search

FLIF-hub / FLIF

Free Lossless Image Format
Other
3.72k stars 229 forks source link

Will FLIF be submitted to standardization? #425

Closed mzso closed 2 years ago

mzso commented 7 years ago

Hi!

I wonder if FLIF will be pushed to become a standard.? (IETF, ISO, ITU, etc)

I feel like it would have a better chance to gain popularity.

saschanaz commented 7 years ago

This depends on #162, not sure about the progress...

mzso commented 7 years ago

@SaschaNaz commented on 2017. máj. 28. 21:30 CEST:

This depends on #162, not sure about the progress...

I see. There doesn't seem to be any activity though. It's weird that the owner is not interested since it seems to me one of the key elements of getting the format popularized. (Right after the format actually being good.)

BTW doesn't this count as a specification? http://flif.info/spec.html

jonsneyers commented 7 years ago

The spec at http://flif.info/spec.html is a good start, but it's not finished yet. Mostly the transforms still need to be written out.

I am interested in making FLIF a standard, but I'm not at all familiar with the process. Maybe someone who has experience in standardization could help us out?

maddin200 commented 7 years ago

Hi, every new standard starts as a proposal in the right ISO group - here probably ISO/IEC JTC 1/SC 29 Coding of audio, picture, multimedia and hypermedia information The standard has a different status from Working Draft to Final Comittee Draft. This process can take a while. Bye Martin

est31 commented 7 years ago

I have no experience with standardisation, but generally ISO puts its published standards behind a (big) paywall, while other standardisation bodies like the IETF make their standards available free of charge, which greatly helps if you just want to look up something about the standard.

jonsneyers commented 7 years ago

If we try to formally standardize FLIF, I think it would make a lot of sense to ensure that the standard does not end up behind a paywall – FLIF is not Free as in beer ;) but I see no reason for its standard to be non-free as in beer :)

Would it make sense to (try to) make it a W3C standard? Or maybe try to get it recognized somehow by the Alliance for Open Media (http://aomedia.org/) or Xiph.org (https://www.xiph.org/) ?

mzso commented 7 years ago

@jonsneyers commented on 2017. jún. 7. 16:12 CEST:

If we try to formally standardize FLIF, I think it would make a lot of sense to ensure that the standard does not end up behind a paywall – FLIF is not Free as in beer ;) but I see no reason for its standard to be non-free as in beer :)

Would it make sense to (try to) make it a W3C standard? Or maybe try to get it recognized somehow by the Alliance for Open Media (http://aomedia.org/) or Xiph.org (https://www.xiph.org/) ?

Any barrier to have it submitted to multiple standardization authorities? I see most formats are standardized by more than one authority. (Jpeg: ISO, IEC, ITU, PNG: ISO, IEC IETF) I wonder how they did it. (For MPEG-4 AVC wikipedia claims that the ITU/ISO standard is jointly maintained.)

est31 commented 7 years ago

I'm not sure about W3C's scope, maybe FLIF falls inside it, no idea.

About the Alliance for Open Media and Xiph, contacting them may never harm, but I guess that both are more actual places which create standards themselves, not who publish them in a standardized (:p) way.

Xiph themselves have submitted opus to the IETF for standardization, where the official place is now. The standards they did themselves were very inconsistently published: The Theora spec is a PDF, the ogg and vorbis specs are html, and the OGG skeleton spec is even just a wiki page. Some of it has been "standardized" by the IETF since, but only as informal document, not actual standard.

mzso commented 7 years ago

@est31 commented on 2017. jún. 7. 16:46 CEST:

Xiph themselves have submitted opus to the IETF for standardization, where the official place is now. The standards they did themselves were very inconsistently published: The Theora spec is a PDF, the ogg and vorbis specs are html, and the OGG skeleton spec is even just a wiki page. Some of it has been "standardized" by the IETF since, but only as informal document, not actual standard.

This probably didn't help them achieve popularity, which they didn't...

About the Alliance for Open Media and Xiph, contacting them may never harm, but I guess that both are more actual places which create standards themselves, not who publish them in a standardized (:p) way.

Well, it's not like AV1 was custom designed by AOM. It manly consists of different parts of codecs that some of the members were developing themselves before the alliance. They don't seem to oppose building on pre-existing stuff. Which they aim for IETF's NETVC.

IETF seems to be the most relevent since as an image format probably browsers are the largest market. (Not sure if FLIF is suitable for hardware encoding and digital cameras or not.)

maddin200 commented 7 years ago

Hi, ISO is not automatically a paywall: http://standards.iso.org/ittf/PubliclyAvailableStandards/index.html I must ask a colleague what the conditions are. Bye Martin

Traneptora commented 7 years ago

The ISO/IEC standards are often paywalled, but amusingly, you can find them elsewhere. In particular, the spec for MPEG-4 AVC is behind a paywall, but the spec for H.264 can be freely downloaded from ITU's website.

maddin200 commented 7 years ago

As far as I know everybody can make a proposal for an ISO standard with a application at the local standard organization (in Germany DIN and so on).

saschanaz commented 7 years ago

Would it make sense to (try to) make it a W3C standard?

PNG already is defined as W3C standard, why not FLIF? :smile:

It seems some early-day standards are first grown on WICG in these days before being a complete standard.

jessetrana commented 6 years ago

Any further movement on this? I've been watching this project with great interest for quite some time and I would love to see it turn into a standard - even if that would be a few years out.

saschanaz commented 6 years ago

AFAIK no progress. I opened a post on WICG, I hope someone be interested...

waldyrious commented 6 years ago

Doesn't this still depend on #162 ("Write out the detailed FLIF stream specification")?

Also, wouldn't the "affects bitstream" issues need to be resolved first as well? Most of them are under the "M3 - Future extensions" milestone, but #362 isn't.

saschanaz commented 6 years ago

I think the community needs some drive to do that, if someone from standardization groups show interests that will probably be a drive.

jessetrana commented 6 years ago

Looks like WICG have started to provide some good feedback about how/where to start, even if it is an uphill battle. I guess a question I have as a newcomer to the community is "Where is FLIF in 5 years ideally?" If standardization is a major desire, that's probably going to be a big, multi-phase push - which may or may not be anyone's passion. But if it's not, it will probably just take time/energy away from the other good stuff happening here.

mzso commented 6 years ago

@jonsneyers commented on 2017. jún. 7. 16:12 CEST:

Would it make sense to (try to) make it a W3C standard? Or maybe try to get it recognized somehow by the Alliance for Open Media (http://aomedia.org/) or Xiph.org (https://www.xiph.org/) ?

How about joining the alliance? Is there any barrier to that?

They not too long ago expressed interest in creating a still image format anyway.

saschanaz commented 6 years ago

AOM is already developing AVIF based on AV1 anyway. https://aomediacodec.github.io/av1-avif/

mzso commented 6 years ago

@saschanaz commented on 2018. jún. 1. 11:52 CEST:

AOM is already developing AVIF based on AV1 anyway. https://aomediacodec.github.io/av1-avif/

That's a lossy format though. If it can do lossless I doubt that it's effective. There's room for a lossless format the way I see it.

wingman-jr-addon commented 4 years ago

@jonsneyers I was excited to see your announcement about JPEG XL, which sounds rather like the standards-based successor to FLIF, in a roundabout way. I'm guessing that means this can be closed?

ziemek99 commented 2 years ago

I'm guessing that means this can be closed?

FLIF has been superseded by JPEG XL, and yes, that means this issue can be closed.

You can read more about JPEG XL on an official website, community website, dig into GitHub repository or join Jon, other JPEG XL devs and the community on a Discord server.

saschanaz commented 2 years ago

Should this repo be archived?

ziemek99 commented 2 years ago

No, there are still bugs to be fixed and issues to be answered. It's just not that urgent. Fortunately it won't consume Jon's precious time (which is already reserved for JPEG XL, his private life and other things) because I can now manage this repo's issues, PRs and releases. Personally I prefer to keep FLIF retired instead of archived, I don't like ending unfinished things. Maybe I'll finish exactly on a 1000th commit (or some other nice round number)? 😜

FUIF repo, on the other hand, should be archived.