Re: clli vs mLUm

The MP4 registry also shows it as clli:

http://mp4ra.org/#/search


-Chris

From: Seeger, Chris (NBCUniversal) <Chris.Seeger@nbcuni.com>
Date: Monday, March 20, 2023 at 11:45 AM
To: Chris Blume (ProgramMax) <programmax@gmail.com>
Cc: public-png@w3.org <public-png@w3.org>
Subject: [EXTERNAL] clli vs mLUm
Hi,

maxFall and MaxCLL are known and “Content Light Level Information” or cLLi.  I should have caught this earlier:

In QuickTime atom:
https://developer.apple.com/av-foundation/High-Dynamic-Range-Metadata-for-Apple-Devices.pdf<https://urldefense.com/v3/__https:/developer.apple.com/av-foundation/High-Dynamic-Range-Metadata-for-Apple-Devices.pdf__;!!PIZeeW5wscynRQ!oDbHitmKWyxAUqiSNmf4aimGrz136V2sHY5-IdskK3SiEdpaG5UPPD4bmx9WcGZFvDnmImlf_msErrQJMZEbD-PW$>

In H.265 (it is not abbreviated but is indicated as “Content Light Level Information”:
https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-H.265-202108-I!!PDF-E&type=items<https://urldefense.com/v3/__https:/www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-H.265-202108-I!!PDF-E&type=items__;!!PIZeeW5wscynRQ!oDbHitmKWyxAUqiSNmf4aimGrz136V2sHY5-IdskK3SiEdpaG5UPPD4bmx9WcGZFvDnmImlf_msErrQJMU9IYhfF$>

In other groups like SCTE, we are abreviating it as clli.

Other references:
https://www.elecard.com/page/article_hdr<https://urldefense.com/v3/__https:/www.elecard.com/page/article_hdr__;!!PIZeeW5wscynRQ!oDbHitmKWyxAUqiSNmf4aimGrz136V2sHY5-IdskK3SiEdpaG5UPPD4bmx9WcGZFvDnmImlf_msErrQJMVW61XCV$>

If it’s OK to change, I will take this work on and change it ASAP.

Best,
Chris



From: Chris Blume (ProgramMax) <programmax@gmail.com>
Date: Monday, March 20, 2023 at 2:07 AM
To: Seeger, Chris (NBCUniversal) <Chris.Seeger@nbcuni.com>
Cc: public-png@w3.org <public-png@w3.org>
Subject: Re: [EXTERNAL] Meeting topics for March 19th, 2023
I don't know a great way to add cICP chunks at the moment.
I manually hex edited the file to add the new chunk.
There is a tool called TweakPNG which is very nice. It doesn't yet know about the cICP chunk but it is open source so I could add support if you would like.

On Sun, Mar 19, 2023 at 6:47 PM Seeger, Chris (NBCUniversal) <Chris.Seeger@nbcuni.com<mailto:Chris.Seeger@nbcuni.com>> wrote:
Hi Chris,

Can you instruct me tools for adding the CICP chunk to a png file so I can make independent tests?

Best,
Chris Seeger

From: Chris Blume (ProgramMax) <programmax@gmail.com<mailto:programmax@gmail.com>>
Date: Sunday, March 19, 2023 at 3:17 PM
To: public-png@w3.org<mailto:public-png@w3.org> <public-png@w3.org<mailto:public-png@w3.org>>
Subject: [EXTERNAL] Meeting topics for March 19th, 2023
Hello everyone,

I apologize for the late email. I had friends visiting this week and lost track of time.

Topics for our March 19th meeting:

  *   Friday, March 24th is when the PNG patent disclosure period ends. Our next step will be Candidate Recommendation, which requires us to close issues marked blocking-3rd-edition-cr<https://urldefense.com/v3/__https:/github.com/w3c/PNG-spec/issues?q=is*3Aissue*is*3Aopen*label*3Ablocking-3rd-edition-cr__;JSslKyU!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BF9AZhAo$> and gathering implementation experience.

     *   We now have 2 WPT tests live<https://urldefense.com/v3/__https:/wpt.fyi/results/png?label=master&label=experimental&aligned&view=subtest&q=png__;!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BOqHRzeE$>.

        *   Although, it looks like the cicp test harness is now failing? I'll investigate.

     *   We have reached to Mozilla<https://urldefense.com/v3/__https:/bugzilla.mozilla.org/show_bug.cgi?id=1821750__;!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BMjWsXgS$> and libpng<https://urldefense.com/v3/__https:/github.com/glennrp/libpng/issues/471__;!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BF1z1bPW$> to learn how they want to proceed with implementation.

  *   Acropalypse was announced.<https://urldefense.com/v3/__https:/twitter.com/ItsSimonTime/status/1636857478263750656__;!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BNACUwYI$>

     *   It works via cropping an image so it uses less data. When the newly cropped image is saved, the file isn't truncated down to the smaller size. This means data from the original, uncropped image is still in the file. Using the Huffman table from the cropped image, parts of the uncropped image can be reconstructed. (Or that is my understanding, at least.)
     *   We already have a bug open to warn about data stored after the end of the image<https://urldefense.com/v3/__https:/github.com/w3c/PNG-spec/issues/170__;!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BG4KcpFB$>.
     *   Should we encourage or require decoders to not display an image with extra data? That might have helped people catch their mistake. But the image is still valid and so not displaying it might be heavy handed.

  *   The Privacy IG has gotten back to us with suggested changes<https://urldefense.com/v3/__https:/github.com/w3c/PNG-spec/issues/282__;!!PIZeeW5wscynRQ!pRT_nke26r6RVrORfn4i-PEba9A9T_MjTfbZH3QHOuHEYSpEMkANxzSDFgNcPXLpNY9UEUVqEVL8BC0vSUsM$>. Would anyone be interested in mentioning these privacy concerns in the privacy section of the spec?

Thanks everyone!
See you at the meeting tomorrow.