It feels like the blink of an eye since I kicked off the Open Source Telecom Software Project Survey in May this year. I shared all the results immediately with everyone who contributed to the survey. This weblog presents the results with some analysis for everyone to see, as promised in my presentations at TADSummit Americas and TADSummit EMEA.
My request to everyone who reviews the material is:
- If I’ve missed any relevant projects please let me know;
- Please ask questions / issues arising from this work; and
- Please suggest topics / questions we should include in a 2020 survey, thank you.
Open source projects are a critical part of the web’s success. As programmable telecoms becomes democratized, as explained in the Astricon Keynote, open source projects are of increasing importance to the telecoms / communications industries. Telcos have tended to shy away from open source telecom software, rather adopting retro-closed source solutions (e.g. IMS and RCS). But even there, open source has made its way in creating greater innovation and lower pricing, thanks to companies like ng-voice.
Most CPaaS / UCaaS / CCaaS providers use open source telecom software projects listed in this survey. Many of the tech-savvy providers have either forked a project, or are confident of being able to fork with no risk. A few of the bigger ones, like Twilio began on open source with Asterisk and FreeSWITCH, but have since built out their closed source platform. This is not a winners takes all business, it’s a dynamic market, meeting a plethora of evolving needs. Where exceptionally smart technologist continue to create and share software that makes the world better.
Given the above importance I wanted to provide to a broader audience an understanding of Open Source Telecom Software Project landscape.
The projects we’re evaluating include:
Telecom App Servers
- Restcomm / Mobicents
- SIPfoundry / sipxcom
Management Interfaces / Solution Wraps / Monitoring and Capture
- Elastix (for Asterisk)
- FreePBX (for Asterisk)
- FusionPBX (for FreeSwitch)
- Sipcapture – Homer
- Sippy Softswitch
- VICIdial (for Asterisk)
- Wazo (forked from XiVo, for Asterisk)
- YETI Switch
- sipsak / SIPp
Other (because there is some overlap / I like them / think the projects a relevant)
I missed Drachtio (Dave Horton) from the list this year. I’ve talked with Dave and I’m impressed with what he’s creating and his commitment to keeping the project non-commercial. His community is small, but this is one to watch.
One’s I’m not considering include: AskoziaPBX, Murmur, Mysipswitch, sipXecs IP PBX, Opencloud Rhino, Metaswitch Clearwater, GNU Gatekeeper, Yate, OpenTok.
I can not review all the insights generated in this summary weblog without making it extremely long and very boring. So just a few highlights are shown below. If you want to know more as you read through the slides, please just contact me, thanks.
For configuration management / scripting, the answer is Ansible. Should we ask addition questions on usage and scripts, should each project share popular scripts to aid adoption and setup?
Native Cloud (AWS, Google, etc.) is not a popular choice
- Is this because Telecom people are suspicious of the cloud?
- Is it because we like to control everything from the metal up?
- Are we always looking to save a dime? RTC is CPU intensive, and price is an issue.
- Or is it that the complexity that AWS, Google, etc, imply are not worth our time? RTC is latency sensitive so it that an issue?
- At TADSummit EMEA this was a discussion topic as serverless is becoming more popular. The FinTech companies attending were strong proponents of serverless. Referring to managing your own environment as “yak-shaving”.
- Should the survey delve more deeply into this topic in 2020?
Is FreeSWITCH losing gas? Or is this a temporary blip given the Signalwire announcement? We’ll be able to monitor the situation with this survey.
Kamailio/OpenSIPS seem to be ridding the xCaaS storm well. Telecoms is still hacking low level stuff and will probably continue to be for some time. Price, security and on-premises requirements make up a very big market that Twilio, RingCentral, Taslkdesk can’t disrupt yet..
Matrix has a lot of promise and support, but we’re still waiting for that break-out moment. They had another great customer win, which we covered in CXTech Week 52. Personally, I think it will be in another 3-5 years when we discover Matrix is simply everywhere for secure, decentralized, (and federated where appropriate) communications.
Geography. Some in Europe were surprised at the size of the Asterisk responses, it would be interesting to map how the projects are adopted around the world.
Please let me know what you think of the survey and where we should focus in 2020, thank you for your attention.
And for those not aware of all the projects here’s a quick summary of what they are: