Keynote Review TADSummit 2022

Running the event on a tight budget means I’m often dealing with logistics during the event. This year Youtube streaming in Portugal was problematic, the event location had 200 Mbit/s bidirectional, I think it was the local ISP messing with the stream. Anyway, these reviews are in part for me to give all the excellent content deserved attention and add my commentary on its importance.

With these links you can get to all the TADSummit content directly: Photos, Videos, Slides, Brief Agenda with videos and slides, Full Agenda with videos and slides.

And I can not say this enough, thank you to STROLIDBroadvoice / GoContactRadisysRingCentralStacuity, and AWA Network / Automat Berlin for sponsoring TADSummit 2022. And thank you to all the presenters and attendees in making the event excellent.

Welcome to vCon! The next leap forward in the programmable communications industry. 

Thomas Howe, CTO STROLID. Slides and Video. Simply, vCon should be part of the product roadmap and strategy for any business that supports customer or employee communications. It will rapidly become a standard from the IETF, its already released as an IETF Internet draft, covered in CXTech Week 44 2022.

We were lucky Thomas made the event as that evening he came down with an illness and spent the second day of TADSummit in bed at his Airbnb.

Earlier this year we wrote a whitepaper on vCon to introduce the concept and explain the value it presents to the industry. Thomas introduced the idea of vCon at TADSummit 2021. Simply, vCon is a PDF for conversations. Its something we should have created over a decade ago, but we didn’t, so let’s get it done ASAP.

Being an open standard is important for adoption and creating an ecosystem of tools, e.g. a redaction tool if third parties are involved in a conversation to meet regulatory requirements.

The four parts of a vCon are:

  • Dialogs – voice, video, text either packed in the vCon or linked.
  • Identities – people or entities (bots) involved in the conversation
  • Analysis – sentiment, translations, fraud scores, etc.
  • Attachments – presentations or URLs

Thomas then ran through some of the interesting use cases. He also raised an interesting point on the energy savings possible through vCon in the training of machine learning models. Simply vCon (robot food) enables the customer data used in training to be tracked.

During the audience discussion the point was raised on whether vCon would be required by regulations to enabled traceability and prove compliance. Thomas pointed out GDPR (General Data Protection Regulation) requires companies not to release data that could identity their customers’ sex, age or race – both audio and video enable that. As customer privacy and the use of customer data continues to grow in importance to businesses, vCon will play a critical role in managing that data.

Founding a Startup in Telecoms. The good, the bad and the ugly. 

João Camarate, CTO at Broadvoice & GoContact. Slides and Video. The insight for me was the viability of buying an existing business with customers, and using that as a stepping stone to grow a business faster.

Joao has recently sold GoContact to broadvoice, so shared his experiences. Most people who have built companies in telecoms say the same thing, don’t do it! Joao shares some of the reasons.. The key advice for him, after ‘don’t do it,’ is find a problem that is either yours or you have direct (daily) access to the customer and solve it. And then find more of those customers and build out the business. Incrementalism versus the grand plan. He have lots of other advice relevant to the current environment in programmable communications.

But back to the insight Joao shared on buying an existing business. He recommends the book “Buy Then Build” by Walter Deibel, and HBR series on “Buying a Small Business.”

80% of SMB are owned by people over 60 years old. Their children are not interested in running the business. I’ve seen it within my own family. Joao highlighted there are low interest rate loans for up to 90% of the purchase value for companies up to $4M in revenue. And 20/30 years ago lots of small businesses where created delivering local telecom solutions that still have recurring maintenance revenues.

I’ve not tried this, of the people I know who’ve done this, due diligence is critical, and do not ‘fall in love’ with the business before you buy, do the math and keep to your purchase ceiling. Believe in yourself not to screw the business up, it’s a going concern.

This type of presentation is unique to TADSummit. Frank, insightful advice from successful entrepreneurs in programamable telecoms. Every other event I know of is either more marketing focused, or focused on selling your business. Which is another problem, but nowhere near as difficult as building it. And Mark White shares his advice on selling a business at TADSummit as he’s moved over the sell side of private equity.

AWA – a Telco bootstrapping product development: Challenges with dynamic market consolidation – an 18 month road trip.

Marten Schoenherr, CEO Automat Berlin/AWA Network/Founding Partner at TheWorkinGroup. Slides and Video. This was great to see Marten proving the concept that could revolutionize SMS aggregation.

TADSummit is the only event where you get in-depth and practical techno-economic discussions that could disrupt the industry. That is deep tech as well as practice focused businesses discussions. TADSummit is often where companies first showcase, for example, Apidaze, Shango, Solaiemes, Matrix, WSO2.Telco, Wazo, Extrogene, STROLID, jambonz, Automat Berlin, Stacuity, and many more.

Following on from Joao sharing his experiences was funny, as Marten went through sharing how he too had learned the same points, and asking each time why Joao could not have presented 18 months earlier 😉

Marten is bootstrapping his business, AWA Network, which slows development, requires more dependence on professional services, and deployment at the customer than hosting a service.

In 2019 Marten presented at TADSummit “Improving the Experience of Realizing CXTech Use Cases.”  Where he presented the CPaaS proxy idea, which became AWA.

AWA sponsored TADHack in 2021 and at TADHack Berlin the winning hack was The AWA Effect by Ronny Graefe, Ali Barin, and Omer Faruk Aydin. The hack visualizes the savings made possible by AWA. It demonstrated the problem Marten knows well both in his business and in the projects for his clients.

The A2P SMS business is complex, there are over 1000 companies offering the delivery of A2P SMS. It’s not just Twilio, Sinch, and Nexmo (Vonage, well now Ericsson). Many operating within one country. But those small national or regional aggregators often offer one hop to the carrier. As well as supporting a range of performance.

For example, an SMS for an OTP (One Time Passcode) needs to be delivered reliably in seconds as someone is waiting to login, complete a transaction, or sign-up. Keep an eye on AWA Networks, and if your business involves A2P SMS, check them out it could saving you money and improve the performance of your business.

Time to ditch the ‘dumb-pipe’ – reinventing the core mobile network, to put developers first.

Mike Bromwich, CEO / Co-Founder Stacuity & Tim Dowling, Co-Founder Stacuity. Slides and Video. While the established IoT companies focus on making it easy to manage and operate IoT. Stacuity is exploring the next phase of IoT programmability.

As mentioned in Marten’s summary above, TADSummit has a long history of showcasing new companies, some of which go on to change programmable communications. We hope Stacuity follows in that tradition.

Mike highlighted the openness of cloud based platforms versus the closed nature of mobile networks. Then explained why telco networks have ended up with such a closed approach. And showed how the approach taken to exposing network capabilities limits what service exposure telecom networks can achieve. For example, send an SMS, make a call between numbers A and B, charge customer account, provision SIM. It’s the most common service capabilities.

Stacuity’s vision is a network built from the ground-up with developers as first-class citizens, think of a mobile core built like a CPaaS, designed as a platform not just a service, with configuration, control, observability and integration built in. While meeting all the telcos needs on interop, protection of the underlying network and mitigation of potential security and fraud risks.

They’ve set up partnerships with IoT and roaming partners, covered in for example CXTech Week 20 2022. As mentioned in Joao’s presentation they need to focus on a problem they know well and where they have direct access. As developers on mobile networks for many years they’ve lived first hand many of the limitations of building on mobile networks.

The key is finding the initial application(s), for example targeting system integrators building networks over mobile networks for clients, where observability is important. Check out Stacuity and have a think about relevant problems you see in your work and let them know.

Advancing Cloud Communications Well Beyond the Basics – Leveraging AI and ML in all aspects of programmable communications applications and media analytics.

Adnan Saleem, CTO – Software and Cloud Solutions at Radisys Corporation. Slides & Video. Radisys had an impressive TADHack this year with 23 hacks. In this presentation they shared how they make it easy for developers to use a range of programmable communications tools with the Radisys EDP.

EDP recognizes there are many different types of developer, so it provides a range of tools to help build stuff fast including simple APIs, SDKs, low code / no code visual interface and off-the-shelf applications. And its beyond the standard communications services (CPaaS, UCaaS and CCaaS) it also includes speech and video analytics, and conversation tools.

This approach works, look at the 23 hacks created at TADHack 2022 over just one weekend. So rather than implementing a CPaaS or partnering with a CPaaS provider and hoping developers sign up to a me-too offer. Take a developer-centric approach in giving them the tools to do a whole lot more to solve problems fast.

Shifting from Voice to Workflow Management.

Filipe Leitão, Global Service Provider Channel SE, RingCentral. Slides & Video. This is an important shift, communications is table-stakes. The focus is now on solving business problems using programmable communications with partners. RingCentral showed where the industry is going, ignore Filipe’s presentation at your peril.

The key points Filipe raised are the growth potential in the market, most enterprise are not yet using RingCentral 😉 The M&A through the pandemic, see my welcome presentation for some examples, has resulted in many similar offers. Joao also mentioned this in his presentation on how crowded the market has become. So does the market simply focus on pricing and feature differentiation?

Filipe neatly summed up a point raised in many of the keynote presentations, focusing on value by solving problems for the customer. In RingCentral’s case with its platform and ecosystem of partners. Or quoting Filipe, an “ecosystem of innovation that not only delivers results, but solves real-life problems for businesses and users alike.”

Future of WebRTC and Web3 Panel Discussion


The idea for this panel came from re-reading Tim Panton’s FOSDEM ‘Future of WebRTC’ and Arin/Alberto’s session on the WebRTC Landscape. And that got me thinking about the growing importance of WebRTC to Web3. David Dias (Protocol Labs) has pointed this out since he first presented at TADSummit in 2017.

There is always a danger in using the term web3, people can get hung-up on ‘What is Web3?’ It’s simply a marketing term that groups concepts such as decentralization, blockchain technologies, and token-based economics. Web3 a backdrop for the discussion about the evolution of WebRTC and what it means to programmable communications.

WebRTC has become the default for real-time communications over the web, and can support many of the concepts listed as Web3. So the intersection of WebRTC and Web3 is interesting to the TADSummit audience in understanding the future of WebRTC/Web3. We’re not saying Web3 is inevitable with the current protocols, nor will it replace Web2. Rather there are some characteristics that are shaping the web, where WebRTC could play a role. Now onto the panel…

Panelists are:

The panel brought a diversity of experiences wrt WebRTC. Jayne brings the developer perspective, Nikki brings a marketing perspective across gaming and telecoms, Arin brings an engineering / development perspective, Tim brings a broad set of experiences from strategy through to deep technology understanding, Florian brings both platform and implementation experience, and Jerod commercial experience in selling solutions built on WebRTC.

This was our only panel, and because of its importance we made it part of the keynotes. We ran the panel hybrid using Jitsi for the online portion, with participants in the US west coast, Germany, UK, and Portugal.

The simple question on the state of WebRTC highlighted the importance of diversity of the panel. From the engineering side the difficulty of interoperability (Jerod), performance, and enabling interchangeable components, e.g. echo cancellation (Florian). Complexity was raised as an issue with a range of views on the limited education resources for developers (Jayne), the steps open source projects and CPaaS providers have made for aiding developers to implement WebRTC (Jerod). And the overall comment that WebRTC has made building video applications much easier than in the past (Florian)

Tim had a great view on the status of WebRTC, it’s a success wrt every browser has an implementation. However, the APIs scare many people off. He reference an initiative he’s working on to bring general developer feedback into the standards process. Please take the time to answer the 5 or so questions in their survey.

On the issue of web transport replacing WebRTC. The bifurcation is WebRTC will continue to be used for end-to-end applications. However, for applications with a server in the middle, web transport could replace WebRTC for some implementation. Zoom is a good example in that it uses components of the standards, but the stack is very much Zoom specific, not WebRTC.

The ‘Who’s making money with WebRTC?’ discussion was interesting. Nikki shared her Subspace war stories. And asked of the panel, given the significant cost in implement WebRTC, even though it’s ‘free’, how can developers align the implementation cost with what customers are willing to pay? Here Jerod and Tim gave similar answers, by focusing on high value niches, e.g. keeping baby monitoring private and a quite unique broadcast experience with the ThunderDome to keep audiences engaged through the pandemic.

One application area I’m seeing more often and mentioned by Jayne is live commerce, it’s like the shopping channel just for niche audiences. In APAC there are personal shoppers where the audience is one, that is a rich person has someone shopping for them and they connect when a decision is needed. That’s back to high value niches focus mentioned earlier. Anyway, it’s a great panel discussion led by Arin on the status of WebRTC and some of the possible future directions of WebRTC.

With these links you can get to all the TADSummit content directly: Photos, Videos, Slides, Brief Agenda with videos and slides, Full Agenda with videos and slides.

And I can not say this enough, thank you to STROLIDBroadvoice / GoContactRadisysRingCentralStacuity, and AWA Network / Automat Berlin for sponsoring TADSummit 2022. And thank you to all the presenters and attendees in making the event excellent.