companies who run their own Asterisk servers and it does not have monthly plans.
Well, someone seems to think this is funny, I don't. Whoever believes this is a joke, please check out the site below ...
http://tuxmobil.org/mobilix_asterix.html
It has details on the case I described with several links to other sources and it also lists other cases where this French publishing company has sued users of far fetched similar names as the characters in their comic books.
Those guys don't shy away from any frivolous lawsuit they can bring and they have the dollars to follow it through.
It would be very sad if Asterisk did show up on their radar screen.
Judging by the amount of comments this article has received so far, it would seem that there was enough interest to justify it showing up even if it wasn't all that "new" as you say.
;)
Also, most of the publicity Skype is getting is more or less repetition, not always, but many times it is. So, shouldn't we be happy that an open source telephony & VOIP project like Asterisk gets it's share of free publicity every now and then?
Remember the saying "Any news is good news"
depending on what stacks we are talking about, your kernel may end up containing 99.99% telco stuff and 0.01% Linux. Of course, I am deliberately exaggerating but the point is that you probably don't want to turn Linux kernel development into a telco R&D lab.
It may seem a little paranoid, but I really believe we should take care not to use the incorrect name "Asterix" when referring to Asterisk.
Evidently, the publishing company of the popular French comic books with the famous character Asterix are extremely litigious. They have sued a German Linux company over the use of the name Mobilix for a Linux distro that was aimed specifically at mobile computing, hence Mobilix.
They don't have any character by the name Mobilix in their comic books, but they claimed that Mobilix was passing off anyway just because it was similar in appearance as all their characters have a name ending in "ix". Worse still, they won the lawsuit and those poor Linux developers had to hand over their domain. AFAIK, the case is now awaiting appeal at the German high court, so not all is lost, but it's pretty scary nevertheless.
Now, if there is enough noise on the net where people use "Asterix" instead of Asterisk, those comic book people may get ideas and sue Digium. They may claim that there is proof of confusion with their Asterix trademark by merely pointing out how many times somebody used "Asterix" instead of Asterisk. And they may just find a judge crazy enough to go along with that.
So, I say, let's be careful, let those sleeping dogs lie.
Did you read the question?
The parent had asked if and how he could build his own skype like service with a tool like Asterisk.
I could name at least half a dozen skype like services that one can just join, but that wasn't asked for.
So you actually meant to say you had a bad experience with Broadvoice full stop. At least I can't see anything in your post that would even hint at something going wrong with Asterisk and the technology behind it.
To the contrary. With Asterisk you can connect with any of a few hundred service providers with next to zero effort. Thus, you can change providers in a heart beat or even use multiple providers concurrently and have Asterisk automatically fail over to another provider when your call doesn't go through. How do you think your local telco achieves quality of service? They do the same thing: using redundant routes.
As far as charging disputes go, Asterisk generates what telephone folks call CDRs, call detail records, one for every single call. This can help you to backup your claim that you have been incorrectly charged.
So don't blame it on the tool.
Yes, Asterisk is an excellent tool for what you want to do.
It has quite a few neat tricks for telemarketer avoidance. First, there is a thing called Zapateller, which if enabled, sends a so called SIT sequence on the line when a call comes in and telemarketer's equipment hangs up on that.
Those telemarketers who don't have equipment that hangs up on the SIT sequence, eg. telemarketers located overseas, can usually be blocked just as easily by sending all calls without caller IDs or with unknown caller IDs to a voice menu that asks the caller to press a touch tone key. This is because telemarketers use so called predicitve diallers, systems that dial and only connect the call to their staff when there is somebody human on the other end of the line. If a predictive dialler hits your voice menu, it will just hang up and call some other number.
Likewise you can do all kinds of smart things with calls from callers you do know. For example, some you may want to forward to your mobile, to some others you may want to announce an alternative number to call, yet others you may want to forward to someone else or to voicemail. Asterisk can send voicemail to you by email as an attachement and it can send you an SMS to your mobile phone with the number of the caller and the time of the call.
You say you aren't keen on using the VOIP features, but VOIP isn't only about making long distance calls over the net. It is also about extending the reach of your home phone line. For example, you may be out of the house but as long as you have internet access, you could still be picking up your phone at home when a call comes in. Or you could make a call using your home phone while you are some place else.
It's pretty addictive. Once you've started using something like Asterisk, you keep using it in more and more interesting and innovative ways.
let's just say that any publicity for Asterisk is good news, belated or not ;)
With the help of Cygwin, a Unix compatibility API layer for Windows, you can now run Asterisk on Windows. Keep in mind though that this has only been a very recent development, so it is probably less mature than Asterisk on other platforms such as Linux, BSD, MacOS X and Solaris.
There is also OpenPBX from Voicetronix, a Perl based PBX which runs on Windows, but it doesn't have the IP capabilities Asterisk has.
The sharing of landlines between Asterisk users is already happening on a community basis. It's an initiative by Jeff Pulver's Free World Dialup, aka FWD, a free VOIP network, and it's called FWDout.
And, yes, most of the open protocols Asterisk supports can be used for both audio and video, namely SIP, IAX and H.323.
"What I would like to know is how I may build a free (as in speech) Skype-like network with my friends, using Asterisk (or something else)?"
That's rather easy to do with Asterisk.
The first thing to do is - surprise - to set up an Asterisk server. Next, you configure a user account for yourself and one for each of your friends.
Then you tell your friends your server's address or DNS name, username and password and ask them to download a software phone that supports any of the open standards Asterisk supports, eg. SIP, IAX and H.323 to name the most important ones.
For Windows, your preferred choice would probably be the Firefly softphone, which supports both SIP and IAX, another one is called X-lite which is SIP only. For Linux there are quite a few open source softphones supporting various protocols, SJphone, Kphone, GnomeMeeting and more. For MacOS X there is X-Lite and the cross platform iaxComm (Win/Lin/Mac). All those are free.
Then all that remains to do is to tell those softphones how to find your Asterisk server and what their username and password is. In some cases a little fine tuning may be needed. For example, if someone is behind NAT, you may have to work around NAT traversal problems.
The easiest way to avoid NAT problems is to use the IAX protocol and a softphone that support IAX, eg. Firefly or iaxComm. IAX doesn't have NAT issues, so no work arounds are needed.
Note, that Asterisk supports multiple protocols concurrently. So, some of your friends might come in using SIP while others use IAX and yet others use H.323. The overhead for Asterisk to translate between protocols is negligible.
Everybody can now call everybody else by their username, which could be a nickname or an internal phone number. If a user isn't logged in, calls will go to voicemail. You can also set up chat rooms for multi-party voice conferencing.
In addition, you can set up so called SIP URIs, which is akin to an email address. In fact, your email address may well be identical to your SIP URI. Using that SIP URI, anybody with a SIP device can now call anyone on your DIY VOIP network, if you want to allow that.
Your friends can also register their ordinary phone numbers with a directory service like E164.org and if somebody with an appropriately configured IP-PBX calls that number, the call would not pass over the PSTN but over the internet via your Asterisk server to the owner of that number.
All this is not very difficult to do and you don't need a very powerful box either. So, all I can say is: Go for it!
"I'm the only one in the family running Linux ..."
Asterisk also runs on *BSD, MacOS X and Solaris. With the help of Cygwin it even runs on Windows now.
In fact, talking about an easy to set up home PBX, you might actually find MacOS X to be far more likely to suit your needs.
There is an Asterisk installer for the Mac, so you don't have to built it yourself and there are GUI based setup wizards, or assistants as they're called in the Mac world, which allow non-geeks without tech skills to set up a basic home PBX in just a few minutes.
A driver for using the Mac's built-in modem as a voice port to connect to a POTS line is on its way.
But even if you don't have a Mac nor want to buy one, I assume that similar tools will eventually show up for Windows now that Asterisk runs under Cygwin.
Asterisk on Linux will probably remain a "mostly for geeks" affair. Then again, there are some promising efforts under way to package Asterisk and Linux in a "works out of the box" fashion, for example Asterisk@Home.
Anyway, you shouldn't compare Asterisk with Skype because Asterisk is a _server_ application that can be linked to just about _any_ service and Skype is a _client_ application that is _locked_ to one single service.
The beauty of Zaptel is that it is a philosophy and a convention more than anything else.
Consequently, there is no such thing as "the zaptel people" or "zaptel management".
Anybody who follows the convention to write a driver automatically becomes part of what you call "the zaptel people". Anybody who participates in maintaining any zaptel code automatically becomes part of what you call "zaptel management".
In this regard, Zaptel is about as anarchist as it gets. It started live as a free BSD driver released under BSD license by Jim Dixon for a T1 card he designed and also released "open source".
Later on, Mark Spencer of Digium ported Jim's zaptel driver to Linux. He also wrote and released more drivers following the same convention to support other hardware. Although Digium deserve a lion share of the credit to make Zaptel successful, they didn't invent it, they don't own it and they don't necessarily control it.
More about Zaptel is at http://www.zapatatelephony.org
With all due respect, your advice seems rather inappropriate for the kind of drivers we are talking about here. This is not about storage controllers or video cards.
If every vendor of telephony gear was to insert their APIs and drivers into the Linux kernel, you will have to rename it to something like "kernel mode telephony library with some minor operating system features attached".
I certainly know who Gerry works for. Yet, considering the wording he chose, "a division, specializing in telecommunications equipment, of a very large hardware manufacturer", I didn't feel it was appropriate for me to spell it out.
But indeed, I agree with you that there is an incentive for Intel to release open source dialogic drivers for Asterisk if only to make sure potential customers don't have to go through Digium if they want to use Intel/Dialogic hardware.
But there is also the factor of increasing competition. Only a year ago, PRI telephony hardware for Asterisk was available only from Digium and Intel/Dialogic. Digium was the default and Intel/Dialogic was the fallback you would choose if you couldn't use Digium. For example, Digium is very weak on international support, they have type approval for their gear only in a very few places while Intel/Dialogic have local presence and type approval just about anywhere.
However, over the last year things have been changing dramatically. Today, pretty much any telephony interface vendor is aware of and interested in Asterisk. Voicetronix have entered the PRI market, Acculab, Eicon, Sangoma also support Asterisk now, others are in the process of development or at least evaluation, eg. Brooktrout. More vendors are likely to follow.
Unlike Digium, those vendors have presence in and type approval for most international markets. This means Intel/Dialogic are no longer the default fallback when Digium doesn't fit. It is therefore in Intel's interest to make it as easy as possible for integrators to be able to use Intel/Dialogic gear. Open source drivers can help to achieve this.
Intel have been very active in the GNU Bayonne community, but they have left Asterisk support entirely to Digium. Of course this has a lot to do with Bayonne's initial focus on IVR applications which is the traditional Dialogic domain. However, one might take the view that Intel have bet on the wrong horse. It certainly wouldn't be to their disadvantage to engage directly in the Asterisk community and release open source drivers. Zaptel compatibility would be the icing on the sugar.
You may want to check out the open Zaptel interface driver suite. [Google for Zapata Telephony]
It was originally developed by Jim Dixon for his Tormenta T1 card (open source GPLed hardware BTW) but has since been used with open source telephony projects such as Asterisk.
Asterisk is an interesting example to study in respect of open versus closed telephony drivers.
Some vendors have closed driver support for Asterisk, eg. Intel/Dialogic which means their drivers can only be sold through a non-GPL Asterisk License. This however means that they rely on sales through Digium, who hold rights in Asterisk. The irony is that Digium are also a telephony interface card vendor and thus a competitor.
Voicetronix have open source driver support for Asterisk through their own GPLed drivers. Yet, the action on open source drivers is with Zaptel and so Voicetronix have to do the work on their open source drivers all by themselves and their drivers lack features that Zaptel drivers have.
Sangoma Technologies support Zaptel in addition to their own drivers. To Asterisk and other telephony packages using Zaptel, a Sangoma device is just another Zaptel device. A significant benefit for end users, open source projects and the vendor.
Zaptel is now supported on Linux (x86 and PPC) and BSD. In addition, work is under way for Zaptel on Solaris and MacOS X.
Are you trying to be funny or are you clueless?
Hydrogen powered engines or fuel cells exhaust water in form of steam. And if you release hydrogen into the atmosphere, it will react with the oxygen in the air and burn to water, again in form of steam.
Water steam in the air tends to eventually come down as rain. No greehhouse gas. That's the beauty of hydrogen.
A problem arises only when the hydrogen is produced through burning fossil fuels. However, it can be produced cleanly, eg. solar powered hydrogen plants.
But even if the hydrogen is produced by burning fossil fuels, you still get a benefit because power plants are typically more efficient than otto engines and also because the emissions are kept out of the cities.
All you have to do is let your telephone portal know all the phone numbers of people you know.
Then, you let your phone portal pick up the line and play a greeting to anybody who is not identified as somebody you know. The majority of telemarketer's systems will hang up on that already. However, after the greeting you let your portal ask for an action to be taken by the caller "press 1". Even the most persistent telemarketer systems will hang up on that.
The reason is that the folks who work for telemarketers don't usually call you themselves. The dialling is automated and calls are screened by automated systems. Only if a real human picks up will the call be connected through to the call center.
No, reason to get paranoid. Blocking telemarketers is far easier than blocking spam.
Forget all you know about SPAM blocking because blocking telemarketers is totally different and by no means anywhere near as difficult as blocking spammers.
If you do already have a problem experiencing too many calls from telemarketers, you should consider using Asterisk as your telephone portal. Asterisk can block those telemarketers efficiently for you no matter if they come in over POTS (plain old telephone system) or VOIP.
The simplest anti-telemarketer tool in Asterisk's arsenal is called Zapateller, a feature that if enabled will play a so called SIT sequence to callers which forces telemarketers' predictive diallers to hang up instantly. You won't even know somebody was calling.
Beyond Zapateller there are various ways to make sure you only get legitimate calls. An easy way to get telemarketers to hang up is to let Asterisk's autoattendant pick up the call and play a welcome message. Most telemarketers will hang up on that recording.
This is because telemarketer call centers are volume businesses where time is money. Every second their marketeers are not talking to potential customers are a waste of their time and money. Consequently they have invested heavily in equipment that makes sure that they only get connected when the system has determined that there is a real human on the line. Such systems are called predictive diallers.
A predictive dialler will pick numbers from a database to call and screen those calls when the line is being picked up. The marketeers will only be connected if the predictive dialler has determined that there is a human at the other end of the line. These systems are quite sophisticated.
This can be very easily used against them. In most cases it takes as little as playing a welcome message first. If the predicitive dialler is programmed to listen and wait for some time to see if the call gets put through to a human, then it takes as little as asking for some interaction, like "press 1 if you are not a telemarketer". Predictive diallers will simply hang up on this.
Asterisk has scripting abilities for building such voice menus which make it very easy to set up an autoattendant that will make telemarketers give up on trying to reach you. Further, you can teach Asterisk to put through calls from people you know immediatly and let them bypass the autoattendant. To do that, you teach Asterisk the phone numbers and names of people you know, and then ask those people to make sure they call you with their caller ID.
Really, I don't think that SPIT is going to be a problem similar to that of SPAM. Filtering out spam is fairly difficult. Blocking SPIT is rather straight forward. No reason to get paranoid IMO.
Actually, Vicimarketing have a predicitive dialler module for Asterisk. If call centres are asking for predictive diallers for Asterisk then it is only a matter of time that they discover IAX.
So, sooner or later you'll get telemarketing calls over IAX, rest assured of that.
Of course, if you are using Asterisk as your phone portal, then it is fairly easy to filter them out. Asterisk has a whole arsenal of tools to block telemarketers very efficiently.
that's fine with me. As I said, you had already disqualified yourself by your inappropriate choice of language and my response was meant not for you but for other readers.
Those Sipuras are -shall we say- well featured. They have a lot of tricks up their sleve that are not immediately apparent. One of those is the ability to define your own dial plan.
This means you can teach it to dial out via SIP network A when dialing a certain prefix, eg. *9 and dial out via another SIP network B when dialling some other prefix, eg. *8.
You can use this to your advantage such that you can participate both in the Free World Dialup network as well as use some commercial SIP provider for PSTN gateway service all from a single device.