The Story of Plaudere: From Concept to a Working Website

This is the story of how Plaudere was imagined and built. It is a story that started long before the first line of code was ever written.

I clearly remember the first time I visited a website in my high school's computer lab. During the late 1990s, I used a search engine called Altavista combined with Microsoft Encarta, which was the main source of information for high school students at that time. In those early days, the internet was very simple, and I never thought that one day I would build a website myself.

Later, at university in the early 2000s, an introductory computer software course taught me some basic HTML. This was the time of Web 2.0, which introduced social networks like Hi5, MySpace and the first version of Facebook. It was exciting times, as these early websites gave us a hint of what was coming in the future. In the late 2000s, services such as YouTube and Facebook became very popular, and most of my friends used Messenger to chat. Mobile phones also slowly started to get internet access. Still, the idea of developing a website for myself was not in my mind.

The First Steps in Web Development

During a professional experience in a procurement department, the team used a static web to run an intranet. This was a private website for the company's employees where documents, links, and guides were available to support our daily tasks. In the early 2010s, I became responsible for this intranet. Without knowing any HTML, I found myself looking at the code. It was not clear at all, but I started to see patterns, tags and links. With some creativity, I was able to change the website completely, making it more modern and adding useful sections for the buyers, including multimedia content and documentation. However, my main programming tools at that time were MySQL and Visual Basic, along with some Excel macros. Web development was still not a key skill for me.

This experience, combined with my love for music, led to a new idea. From time to time, I got together with musician friends in person to practice, write, and record music. When I started a project management course along with advanced English classes, I felt a strong need to find a solution to play music online. At that time in the early 2010s, there were some tools that allowed musicians to play together, including websites and desktop programs. But there were problems like a delay between streams, a need for a high internet speed, and user interfaces that were difficult to use, with complicated settings and parameters. Because of these issues, my friends and I could not find a good away to collaborate online. It also became harder for me to play live with them because of my increasing work responsibilities.

I researched the state of web development at that time and found it quite challenging, especially because scripting languages such as HTML and CSS were still fairly new to me. A colleague told me to learn PHP along with HTML, CSS, and JavaScript using w3schools, a great online platform, but because of some big changes in my life, learning to code became less important for my career and I put it aside.

From MBA to the Birth of an Idea

The 2010s were a decade of professional growth for me. I completed my MBA and started my career in Europe, but in the late 2010s, far from the friends I used to make music with, the idea of collaborating online came back to me. I felt a strong need to find a solution and decided to do something about it. After exploring different options, I was not satisfied with the existing solutions for real-time streaming between users, so I decided to try creating my own. When I considered what kind of software I could build, I realised a desktop application would not be as easy to use and share as a website. With that, I decided to start learning HTML, CSS, and JavaScript, beginning with programming books from the library and then taking online courses to learn the fundamentals.

At first, it was difficult to learn how to build websites. But during the pandemic, the lockdown gave me time to think about this crazy idea of helping musicians play together online. This motivated me to start developing "Plaudere", which is a Latin word meaning "to applaud and approve", connected with the idea of sharing a performance with an audience and having them applaud after the show.

Developing Plaudere: The Prototypes

From early 2020 to 2025, I created four different prototypes while working on the idea for Plaudere. At the beginning, Plaudere was more focused on live streaming. I planned to use open-source libraries like MongoDB for the database, Node.js and Express for the server, and APIs like the Web Audio, Media Recorder, and Web Socket APIs to capture and send a stream in small parts to the audience. The first two prototypes were SPAs (Single Page Applications), which I believed would make it easier to manage the entire streaming session. However, they turned out to be more complex. A single page requires managing the user interface with JavaScript, which left less capacity for handling media content like audio and video.

The next prototypes used a more traditional website with multiple pages managed by an Express router and controller in Node.js. It used cookies to remember the session, a login with Google or Microsoft, a database (this time using MongoDB), and a background worker to do calculations. Even though I hosted it on Google Cloud using Google App Engine, I later preferred a specific hosting service called Railway. Finally, I launched the domain "Plaudere.com," and this is how the fourth prototype is moving forward.

Challenges and My Approach

From the first to the fourth prototype, I had to spend a lot of my personal time understanding and solving problems. I can mention a few of the key challenges. For example, an SPA, if not managed well, can make the website slow. The server-side JavaScript code using Node.js and the Express framework is very open, which can lead to inefficient programming. In later prototypes, I started to create reusable parts of the code to avoid repeating myself. I made the server prepare the HTML pages as much as possible to reduce the work for the client-side JavaScript.

I know that using web frameworks like Bootstrap or React and Next.js would have made my life easier. These frameworks help build websites with fewer steps but have a more complex structure, as they need a compiler and can slow down the website because of pre-loaded libraries. I looked at what is used for web development today, and I decided to build the website on my own using pure HTML, CSS, and JavaScript for the frontend and Node.js and Express for the backend. My main goal was to learn the basics of web development and to control the website with as few libraries and external tools as possible. This is a difficult goal to reach today, but the exercise helped me decide if a library or a third-party API was truly necessary or if I could do it with some extra JavaScript. In the future, I plan to use more modern tools, but step by step, this is a starting prototype, and it will only use more advanced solutions if needed to improve the experience.

The streaming tests also taught me many lessons. For a shared streaming session, the website's interface and background tasks must be flexible enough for different streamers. However, adding features like a 'ladder' to join another stream in real-time was very difficult. I included it in the first three prototypes but removed it from the fourth, until I could find a way to create a working two-streamer 'ladder' approach. I decided to focus on building a stronger foundation for sending and receiving streams and avoiding errors when switching between streamers. I also added the ability to combine the stream with video from a camera or a video file, which gives the streamer more options. I am sure that in future versions, I will look again at the streaming technology to expand the functionality and make the stream feel more like a theatre show than a simple stream through a website.

I also explored adding interesting features to the website, such as an AI agent that could read user content, answer questions, and provide support on how to use the site. To do this, I had to test some open-source LLM models on the backend to see if it was feasible given the site's constraints and infrastructure. It was. But as you may know, integrating an AI agent is a new frontier, and this is just the tip of the iceberg. I will need to do more research to create a more robust solution for the website.

Conclusion

In this post, you've learned about the ideas and work behind Plaudere. I'm very happy with this fourth website prototype, which has expanded the concept beyond a simple live-streaming website. Plaudere can now be described as a mix of a blog and a streaming platform, allowing people to create spaces and posts and use them to interact with an audience through a live show, with more features coming soon.

Let's see what's next for Plaudere. I hope this post has helped you understand the idea and the first steps of its development.

Joe Esteves