June 2022 Summaries

This post discusses how to add a text chat feature to a custom React video application using Daily's real-time video and audio APIs and the new custom React hooks library, Daily React Hooks. The goal is to allow participants to send and receive messages during a call. The app will have a toggleable chat UI, notifications when a new message is received by a participant who doesn't have the chat open, and the ability for new participants to read older messages after joining the meeting. The chat feature includes a form where users can input their messages, which are then broadcasted to other participants on the call using the `useAppMessage()` hook from Daily React Hooks. The app's state is managed using React hooks, such as `useState` and `useCallback`. The post assumes that the reader has basic knowledge of React and JavaScript.

Daily's WebRTC calls utilize a signaling server to manage call state in both Peer-to-Peer and Selective Forwarding Unit modes, with the latter also sending and receiving audio and video data between participants. A single server can function as both a signaling server and an SFU in a given region, which is determined by geographic location, potentially impacting latency experienced by call participants. The addition of an EU-based "eu-central-1" region in Frankfurt, Germany offers room and domain owners more control to optimize their participants' call experience. A signaling server acts as an intermediary between two peers, tracking call state information, while an SFU receives all participants' video and audio tracks and forwards them to relevant call participants. Developers can manually toggle between P2P and SFU modes using the API's setNetworkTopology() method, but letting Daily handle these states is often the most efficient option. The location of a signaling server can affect WebRTC call performance, with default settings potentially leading to latency issues for participants in different regions. The introduction of the "eu-central-1" region aims to address GDPR concerns related to transferring call participant data outside of the EU.

Daily's APIs allow developers to configure their domains, rooms, and meeting tokens for building WebRTC video and audio-only apps. Domains represent the top-level object in the Daily API and are associated with multiple accounts, while rooms host video calls and have unique room URLs that include the domain and room name. Meeting tokens provide a way to update session configurations on a per-user basis and override room settings when the same property is set on them. The order of priority for these settings is domains (most general), rooms, and meeting tokens (highest priority). Developers can use the Daily dashboard or REST API to configure these settings and override shared properties in various scenarios, such as turning off prejoin UI or enabling recording for a specific meeting owner.

This post is part two of a series on building a custom React video chat app using Daily's real-time video and audio APIs, and Daily's custom React hooks library. In this post, we'll be adding a prejoin UI to our application, which allows participants to preview their video before joining, select their preferred devices (camera and microphone), and set their username. We'll use the preAuth() and startCamera() daily-js methods to authenticate ourselves and see ourselves on the screen without having to join the call first. The new <HairCheck/> component will be implemented using Daily React Hooks extensively.

Daily and Confrere are now one company, merging their expertise in video APIs and infrastructure with the telehealth platform's innovative product work. The combined team will focus on building a full-stack of enabling technologies for live and low-latency video, expanding global infrastructure, optimized media handling and routing, and features like recording and transcription. The goal is to make video calls always work and video features easy to deploy on any device, network, anywhere in the world, with a focus on reliability, user experience, and customer satisfaction.

Today, Daily is announcing expanded large call support, allowing developers to build video and audio experiences with up to 300 participants, all with cameras and mics on. The platform also supports interactive live streams with up to 15,000 viewers and 10 presenters at a time, with real-time latency of 82ms median end-to-end network latency for all sessions across its global user base. This new support is available in Daily Prebuilt, which lets developers add a fully working video call to any app with just a few lines of JavaScript code, or through the lower-level APIs. The platform aims to power large-scale use cases such as education, events, social, and live commerce applications, while providing full control over UX and interactivity for developers. With this release, Daily is raising limits on participants, participants with mics/cameras/screen shares on simultaneously, and maintaining its best-in-class real-time latency, making it the only video API that allows 300 active participants with complete control over cameras and mics.

RTMP is an older technology widely used for sending video to a live session, but rarely used for viewing streams. HLS, on the other hand, is commonly used by platforms like Twitch to deliver live streams to large audiences. WebRTC is a newer standard that supports interactive use cases like video calls and has low latency, making it suitable for real-time applications. The choice between RTMP, HLS, and WebRTC depends on factors such as latency requirements, scalability needs, and the type of application or service being built. While HLS offers advantages in terms of resilience to network issues, it comes at a cost of higher latency compared to WebRTC. WebRTC's low latency makes it suitable for interactive applications like video calls, but its trade-offs include lower average video quality and more complex infrastructure requirements.

We're launching our next version of pricing, focusing on simplicity and transparency at scale, based on customer feedback that there was room for improvement in making their lives easier. Our new pay-as-you-go pricing removes friction for developers, providing 10,000 free minutes to start with support, simple usage-based pricing, volume discounts that scale as you grow, and add-ons for premium support, SLAs, and enterprise functionality when needed. We've designed our pricing around simplicity, flexibility, and enterprise benefits, offering a cost calculator, no extra charge for HD-quality video, and HIPAA compliant video for healthcare teams. For existing customers, there's no action needed as they'll be matched with a new add-on at an equivalent level of service, while new customers can immediately take advantage of our new pricing, including $1,000 in free credits and dedicated support for startups.