CTserver is now actually ahead of bayonne in support for the OpenSwitch cards because it's just better tested. I imagine that bayonne will catch up soon, as we'll release a lot more code and docs in the next few days.
All these projects are functional and each have their niche, but asterisk is obviously way more mature as a pbx than either bayonne or CT server.
Telecom application developers want nothing more than good docs. I've been through enough pain with bayonne to feel any other way. If those aren't there, fuck the support revenue. Seriously. If the topic or the app is complex, that's one thing. I'll support that as necessary. But operating with incorrect/missing docs is no business model at all. It's pretty obvious that asterisk wants to improve their docs as badly as the bayonne project does; I bet they feel the same way.
The bayonne docs are improving, BTW. Check out David's latest scripting guide. It's actually complete. When I've verified with my own eyes that it's a little better tested, I might even consider taking on another bayonne project.
I have far, far more trouble with TDM service providers of my existing customers misconfiguring or screwing up service or lying to my face than I do with bayonne. I thought CTI hardware and software was out of control, until I tried dealing with MCI.
The point is, Voicetronix is a kick ass company that offers good support and put has put a lot of effort into open source projects over the years, of which this is just one. In addition to projects they themselves have initiated, they have contributed to speex, bayonne, and openh323. The feature set will, inevitably, expand.
Bayonne is by far the most mature telephony server out there, having been in existence as ACS long before even asterisk was a twinkle in kram's eye. Best of all, it's hardware independent, unlike either asterisk or openpbx. It supports just about every CTI card line that supports linux. They just added both PBX support and H.323 support. SIP is not far off, and if you want to be able to script complex applications in no time flat, optionally dropping out to perl, python, or php for logic, Bayonne is unparalleled. Add to that direct mysql, postgres, support, text to speech support, internatilalization, and so on and so on...
That this thing uses VNC is speculation. As far as anyone here knows, this phone may be ENTIRELY proprietary, and coming from AT&T, it probably is. 70's video phone updated to use SIP. Even if you do get a tiny little VNC viewer running on your desk phone, it's still sitting next to your spacious monitor and input devices. Why bother? B. F. D.
The value of the software in question is largely in it's reference designs for a variety of protocols heretofore without open source stacks. In telecom, a large amount of equipment is shipped with a proprietary protocol of some kind, so for those who have such equipment (most businesses), a reverse engineered stack is a must if you are to gainfully interface to it, (i.e. with a linux-based voicemail system). The second area where the code is valuable is in ten yers of hardcore experience with messaging systems. The data structures that are outlined by the code are not obvious and did not exist prior to the code release in question, basically saving open-source programmers ten years of grief.
Your hopes have been realized. See the Bayonne project for a next-gen open source IVR system running on Linux or FreeBSD and OST's website for commercial support.
Listen, a full implementation of an H.323 stack is a softswitch. For an open one, see openh323.org. Every cisco switch shipped within the last year is a softswitch, it's in IOS. A softswitch is a call manager the responds via a standard protocol interface and carries out switching logic. THE GOAL IS TO CREATE OPEN SOFTSWITCHES. Your PBX doesn't use SS7, but when you are getting 32 channels of voice and 1M of data from Sprint next year over an MMMD wireless connection, you will be happy to find that your linux box can split this out to USB phones on your LAN. Why? Because some people are out there working their asses off to create softswitches that are open-source. That's why.
No one cares how it works? Where do you think hackers came from? Or UNIX? The phone system is interesting as all hell and getting better all the time. Why do you think Cisco includes VOIP software in every copy of IOS that ships? Because it's not just the pphone companies and the hackers that are interested, it's the new generation of communications companies as well.
You obviously have never looked at a phone system. Software does not = hardware. Telephony systems are being developed out of commodity PC hardware these days, just like everything else. Application cards that handle analog lines and voip functions are becoming just as much a commodity.
The article was about breaking through the walls of one of the last and most proprietary software industries, and the one with the most to gain by this happening. The history at the start just highlighted the difficulty in changing legacy thinking. In this industry, it's earth shattering.
Well, it isn't perfect, but it is evolving. Calls over the internet are growing by every measure month over month, including percentage of total voice traffic. Speaking from experience, our Cisco 2600 with the latest IOS load and a 4-port fxo plugin makes very good sounding gateway calls from europe over the internet to the local 650 area code daily. However, speakfreely with GSM compression over the same distance sounds pretty bad, but is useable. GSM compression is actually pretty good, and it's unencumbered. I feel it's a matter of time before we have an open-source app that approaches the quality of our hardware solution, but not reliability. At that point, 50% of voice traffic over the internet is probably a feasible projection.
Our company (www.ostel.com) supports open source telecommunications solutions. We have commercially installed and supported speakfreely, with and without hardware compression cards from Quicknet. We have also done full telecommunications infrastructures using entirely open source software (ACS, Hylafax, Speakfreely, custom code -> implementing messaging, conferencing, and e-commerce apps), and developed embedded telecommunications servers for a number of clients. We currently have 10 active development projects. It can certainly be done. We would be happy to talk to anyone who is interested in working on such projects. Check out www.voxilla.org for some interesting project listings.
Linus used to say that databases would never have a crowd of open source developers because they are too boring, but he had to take it back. Check out the GNU enterprise project on www.gnu.org. These guys are making headway into ERP already, and would definitely be interested in the code. Hackers don't just hack out of interest, they understand the economics of being an expert in a highly valued application field.
All of them involve some coding, as this is a fairly new thing to do with a Linux box.
If you want only one port, you can grab a quicknet linejack and use their API. The card will run a couple hundred bucks. You could put more than one card in a box as well.
If you want a large number of ports, and money is no object, you can grab any of the cards from Audiocodes or NMS and work with them. They both have linux support, although NMS is the only open source driver you'll find.
If you are looking for more of a challenge, you could grab any number of products that weren't exactly made for the task but has some Linux support. Pika has supported linux for quite some time and their cards make a great linux based pbx (see ACS at voxilla.org). Dialogic is rumored to be releasing drivers for their line of cards. Also, Linmodems.org is gradually reverse engineering drivers for "win-modems" with the added advantage of no command set. There are already voip test apps available for these things, and they only cost 15 bucks. If you have and AD-TRAN frame relay switch, there is an open-source PBX available for that at linux-support.net (The ad-tran does voice over frame relay). And I suppose you could control some of the commercial voip boxes with just about any PC.
They can't easily keep the good reporting out, but they can sure as hell put biased reporting in. Encouraging a novice to set up an NT and a Linux box for a security test is a no-brainer for MS. Out of the box, linux is pretty darn far from secure. It is our responsibility in such tests to crack both boxes, and secure the linux box.
I don't think so. Sure, MS will try to control and obfuscate it, but so can you. If free software implementing this protocol makes useful improvements (as opposed to "ehance-embrace-destroy" strategic "improvements" ) then people will start using them and the protocol will change. Didn't someone reverse engineer the AOL protocol? That must have been a pain in the ass. I don't have a problem with protocols developed by commercial companies. It makes sense that open-source communications clients should be able to auto-detect several protocols, as well as implementing their own protocol. We should get together, check out the spec, compare it to the AOL stuff, and come up with a really open non-commercial instant messaging protocol. Then submit it to the IETF, the only remotely trustable standards body.
Nothing against Craig, but David Sugar has also been contributing to Linux for some time now. He has written BBS's, Getty's, protocol stacks for switch communications, and his latest project, ACS, is a very promising communications server. Check out his stuff at http://www.tycho.com
Pika, Aculab, and NMS all have cards that support switching...but they are also all in early beta as far as driver development goes. The best you can do about switching at the moment is to lobby pika and aculab to open source their drivers or work with nms on porting their open source driver to one of their cards that does support switching.
now, now. there are drivers for the pika line of telephony cards (http://www.pika.ca) that support fax (without artifacts like modems) voice, switching, tode detection, caller id, etc. You can find a Linux based phone system that can be used with these cards on http://www.tycho.com. It's called ACS.
You can also use the AT+V command set with any voice modem card. I have the chase PCI-RAS4 at home running under linux and it works quite well.
The reason I haven't had time to work on voxilla is that I have been working at VA Research fo the last 6 months and they keep people busy. I'll update the site tonight with a bunch of links to the developments that have been going in all over the place that almost no-one has asked about.
Slashdot Mirror
CTserver is now actually ahead of bayonne in support for the OpenSwitch cards because it's just better tested. I imagine that bayonne will catch up soon, as we'll release a lot more code and docs in the next few days.
All these projects are functional and each have their niche, but asterisk is obviously way more mature as a pbx than either bayonne or CT server.
Telecom application developers want nothing more than good docs. I've been through enough pain with bayonne to feel any other way. If those aren't there, fuck the support revenue. Seriously. If the topic or the app is complex, that's one thing. I'll support that as necessary. But operating with incorrect/missing docs is no business model at all. It's pretty obvious that asterisk wants to improve their docs as badly as the bayonne project does; I bet they feel the same way.
The bayonne docs are improving, BTW. Check out David's latest scripting guide. It's actually complete. When I've verified with my own eyes that it's a little better tested, I might even consider taking on another bayonne project.
I have far, far more trouble with TDM service providers of my existing customers misconfiguring or screwing up service or lying to my face than I do with bayonne. I thought CTI hardware and software was out of control, until I tried dealing with MCI.
The point is, Voicetronix is a kick ass company that offers good support and put has put a lot of effort into open source projects over the years, of which this is just one. In addition to projects they themselves have initiated, they have contributed to speex, bayonne, and openh323. The feature set will, inevitably, expand.
Bayonne is by far the most mature telephony server out there, having been in existence as ACS long before even asterisk was a twinkle in kram's eye. Best of all, it's hardware independent, unlike either asterisk or openpbx. It supports just about every CTI card line that supports linux. They just added both PBX support and H.323 support. SIP is not far off, and if you want to be able to script complex applications in no time flat, optionally dropping out to perl, python, or php for logic, Bayonne is unparalleled. Add to that direct mysql, postgres, support, text to speech support, internatilalization, and so on and so on...
That this thing uses VNC is speculation. As far as anyone here knows, this phone may be ENTIRELY proprietary, and coming from AT&T, it probably is. 70's video phone updated to use SIP. Even if you do get a tiny little VNC viewer running on your desk phone, it's still sitting next to your spacious monitor and input devices. Why bother? B. F. D.
Because sound cards suck.
ibm makes powerpc chip for linux appliances.
Why don't you look at running ACS on a quicknet phonejack?
The value of the software in question is largely in it's reference designs for a variety of protocols heretofore without open source stacks. In telecom, a large amount of equipment is shipped with a proprietary protocol of some kind, so for those who have such equipment (most businesses), a reverse engineered stack is a must if you are to gainfully interface to it, (i.e. with a linux-based voicemail system). The second area where the code is valuable is in ten yers of hardcore experience with messaging systems. The data structures that are outlined by the code are not obvious and did not exist prior to the code release in question, basically saving open-source programmers ten years of grief.
Your hopes have been realized. See the Bayonne project for a next-gen open source IVR system running on Linux or FreeBSD and OST's website for commercial support.
every linux company sells GPL'd software. SUN makes over half their consulting revenue off of sevices.
Listen, a full implementation of an H.323 stack is a softswitch. For an open one, see openh323.org. Every cisco switch shipped within the last year is a softswitch, it's in IOS. A softswitch is a call manager the responds via a standard protocol interface and carries out switching logic. THE GOAL IS TO CREATE OPEN SOFTSWITCHES. Your PBX doesn't use SS7, but when you are getting 32 channels of voice and 1M of data from Sprint next year over an MMMD wireless connection, you will be happy to find that your linux box can split this out to USB phones on your LAN. Why? Because some people are out there working their asses off to create softswitches that are open-source. That's why.
No one cares how it works? Where do you think hackers came from? Or UNIX? The phone system is interesting as all hell and getting better all the time. Why do you think Cisco includes VOIP software in every copy of IOS that ships? Because it's not just the pphone companies and the hackers that are interested, it's the new generation of communications companies as well.
You obviously have never looked at a phone system. Software does not = hardware. Telephony systems are being developed out of commodity PC hardware these days, just like everything else. Application cards that handle analog lines and voip functions are becoming just as much a commodity.
The article was about breaking through the walls of one of the last and most proprietary software industries, and the one with the most to gain by this happening. The history at the start just highlighted the difficulty in changing legacy thinking. In this industry, it's earth shattering.
Well, it isn't perfect, but it is evolving. Calls over the internet are growing by every measure month over month, including percentage of total voice traffic. Speaking from experience, our Cisco 2600 with the latest IOS load and a 4-port fxo plugin makes very good sounding gateway calls from europe over the internet to the local 650 area code daily. However, speakfreely with GSM compression over the same distance sounds pretty bad, but is useable. GSM compression is actually pretty good, and it's unencumbered. I feel it's a matter of time before we have an open-source app that approaches the quality of our hardware solution, but not reliability. At that point, 50% of voice traffic over the internet is probably a feasible projection.
Our company (www.ostel.com) supports open source telecommunications solutions. We have commercially installed and supported speakfreely, with and without hardware compression cards from Quicknet. We have also done full telecommunications infrastructures using entirely open source software (ACS, Hylafax, Speakfreely, custom code -> implementing messaging, conferencing, and e-commerce apps), and developed embedded telecommunications servers for a number of clients. We currently have 10 active development projects. It can certainly be done. We would be happy to talk to anyone who is interested in working on such projects. Check out www.voxilla.org for some interesting project listings.
Linus used to say that databases would never have a crowd of open source developers because they are too boring, but he had to take it back. Check out the GNU enterprise project on www.gnu.org. These guys are making headway into ERP already, and would definitely be interested in the code. Hackers don't just hack out of interest, they understand the economics of being an expert in a highly valued application field.
All of them involve some coding, as this is a fairly new thing to do with a Linux box.
If you want only one port, you can grab a quicknet linejack and use their API. The card will run a couple hundred bucks. You could put more than one card in a box as well.
If you want a large number of ports, and money is no object, you can grab any of the cards from Audiocodes or NMS and work with them. They both have linux support, although NMS is the only open source driver you'll find.
If you are looking for more of a challenge, you could grab any number of products that weren't exactly made for the task but has some Linux support. Pika has supported linux for quite some time and their cards make a great linux based pbx (see ACS at voxilla.org). Dialogic is rumored to be releasing drivers for their line of cards. Also, Linmodems.org is gradually reverse engineering drivers for "win-modems" with the added advantage of no command set. There are already voip test apps available for these things, and they only cost 15 bucks. If you have and AD-TRAN frame relay switch, there is an open-source PBX available for that at linux-support.net (The ad-tran does voice over frame relay). And I suppose you could control some of the commercial voip boxes with just about any PC.
Anything else you wanted to know?
They can't easily keep the good reporting out, but they can sure as hell put biased reporting in. Encouraging a novice to set up an NT and a Linux box for a security test is a no-brainer for MS. Out of the box, linux is pretty darn far from secure. It is our responsibility in such tests to crack both boxes, and secure the linux box.
I don't think so. Sure, MS will try to control and obfuscate it, but so can you. If free software implementing this protocol makes useful improvements (as opposed to "ehance-embrace-destroy" strategic "improvements" ) then people will start using them and the protocol will change. Didn't someone reverse engineer the AOL protocol? That must have been a pain in the ass. I don't have a problem with protocols developed by commercial companies. It makes sense that open-source communications clients should be able to auto-detect several protocols, as well as implementing their own protocol. We should get together, check out the spec, compare it to the AOL stuff, and come up with a really open non-commercial instant messaging protocol. Then submit it to the IETF, the only remotely trustable standards body.
Nothing against Craig, but David Sugar has also been contributing to Linux for some time now. He has written BBS's, Getty's, protocol stacks for switch communications, and his latest project, ACS, is a very promising communications server. Check out his stuff at http://www.tycho.com
Pika, Aculab, and NMS all have cards that support switching...but they are also all in early beta as far as driver development goes. The best you can do about switching at the moment is to lobby pika and aculab to open source their drivers or work with nms on porting their open source driver to one of their cards that does support switching.
now, now. there are drivers for the pika line of telephony cards (http://www.pika.ca) that support fax (without artifacts like modems) voice, switching, tode detection, caller id, etc. You can find a Linux based phone system that can be used with these cards on http://www.tycho.com. It's called ACS.
You can also use the AT+V command set with any voice modem card. I have the chase PCI-RAS4 at home running under linux and it works quite well.
The reason I haven't had time to work on voxilla is that I have been working at VA Research fo the last 6 months and they keep people busy. I'll update the site tonight with a bunch of links to the developments that have been going in all over the place that almost no-one has asked about.