Carsten?
", "time": "2025-04-30T14:01:49Z"}, {"author": "Joe Hildebrand", "text": "I can hear you
", "time": "2025-04-30T14:02:28Z"}, {"author": "Rohan Mahy", "text": "hi
", "time": "2025-04-30T14:02:50Z"}, {"author": "Rohan Mahy", "text": "i proposed focusing on a single topic today (any topic) and driving to completion.
", "time": "2025-04-30T14:03:59Z"}, {"author": "Joe Hildebrand", "text": "agree w/ rohan above. i haven't had time to digest last night's EDN PR, so can it not be that, please?
", "time": "2025-04-30T14:08:21Z"}, {"author": "Rohan Mahy", "text": "A= number of simple values (in economic 16)
\nB= number of straight refs (in economic 32)
\nC= number of inverted refs (in economic 8)
I can adapt dns+cbor and rerun my experiments with the new packed proposals, but not before end of May.
", "time": "2025-04-30T14:14:51Z"}, {"author": "Rohan Mahy", "text": "Thanks Martine. That would be very useful.
", "time": "2025-04-30T14:16:06Z"}, {"author": "Rohan Mahy", "text": "Do you have a feel for what would work for you. How sensitive is your application to an extra byte per reference?
", "time": "2025-04-30T14:16:56Z"}, {"author": "Rohan Mahy", "text": "Vadim, could you please post in the chat the original (unpacked) CBOR
", "time": "2025-04-30T14:20:23Z"}, {"author": "Carsten Bormann", "text": "[\"foobarbaz1\", \"foobrbaz2foobarquux\", \"foo\", \"bar\", 'bar', \"baz\"]
\n0000 86 6a 66 6f 6f 62 61 72 62 61 7a 31 73 66 6f 6f |.jfoobarbaz1sfoo|
\n0010 62 72 62 61 7a 32 66 6f 6f 62 61 72 71 75 75 78 |brbaz2foobarquux|
\n0020 63 66 6f 6f 63 62 61 72 43 62 61 72 63 62 61 7a |cfoocbarCbarcbaz|
Maybe on the mailing list after?
", "time": "2025-04-30T14:20:49Z"}, {"author": "Orie Steele", "text": "This is like a \"codecs\" approach for CBOR... sorta.
", "time": "2025-04-30T14:24:32Z"}, {"author": "Carsten Bormann", "text": "it's redefining all that CBOR (and UTF-8!) is, which of course can be done, but is quite different from cbor-packed.
", "time": "2025-04-30T14:25:21Z"}, {"author": "Orie Steele", "text": "It requires changes to the encoder / decoder correct?
", "time": "2025-04-30T14:28:05Z"}, {"author": "Carsten Bormann", "text": "Yes
", "time": "2025-04-30T14:28:10Z"}, {"author": "Orie Steele", "text": "reminds me a bit of multicodec: https://github.com/multiformats/multicodec
", "time": "2025-04-30T14:29:05Z"}, {"author": "Carsten Bormann", "text": "No, it is not that bad
", "time": "2025-04-30T14:29:16Z"}, {"author": "Carsten Bormann", "text": "Great ideas, but not all fit very well into CBOR
", "time": "2025-04-30T14:29:39Z"}, {"author": "Carsten Bormann", "text": "Using the gaps in UTF-8 as a bytecode certainly is innovative...
", "time": "2025-04-30T14:30:11Z"}, {"author": "Orie Steele", "text": "is reserving a tag that changes semantics of nested tags dangerous to programs that don't understand the tag?
", "time": "2025-04-30T14:33:37Z"}, {"author": "Carsten Bormann", "text": "some decoders will just give up on 0(\"foo\")
", "time": "2025-04-30T14:34:10Z"}, {"author": "Orie Steele", "text": "failing closed for compression you don't understand seems familiar
", "time": "2025-04-30T14:34:31Z"}, {"author": "Marco Tiloca", "text": "Nested tagging here: https://www.rfc-editor.org/rfc/rfc8392#section-6
", "time": "2025-04-30T14:35:23Z"}, {"author": "Orie Steele", "text": "its a lossless codec, that you can operate on
", "time": "2025-04-30T14:38:03Z"}, {"author": "Rohan Mahy", "text": "I dont think that statement is correct.
", "time": "2025-04-30T14:38:15Z"}, {"author": "Joe Hildebrand", "text": "More similar to EXI for XML.
", "time": "2025-04-30T14:38:16Z"}, {"author": "\u0412\u0430\u0434\u0438\u043c \u0413\u043e\u043d\u0447\u0430\u0440\u043e\u0432", "text": "CBAR is like EXI
", "time": "2025-04-30T14:38:49Z"}, {"author": "Rohan Mahy", "text": "you still need to \"unpack\" the packed CBOR before you can process it.
", "time": "2025-04-30T14:38:52Z"}, {"author": "\u0412\u0430\u0434\u0438\u043c \u0413\u043e\u043d\u0447\u0430\u0440\u043e\u0432", "text": "you may without using indefinite-length strings api
", "time": "2025-04-30T14:39:24Z"}, {"author": "Orie Steele", "text": "i disagree regarding the bytestring part of this... the same cbor data when rearanged... has different bytes... if it didn't... it wouldn't be useful...
", "time": "2025-04-30T14:39:26Z"}, {"author": "Orie Steele", "text": "consider for example a transformation to json that removed base64 inflation, by moving in and out of an intermedia format... you can't do that with JSON, but you can do it with tags.
", "time": "2025-04-30T14:40:39Z"}, {"author": "Rohan Mahy", "text": "\n\ni disagree regarding the bytestring part of this... the same cbor data when rearanged... has different bytes... if it didn't... it wouldn't be useful...
\n
+1
", "time": "2025-04-30T14:41:13Z"}, {"author": "Orie Steele", "text": "but I do agree, that there is a way to do this in the app layer, and in the encoder / decoder... and they are different things imo.
", "time": "2025-04-30T14:41:31Z"}, {"author": "Mikolai G\u00fctschow", "text": "in particular, you don't need to allocate an additional buffer to copy all the data to during processing
", "time": "2025-04-30T14:41:38Z"}, {"author": "Rohan Mahy", "text": "not sure it matters, but I think we are struggling to define the requirements or at least when we know we are \"done\"
", "time": "2025-04-30T14:41:51Z"}, {"author": "Rohan Mahy", "text": "@mikolai and @vadim that's useful
", "time": "2025-04-30T14:43:35Z"}, {"author": "Orie Steele", "text": "the primary use case requirement for this sorta thing is operating on the \"packed\" representation in intermediate processors... changing the parameters of how \"packed / small\" might make it intermediate processing easier for constrained intermediates i suppose.
", "time": "2025-04-30T14:45:20Z"}, {"author": "Carsten Bormann", "text": "... enabling operating on ... (while also having a more concise representation)
", "time": "2025-04-30T14:46:02Z"}, {"author": "Orie Steele", "text": "yep, but in in a format that is already concise, its about that sweet spot... where the packing is useful in removing redundancy... so that it improves operations.
", "time": "2025-04-30T14:47:08Z"}]