Hacker News new | past | comments | ask | show | jobs | submit login

The code shows the first three video formats being available and prioritized before gif, but the sample videos do not include gifs. It'd be nice to see the quality difference between the format that is being discouraged and the format being encouraged.

Personally, I love gifs. They work with no fuss.

GIFs work with no fuss only for small file sizes. Once you get over a fairly low threshold the size differences have an enormous impact both for network performance — like a GIF takes tens of seconds longer to start — or CPU load because all common browsers even on low-end phones have hardware video decoding but have decode GIFs on the CPU. Since people increasingly use them for video clips I encounter that more than I would have expected 10 years ago.

At similar bitrates, they would be barely disinguishable visual garbage. At similar visual qualities, they'd be 10-20x filesize. It would be useful for the author to make this point by example though.

Aside from the size issue, GIF only support 256 colors per frame, so photos and real-life videos almost always have visible dithering artifacts.

It's a terribly shitty format, but remains popular because it has universal support and so works with no fuss. We need to get the same amount of support for a better format.

Unless you're willing to backport that support to 10+ year old browsers, it will never happen. You need UNIVERSAL support. Edge cases, corner cases, and plain old "bad ideas" included. This is what the incumbent has going for it.

The polyfill principle addresses this. As long as transparent fallbacks exist, there's no reason new technology can't be rolled out for the benefit of the significant userbase that can take advantage of it.

Anyone who cares about their data bill should be asking very serious questions about why we're still stuck with the massively inefficient 32-year-old GIF format as the only widely supported way to display an inline animation, especially as many superior alternatives like FLIF, BPG, etc. have been released over the last several years.

GIFs waste absolutely ungodly quantities of bandwidth. Any semi-modern video codec is at least an order of magnitude more efficient. Any way you slice it, the continued dominance of GIF is, at best, an egregious oversight, and it has a direct dollar impact on anyone with a metered connection (this is your phone, but increasingly likely to be your home connection too).

As a community, we need to make an alternative to GIF a real priority.

Why? Getting the major browsers on board going forward has always worked fine before. Most of them update automatically now, anyway.

I like GIFs too, they do indeed work well on all platforms. But I agree with OP, in that we should be switching over to more sustainable standards. Think about the energy savings we could get across the web if even 25% of GIFs were converted to newer and more efficient formats.

I have no problem being fed the fad of the month codec, as long as a reliable fallback is available.

Hold up, we have those for codec's now? Gee, what's next? Linux distro flavors?

Author here: I've uploaded the GIFs here https://github.com/singhkays/its-time-replace-gifs-with-av1-...

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact