Hacker News Reader

Comment by jcutrell | Hacker News Reader

I think the mp3 is responsible for much of the positive momentum of podcasts and audio in general. (That's pretty obvious.) But there is not a widely accepted analog in video. Video encoding is complicated. And since we are at a particular juncture where video is large enough relative to internet speed, compression and encoding are still very important.

So the complexity of that is what makes it hard to go totally open in my opinion, at least from the creator standpoint.

That and the fact that there is just too much momentum with YouTube's audience volume, if your goal is viewership.

View on HN

Video also has the twist that, unlike audio, it has to support a bunch of different hardware configurations. You’ve got to support multiple video codecs, multiple resolutions and bitrates. And to be good you might even be serving several of those variants to the same client in one viewing—which means you have to intelligently chop up the video into smaller pieces. Plus the supporting client side tooling is more important—for example those little thumbnails that show up while seeking through a clip.

Audio is easy. Throw up an mp3 and you are done. The basic fundamentals of an audio client have remained largely unchanged since the invention of the tape deck. Video is a while different animal.

That being said, moving to an RSS based model for video would be pretty interesting. I just imagine there would be a lot of work on whatever system consumes those RSS files to make the video playable across the wide spectrum of video players.

You're suggesting that in 2024, I can't just throw up an mp4 or webm and "be done", in the same sense that an mp3 covers "be done" for audio?

Basic video is good enough for a lot of purposes. But with minimal gear and software, I can clean up most speakers with very little work in audio (and there are better AI cleansing tools these days as well). For a given quality level, the bar is much lower for audio only.

This really isn't true. Mixing/mastering if you want to target:

  * in ear devices
  * vehicle audio systems
  * phone speakers
  * laptops
  * mid-range home stereo systems
  * high end home stereo/studio monitoring

is quite complex to get right, and generally you can't optimize for more than one at a time. That's even more so if you actually buy into the "immersive audio" hype, where playback is not even stereo anymore.

Audio can certainly get complex. But per the upthread query I'd argue that it's still easier to get understandable audio in an interview in a quiet location than it is to shoot video, especially outside of a studio setting.

and yet ... if the video quality is sub-par people care <--- this much --->, whereas if the audio is sub-par people care <----------- this much ------------->

Fair enough. For people speaking, we'll tolerate mediocre video with good audio over vice versa.

Arn’t video podcast still a thing? I mean I don’t watch any myself, but the technology has been there for at least a decade. If people aren’t using it, it’s probably more of a user experience or discoverability or other issue.

Reply on news.ycombinator.com