It's somewhat odd to suddenly see lots of excitement in the 'continuous-tone still images' space, and I can't help but think that on some level, JPEG the group to wants to retain relevance at a time when the world is finally looking poised to move on from using its oldest work.

A still image file containing one frame of a video codec has been long theorized, pushed very aggressively since 2010 by Google as WebP (i.e. VP8-in-RIFF), and as a fully-functional thought experiment by ffmpeg's Fabrice Bellard since 2014 in BPG (HEVC-with-custom-header), the latter of which jumped the gun on HEIF before it was done.

Now that HEIF is here, we have a better container encapsulating an advanced codec, and suddenly we have hardware support with premium phones generating files in this format. The barrier is HEVC's hefty licensing pool, but a parallel effort exists in AVIF [1], which takes a frame of the royalty-free, HEVC-competitor AV1 video codec, and stuffs it into a HEIF container.

It seems a bit superfluous for JPEG to barge into this space right now; I'm confident they'll do technically solid, and even innovative work, but proliferating an incompatible implementation right when it looks like there's a realistic chance of a newer still-image codec replacing one that dates back to 1992 is a bit counterproductive.

But perhaps I should be more optimistic. Their calls-for-proposal document [2] lays out recent competitors, admits their technical merits, but concludes that they're not in widespread use. Then, it lays out that their vision to produce a state-of-the-art format that will achieve widespread adoption, helped by a "royalty-free baseline".

[1] https://news.ycombinator.com/item?id=16803363 [2] https://jpeg.org/downloads/jpegxl/jpegxl-draft_cfp.pdf