In the summer of 1996, my team at CNN Interactive produced what was, to our knowledge, the first commercial internet video webcast. We covered the Presidential conventions — both of them — live, streaming video over a medium that most of our audience accessed through a 28.8k dial-up modem.

To put that in context: 28.8k means 28,800 bits per second. A standard definition video stream today runs at roughly 200 times that. What we were delivering was small, blocky, and buffered constantly. It looked like watching a news broadcast through a keyhole in a fogged-up window. We thought it was the most remarkable thing in the world.

The technology we were working with was barely real. RealNetworks — then called Progressive Networks — had released RealAudio in 1995 and RealVideo shortly after. Encoding live video into a format that could travel over phone lines required hardware that filled a rack and threw heat like a furnace. The stream could drop without warning. The servers could fall over. At any moment the whole thing could just stop.

It didn’t stop. We ran for days. People watched — not many by any standard that would matter now, but they watched. Journalists, political junkies, early adopters who had fast enough connections to get something approaching a coherent image. We got press coverage. We felt, with full sincerity, that we were witnessing the beginning of something.

What I remember most isn’t the technology. It’s the feeling in the room — the particular electricity of people doing something that has never been done before. There’s no manual for that. You make decisions based on incomplete information, under time pressure, with equipment that wasn’t designed for what you’re asking it to do. You solve problems you couldn’t have anticipated because they couldn’t have existed until this moment. And then it works, and for a few seconds everyone in the room looks at each other with the same expression: we just did that.

I’ve been chasing that feeling my whole career. Some of the rooms I’ve been in since have been bigger, with higher stakes and larger budgets. A few of them have had it. Most haven’t. It turns out that feeling isn’t a function of scale or resources. It’s a function of doing something genuinely new with people who care about it.

The 1996 webcast was watched by a fraction of a fraction of the audience that watched the conventions on television. Nobody remembers it. The technology it was built on is completely obsolete. But somewhere in the Internet Archive there are probably a few frames of blocky, compressed video of a Presidential convention that were delivered over a phone line to someone sitting at a desk in 1996, watching the future happen on a 15-inch CRT monitor.

I’m glad I was there for it.

This is the first in an occasional series about the early years of the web — what it felt like to build things when the medium was new enough that nobody really knew what they were doing.