It's a story repeated ad nauseam, but the founder of Zoom came from Cisco, after having been in charge of WebEx there. Cisco had refused to let him do what he saw needed done to radically improve WebEx, and so he left and created Zoom and did it there.

Cisco just saw a product that was successful at the time, and a senior exec that wanted to spend a lot of money to ultimately still only have video conferencing software. They were wrong with their decision, but it's an entirely understandable one.

The weird thing is that even with Cisco $$$ Webex still isn't as good as Zoom at streaming. I wanted to watch a Youtube stream simultaneously with friends and tried it out with Webex, Hangouts, and Zoom. Hangouts was unwatchable, maybe 1-2 frames/second, but hey, it's a "free" product. Webex made it possible to see what was going on, but honestly it was pretty crappy. Zoom looked more or less like the original stream. Pretty amazing for a product that until last year was mainly known security holes.

I've been using Zoom via work for over 4 years now. It has been a stable and very good video conferencing solution for all that time. It is the first time I've honestly dealt with video conferencing software that "just works", which is a huge deal. I've never sat in meetings with people dithering about trying to get it to do what they needed.

Have you tried using Google Meet instead of Hangouts? I'm just curious if it's noticeably different.

I think Meet is just a rebranding of Hangouts. Meet just sounds better for enterprise than Hangouts.

Zoom is still in "growth phase" which means they aren't under a lot of pressure to turn a profit at this particular moment. So they proactively offered their free service to schools, the free tier isn't time-limited, plus their corporate product is basically the same as the consumer/free product.

And on the other side, Google was totally caught napping and assumed that everyone knew about Meet. https://www.nytimes.com/2020/04/24/technology/zoom-rivals-vi... "[Google’s chief business officer] Mr. Schindler tried placating the engineer’s concerns, the people said. Then his young son stumbled into view of the camera and asked if his father was talking to his co-workers on Zoom."

Zoom actually is profitable and was one of the rare technology companies to be profitable at IPO. Although they do offer a lot of free services they nailed the freemium line - 40 minute free calls - since the most effective business meetings are 45 minutes.

I would say the most important part, and where every company has shot themselves in the foot is the need for an account. With Zoom you can join meetings with just a link/code, and while it may seem trivial to us thats a huge plus for nontechnical people.

That and the fact Zoom had seamless integration with large meetings with relatively little-to-no performance/quality costs is a huge benefit. The luck of timing & advertising is what pulled it all together.

So much this. My kid's school went with Google Meet for the online class chats, and gave us meeting codes. So far so good. I get the app, wants a Google account. I would have stopped right there if it was an option. Then, constant problems and apologies from presenter, just like in the bad old days of laptop presentations and projector resolutions. App is not the best either - rookie UI mistakes like the meeting code text box obscured by a CTA. A UX so bad that it boggles the mind.

The school district ended up distributing their own Gmail credentials to use for the meetings, too, which was behind a hard-to-find edu-website login. That's a lot of friction that Zoom solves.

This is exactly it.

In March, when everyone suddenly wanted to do group video calls with their friends and family, most of whom are nontechnical, we settled quickly on Zoom for ease-of-access. Of all the options, it was the easiest one for me to get my parents into a family call.

You message someone a link, and they click the big buttons until they are in a call. No account, no confusion. We had 100% success rate on getting people in.

Also, it's asynchronous unlike something like FB messenger or Facetime (ie it doesn't push you the meeting, you can pull it at your convenience).

This was another annoying feature of Meet. The meeting organizer is special and needs to be connected to let people in.

I think this makes sense, and it is indeed a major contributing factor to the UX. The hunger for data from all the other players essentially hampered the product adoption.

It's also a huge plus for places that lock down their computers so you can't install software on them.

I used to use WebEx on a personal laptop for work conversations with other organisations because my work machine wouldn't allow it to run properly.

I would really like to know what went wrong with Google especially. They've had years to develop Hangouts and Meet, yet it's so, so behind in both performance and quality. Are both of these products just riddled with technical debt to the point that these improvements take significantly longer than normal?

I would venture that the problem with Google's products is that they run in the browser and in spite of all the improvements in browsers these days its just not ready yet when you need the kind of quality and performance that huge meetings demand.

Also and perhaps more importantly, Google's approach is tied into its own ecosystem but anybody can join a Zoom meeting with just a link/code. Google had everything they needed but are too tied into the everything on the browser mindset and their own ecosystem to see they're shooting themselves in the foot before their horse is even out the door. At this point its too late, Zoom's aggressive pre-marketing, simplicity, and ubiquity is a real force now and everyone else is just playing catch up.

> I would venture that the problem with Google's products is that they run in the browser and in spite of all the improvements in browsers these days its just not ready yet when you need the kind of quality and performance that huge meetings demand.

Disclaimer: I have not investigated how Zoom and other video conference apps work, but I do work with real-time video.

I don't see why the browser would be the limiting factor in performance / quality for videoconferencing.

What is the local client doing? Recording video and audio, and playing a real-time video stream. Sure, you want it to be low-latency, which involves its own bag of tricks.

But native-code libraries and/or hardware are doing all the heavy lifting (like H.264 encoding or whatever they use), the browser (and Javascript) is just the UI.

You have much more leeway if you can tune ffmpeg decoding settings manually, skip frames whenever you want, etc than if restricted to what webrtc gives you. Also Zoom is built with Qt which is pretty efficient :-)

So it seems the limitation is more the API for recording video, is this correct?

Even the Hangouts native app was a slow battery hog on anything but the absolute latest generation phone, so it's not a browser problem.

I have repeatedly found Meet much higher quality than Zoom - straightforward to launch from the browser with a single click, clearer settings, better video quality, and no scummy installed app practices.

That might be because of my location or some other factor, but I really don't understand why people find Zoom so much better - it doesn't match my experience at all.

I doubt it's anything more than the cyclical nature of trends, and chance. There was already a movement away from Zoom to platforms that were less closed, but then the pandemic happened.

But the movement to leave Zoom happened due to the prominence of Zoom in many B2B companies that didn’t realize they were risking their security. Just reaching the status Zoom reached in 2019 is bonkers considering the competition had been at it for decades.

They do have some patents https://patents.justia.com/assignee/zoom-video-communication... so there was some innovation required. I thought they may use middle out compression, but couldn't see its mentioned

Patents are an extremely poor signal of required innovation; I strongly suspect they’re worse than random. Most innovation is retained as trade secret rather than patented. Patents are more an indicator that you’re a big business, or that you are operating in a domain where hostile use of patents is rife (and video is one such area).

I just did a very quick skim through those patents (just reading what’s on that page and not the actual claims), and these are my impressions. ’390: obvious, prior art may also exist, as what it describes has been a standard technique even within video codecs for many years. ’900: obvious, also an abstract idea so it shouldn’t be patentable. ’511: obvious and probable prior art, the technique has been used for DRM for years. ’454/’218: no idea, could be valid, but (personal opinion) it doesn’t sound like it should be patentable anyway.

Funny, patents were originally intended to stop people keeping trade secrets, now so many patents are just nonsense, all the important stuff is kept secret again.

Network software has made it possible to keep trade secrets perfectly. When you deployed something physical, people could analyse it in detail, trying to figure out what the ingredients to your drink were, pulling apart your widget and measuring it very precisely, that kind of thing. Some would be easy, some would be hard, but it was all at least physically theoretically possible. So the patent system said “save us the trouble of figuring it out, and we’ll grant you a monopoly on this particular approach for a while”.

But with network software, the interesting part is out of reach, so you can’t reverse-engineer it.

M$ charges money.

Cisco charges money.

Google discontinued their video/messaging services too many times for people to get used to.

Heck for a moment I thought "Houseparty" was going to be a new social media, but I think their push notifications were too annoying.