Re: [Cbor] Private tag numbers

On 2023-04-24 13:27, Tony Putman wrote:
> Hi Anders,
> 
> Thanks for the suggestion.  I had somehow missed your COTX draft.  It looks like a mixed approach is the way to go if the contents schema is changing often.  I don't think this is the case for what I'm investigating right now, but we're trying to use CBOR in more areas, so it's good to keep this solution in mind.
Hi Tony,

If URLs are not for you, wouldn't the other solution I sketched up fill the bill?  With a specification, you can get a 3-byte master tag (like COTX's 1010). Then you get an unlimited number of private tags at a low cost (in bytes).

COTX on the other hand, builds on the same trust model as the millions of projects of all kinds depending on public repositories like GitHub and mavenCentral.  De-referencing URL "tags" is pretty handy: https://fido-web-pay.github.io/ns/p1

Best regards,
Anders

> 
> Best regards,
> Tony
> 
> -----Original Message-----
> From: Anders Rundgren <anders.rundgren.net@gmail.com>
> Sent: 22 April 2023 10:07
> To: Tony Putman <Anthony.Putman@dyson.com>; cbor@ietf.org
> Subject: [EXTERNAL EMAIL] Re: [Cbor] Private tag numbers
> 
> CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe.
> 
> On 2023-04-21 18:53, Tony Putman wrote:
>> Hi all,
>>
>> I'm defining a data schema for purely internal use (no interoperability needed) and encoding it in CBOR. There are points where we may want to vary/extend the contents with new data, and I thought the natural way to do this would be to define different tags to hold the data. I expected to find a set of tag numbers defined in IANA for private use, but there don't appear to be any.
>>
>> - Is this a misuse of tags?  And if so, what would be the more natural way to specify this?
>>
>> - Is this a general problem that people face and would it be worth defining a set of private tags which are known to be context-sensitive and not interoperable?
>>
> Hi Tony,
> 
> Personally I find the tag system less useful if you have a large bunch of ever changing objects. Obviously reusing tag numbers for revised objects isn't a good idea.
> 
> One way to address this issue would be to register a *single* master tag, and then use locally defined sub-tags to specify the actual object id:
> 
>     78564([2, {
>       "hi": "there"
>     }])
> 
> Where 78564 would be your master (registered) tag, and 2 the locally defined id. The overhead in the sample is just two bytes compared to specifying a unique tag for each object.
> 
> 
> If saving bytes is not your main concern, you may find a concept borrowed from the XML world useful:
> https://www.ietf.org/archive/id/draft-rundgren-cotx-04.html
> With public repositories like GitHub, anybody can enjoy a robust and project specific namespace.
> 
> Example: https://fido-web-pay.github.io/specification/crypto.html#5
> 
> Anders
> 
>> Thoughts/comments on this will be gratefully received.
>>
>> Tony
>>
>>
>> _______________________________________________
>> CBOR mailing list
>> CBOR@ietf.org
>> https://www.ietf.org/mailman/listinfo/cbor
> 