Which tool for which type of videoconference?

Date of publication

"Take Zoom. It can handle hundreds of participants. My company also uses it worldwide. And yesterday we had a meeting with 1,500 colleagues." Let me guess: for most of us, a video conference looks a little different. But where are the differences?

No, this will not be a text about "why we don't use Zoom".

This already exists in other places. For example, here and here. We also discovered this one and found it quite good.

Today, we want to talk about the tool of choice for the right purpose.

1:1 meetings (or conferences with up to 10 participants)

Two people used to talk on the phone. Today they like to chat. Whether Rocket.Chat or Threema or Signal. Telegram, too, for all I care. If they want to see each other, tools should be chosen that create as few hurdles as possible and only focus on the relevant functions. From our experience, one that is particularly well-suited for this:

  • Jitsi Meet
  • Kopano Meet
  • Nextcloud Talk

All immediately provide a shared "space" for seeing, speaking, listening and, if needed, screensharing. Jitsi Meet can be easily integrated into Rocket.Chat and you can escalate directly from a chat to the "video call" :). Sometimes talking is faster than writing. And looking into each other's eyes helps to assure that both mean the same thing.

Nevertheless, it requires discipline on the part of all participants. Turning off microphones as a self-imposed "default setting" should be a matter of course when you have nothing to say. And the fact that everyone can "take away" a screen-sharing session from the other is intentional, because it is trusted that one does not do this if there is no reason for it.

The tools mentioned above can of course also handle larger rounds. Rough guideline: 5 - 15 (jitsi also considerably more.)

Adminforge runs a public instance in Germany that gives quite exciting insights into the current workload. The German-speaking Drupal community has just held its first virtual Meetup and at peak times 50 simultaneous participants were no problem.

Image
Grafana-Dashboard Jitsi-Instanz von adminforge

Trainings, sales events and organized meetings

"Many participants, but more restrictions for them" - this is how I would roughly summarize the requirements for such video conferences.

We are talking about roles here. Not everyone is the same anymore. That's another reason why there's no trust that people will sort things out among themselves. A "moderator" wants to be able to mute the participants. Maybe they should not be allowed to see each other at all. Neither the picture nor the presence at all. If you ask sales people, the simultaneous visitor numbers are of course huge. The infrastructure must be able to cope with 75+. All with 4K video and Dolby Surround. Below that, nothing works.

Of course, the room should not be opened by everyone. Only when the moderator is present may participants enter the room. They are welcome to enter by invitation only. Then they are invited in individually. Speaking is by hand signal only.

We consider room sizes of around 30 people to be realistic. They can then definitely switch on their video image and are also allowed to say something.

From the open source area, we can only think of BigBlueButton. This software is popular in schools, universities and also more and more in business. By the way: It's also our tool of choice and since we hate repetitions, we love the recording function. Record a good session once and play it back as many times as you like as a Recorded Show :)

The integration of BigBlueButton into content management systems like WordPress and Drupal or learning platforms like Moodle make the instance a Swiss Army knife for video conferencing.

But 100+ participants in one room? We wouldn't do that even with BigBlueButton.

1:countless

The Chaos Computer Club's Easterhegg event, the re:publica Berlin Internet congress, or the above-mentioned management address to employees: It's not about 1:1 or 1:15 or 1:50. It's about streaming. Feedback from participants must be captured differently and one-on-one conversations are certainly curated.

RTMP streaming is a good choice for this. If we stick with open source solutions, we use OBS and set up a restreaming server. We'll leave the distribution to the masses to YouTube and Facebook. They can do that.

Image
Ein Screenhot einer Videokonferenz mit mehreren musizierenden Menschen

And for conferences?

So far, I have mainly experienced a tedious stringing together of webcasts. What we love as hallway tracks at real conferences - the casual, informal encounters at the coffee machine, the (sought-after or accidental) getting to know other guests in the smokers' corner - is difficult in the digital version. The re:publica has at least tried to transfer the station's courtyard meetings into the virtual, and switched to single rooms in Zoom. Not pretty, but pragmatic. The WordPress community has introduced a kind of chat roulette to bring wildly unfamiliar people together. The kind we're familiar with from conference breaks: "I ran into you there once. Hello, stranger."

An entire industry is here in search of the right toolset. Let's do it with open source software!

And so with looking at and approaching each other at conferences?

I experienced VRML in the late 90s. Later Second Life. Today, the Oculus Quest is the first VR goggles available on the market that no longer require cables to a powerful Windows PC. I believe this is a gamechanger.

After my first experiences, the approaches of VRChat and AltspaceVR are exciting. But I am really happy that there is already an open source community in this area as well, which is trying to counter the future "Facebooks in VR" with Alloverse.

Imagine your avatar walking up to other avatars representing other people. You engage in a conversation - and when it gets boring, you can't just close the browser window. You say goodbye. Today we can always blame that on connection problems :)

But for real. That's a whole other way of being together.

So skip everything and go straight to VR?

It's too early for "Forget video conferencing!" today. Think about what you need. Choose the tools that are right for the job. Choose free software that is under your control. Don't trust advertising promises. Make your own experience.

Profile picture for user DeepL

DeepL is a deep learning company that develops AI systems for languages. The company, based in Cologne, Germany, was founded in 2009 as Linguee, and introduced the first internet search engine for translations. Linguee has answered over 10 billion queries from more than 1 billion users.

Profile picture for user luckow

Stephan Luckow

Stephan is an open source evangelist and constantly curious about technologies. Thematically, his blog posts can best be summarised as "curiosity satisfied".