zlacker

[return to "Tell HN: We should start to add “ai.txt” as we do for “robots.txt”"]
1. qbasic+Ol[view] [source] 2023-05-10 14:11:34
>>Jeanne+(OP)
Your HTML already has semantic meta elements like author and description you should be populating with info like that: https://developer.mozilla.org/en-US/docs/Learn/HTML/Introduc...
◧◩
2. burnte+O51[view] [source] 2023-05-10 17:25:56
>>qbasic+Ol
How do I add a semantic definition in an HTML tag to a JPEG, or MP4, or WAV, or any non HTML format? HTML tags fix HTML, not other formats.
◧◩◪
3. cornst+xe1[view] [source] 2023-05-10 18:04:56
>>burnte+O51
JPEG has EXIF, MP3 has ID3 tags, MP4 has ilst, MKV has Tags, etc. We don't need xkcd/927 for these other formats that already have standard metadata mechanisms.
◧◩◪◨
4. Rambli+6r1[view] [source] 2023-05-10 18:57:38
>>cornst+xe1
Yeah, and if you make it that complex to extract consent you won't get any. Think one step ahead maybe. One switch, one thing to parse.
◧◩◪◨⬒
5. cornst+7C2[view] [source] 2023-05-11 01:55:26
>>Rambli+6r1
1. OP said “what it is about, when was it published, the author, etc.” That’s what these mechanisms already cover. Consent is an interesting possibility that I’ll admit something like ai.txt might be better for, but my post was largely focused on the OP.

2. These are all complex formats. If you want to ingest and process them then you already have to build all the hard parts. Getting the metadata out is dead simple compared to parsing, decoding, and then processing an image, for example.

[go to top]