Seems that there is some uncertainty about this "special" serialisation.
I would actually vote for replacing the flatened JSON serialisation with
one, that provides a real benefit. Taken up on a discussion I read
earlier on the list, wouldn't it be more sensible to have a "readable"
JSON serialisation (i.e., leaving the signed payload "human readbale")!?
This would of course require some form of normalisation/canonicalisaton
as used e.g. in XML Security. Still, this would be something valuable to
have and a real distinguishing point in comparison to the other
serialisations.

If people think that this is worth a discussion, then maybe we should
kick-off an explicit thread on it.

Another point I would like to raise is to have an additional mandatory
header entry specifying the JWS/JWE version used. As the standards will
evolve over time, it would be good to know, with what version of the
JWS/JWE standards a given object have been protected.

BR, Luigi.


Am 12.03.15 um 18:20 schrieb Justin Richer:
> My bad, I totally missed that.
> 
>  -- Justin
> 
> On 3/12/2015 11:38 AM, John Bradley wrote:
>> Yes flattened is a special case of the JSON serialization that has
>> only one signature.
>>
>> I don't know that there is a huge benefit to it over the regular JSON
>> serialization that supports multiple signatures.
>>
>> Some people really wanted it at the time.  
>>
>> Supporting the general JSON serialization would be a higher priority
>> for me.  In most cases I have seen, as Justin points out people
>> wanting a single signature go with compact.
>>
>> John B.
>>
>>
>>> On Mar 12, 2015, at 11:49 AM, Brian Campbell
>>> <[email protected] <mailto:[email protected]>> wrote:
>>>
>>> flattened is not the same as compact FWIW
>>>
>>> compact ->
>>> https://tools.ietf.org/html/draft-ietf-jose-json-web-signature-41#section-7.1
>>>
>>> flattened ->
>>> https://tools.ietf.org/html/draft-ietf-jose-json-web-signature-41#section-7.2.2
>>>
>>> On Thu, Mar 12, 2015 at 8:08 AM, Justin Richer <[email protected]
>>> <mailto:[email protected]>> wrote:
>>>
>>>     Just to add some perspective, if by “flattened” serialization,
>>>     you mean the compact serializations of JWS
>>>     (header.payload.signature) and JWE
>>>     (header.stuff.stuff.stuff.stuff I forgot the order), then there
>>>     are huge advantages to these, and they’re the only ones that I
>>>     personally use.
>>>
>>>     The simplicity gained in processing the compact forms, both in
>>>     terms of generating and consuming. With the compact forms, you
>>>     get something that can be dropped on the wire into an HTTP
>>>     header, form parameter, query parameter, a string in just about
>>>     any language, all without any quoting or further processing.
>>>     Plus, to get back to the crypto calculations, you use the literal
>>>     strings sent across the wire, which is a really nice feature.
>>>
>>>     I’ve personally yet to have a use case that required the multiple
>>>     signatures or other features of the JSON serialized flavor, nor
>>>     have I seen much uptake of it in the wild compared to the compact
>>>     forms.
>>>
>>>      — Justin
>>>
>>>     > On Mar 12, 2015, at 9:18 AM, Prof. Dr.-Ing. Luigi Lo Iacono
>>>     <[email protected]
>>>     <mailto:[email protected]>> wrote:
>>>     >
>>>     > Dear Vladimir,
>>>     >
>>>     > yes, we do support the flattened serialisation, but you are
>>>     right, we
>>>     > did not mention it in the "feature list". This is mainly
>>>     because we do
>>>     > not see any benefit in having this serialisation. We followed the
>>>     > discussions on it and had own discussions internally. I do not
>>>     see the
>>>     > point to define a distinct serialisation which is very close to an
>>>     > already existing one. This increases complexity only and that's
>>>     it. The
>>>     > marginal amount of data reduction coming with the flattened
>>>     > serialisation is ridiculous in comparison to the JSON
>>>     serialisation with
>>>     > only one signature. I personally like the second approach more,
>>>     since it
>>>     > still give you the flexibility to add further signatures along
>>>     the way.
>>>     > Basically, this is the way our architecture supports multiple
>>>     > signatures. They are added by the signing parties one after the
>>>     other.
>>>     > From our perspective, this is the only realistic use case here.
>>>     Having a
>>>     > signing process in possession of multiple distinct private key
>>>     breaks
>>>     > with a lot of security principles. At least in my understanding
>>>     or do I
>>>     > miss something here!?
>>>     >
>>>     > In our architectural approach a JwsDocument is constructed by a
>>>     JwsMaker
>>>     > (either from scratch or by parsing an existing one):
>>>     >
>>>     > JwsDocument jws = JwsMaker.generate...
>>>     >
>>>     > Having such an object, getting a particular serialisation is just a
>>>     > matter of calling the respective method:
>>>     >
>>>     > - Compact: jws.getCompactSerialisation();
>>>     > - Compact DETACHED: jws.getCompactDetachedSerialisation();
>>>     > - JSON: jws.getJsonSerialisation();
>>>     > - JSON PRETTYPRINTED: jws.getJsonSerialisation(true);
>>>     > - JSON DETACHED: jws.getJsonDetachedSerialisation();
>>>     > - JSON PRETTYPRINTED & DETACHED:
>>>     jws.getJsonDetachedSerialisation(true);
>>>     > - JSON Flattened: jws.getJsonFlattenedSerialisation();
>>>     > - JSON Flattened PRITTYPRINTED:
>>>     jws.getJsonFlattenedSerialisation(true);
>>>     > - JSON Flattened DETACHED:
>>>     jws.getJsonFlattenedDetachedSerialisation();
>>>     > - JSON Flattened PRITTYPRINTED & DETACHED:
>>>     >  jws.getJsonFlattenedDetachedSerialisation(true)
>>>     >
>>>     > Hope that helps!?
>>>     >
>>>     > Do not hesitate to ask further questions. We are happy to help and
>>>     > further feedback is welcome!
>>>     >
>>>     > BR, Luigi.
>>>     >
>>>     >
>>>     > Am 11.03.15 um 17:38 schrieb Vladimir Dzhuvinov:
>>>     >> Thanks for sharing this.
>>>     >>
>>>     >> I see that you support JSON and compact serialisation, but what is
>>>     >> flattened serialisation?
>>>     >>
>>>     >> Thanks,
>>>     >>
>>>     >> Vladimir
>>>     >>
>>>     >> On 5.03.2015 14:04, Prof. Dr.-Ing. Luigi Lo Iacono wrote:
>>>     >>> Dear all,
>>>     >>>
>>>     >>> we developed an own JOSE implementation in Java, mainly
>>>     because we
>>>     >>> missed the JSON serialisation in almost all of the available
>>>     libs. You
>>>     >>> can grasp it here:
>>>     >>>
>>>     >>> http://jw-asterisk.realsoasecurity.de/
>>>     >>>
>>>     >>> We are still doing some polishing, that is why the sources
>>>     are still
>>>     >>> lacking. Stay tuned, though, updates will follow soon...
>>>     >>>
>>>     >>> The documentation and especially the unit tests should help
>>>     in taking
>>>     >>> the first steps.
>>>     >>>
>>>     >>> Let us know what you think about it...
>>>     >>>
>>>     >>> BR, Luigi.
>>>     >>>
>>>     >>>
>>>     >>>
>>>     >>> _______________________________________________
>>>     >>> jose mailing list
>>>     >>> [email protected] <mailto:[email protected]>
>>>     >>> https://www.ietf.org/mailman/listinfo/jose
>>>     >>
>>>     >> --
>>>     >> Vladimir Dzhuvinov :: [email protected]
>>>     <mailto:[email protected]>
>>>     >>
>>>     >>
>>>     >>
>>>     >> _______________________________________________
>>>     >> jose mailing list
>>>     >> [email protected] <mailto:[email protected]>
>>>     >> https://www.ietf.org/mailman/listinfo/jose
>>>     >>
>>>     >
>>>     > _______________________________________________
>>>     > jose mailing list
>>>     > [email protected] <mailto:[email protected]>
>>>     > https://www.ietf.org/mailman/listinfo/jose
>>>
>>>
>>>     _______________________________________________
>>>     jose mailing list
>>>     [email protected] <mailto:[email protected]>
>>>     https://www.ietf.org/mailman/listinfo/jose
>>>
>>>
>>> _______________________________________________
>>> jose mailing list
>>> [email protected] <mailto:[email protected]>
>>> https://www.ietf.org/mailman/listinfo/jose
>>
> 

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
jose mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/jose

Reply via email to