Re: New Work Item Proposal: Universal Wallet 2020

A brief reply regarding "reference implementation".  W3C (and IETF, for
that matter) has generally not encouraged the creation of a reference
implementation because it encourages implementers to copy the code's
behavior, even if buggy, rather than implementing to the spec.  As you
quoted from Wikipedia, it is the Gold Standard.  W3C generally believes
that the Gold Standard should be the standard (specification) itself.

-- dan

On Wed, Jun 24, 2020 at 8:07 PM Orie Steele <orie@transmute.industries>
wrote:

> Everytime this comes up, I wonder why members of this community seem
> opposed to the wikipedia definition:
> https://en.wikipedia.org/wiki/Reference_implementation
>
> Characteristics of a Reference Implementation:
>
>    1. Developed concurrently with the specification and test suite;
>    2. Verifies that specification is implementable;
>    3. Enables the test suite to be tested;
>    4. Serves as a Gold Standard against which other implementations can
>    be measured;
>    5. Helps to clarify the intent of the specification in situations
>    where conformance tests
>    <https://en.wikipedia.org/wiki/Conformance_testing> are inadequate
>
>
> Does W3C specifically discourage the use of the term "reference
> implementation".... I assumed this was a common practice:
> https://github.com/webofthings/webofthings.js
>
> Regardless of the name, these are the characteristics I'm looking for. As
> wikipedia notes, "sample" or "model" implementation are less
> frequently used terms for the same thing.
>
> There are some conformance tests, and I can see a lot of value in keeping
> them next to the spec and "sample implementation", I'm generally a fan of
> starting with a mono repo and splitting things up when it becomes obvious
> that it is the right thing to do because there is too much contribution for
> one repo to handle, or because sub components are getting consumed
> independently almost as frequently as they are getting consumed together.
>
> If there is reason to believe there will be more contribution overall if
> the "sample implementation" is also in W3C CCG repos, I'm happy to do
> that... I saw splitting them up as a potential opportunity to
> engage multiple communities and AFAIK you can contribute to Apache 2.0
> software regardless of the github organization that hosts it, for example:
> https://github.com/adobe/experience-platform-dsw-reference is also Apache
> 2.0 with a CLA...
>
> Leonard, should I interpret your strong desire that "an example
> implementation" stay in CCG as an indication of a strong desire / plan to
> contribute to it here?
>
> OS
>
>
>
> On Wed, Jun 24, 2020 at 5:23 PM Leonard Rosenthol <lrosenth@adobe.com>
> wrote:
>
>> "Reference Implementation" means that it is the "blessed/approved "
>> implementation which implements 100% of a specification to the agreement of
>> all parties involved.
>>
>> "Conforming" implies that (a) the specification has normative
>> requirements to which the software can conform and that (b) someone(s) have
>> all agreed that it meets 100% of the requirements.
>>
>> Example implementation is probably best.
>>
>> Leonard
>>
>> On 6/24/20, 4:47 PM, "Wayne Chang" <wyc@fastmail.fm> wrote:
>>
>>     Manu, could you please elaborate on your concerns about calling
>> certain code a "reference implementation"? What drawbacks are there, and
>> what else could we call it? "Example implementation"? "Conforming
>> implementation"?
>>
>>     On Wed, Jun 24, 2020, at 4:26 PM, Manu Sporny wrote:
>>     > On 6/24/20 3:31 PM, Orie Steele wrote:
>>     > > We're looking for additional organizations to co-edit / sponsor
>> the
>>     > > development of the specification in the W3C CCG.
>>     >
>>     > Digital Bazaar is supportive of this specification and hosting it
>> in the
>>     > W3C CCG.
>>     >
>>     > We have no strong opinion on hosting an implementation at DIF as
>> long as
>>     > doing so wouldn't create IPR issues that would restrict anyone from
>>     > contributing to the work. We continue to be skeptical of calling it
>> a
>>     > "reference implementation".
>>     >
>>     > While we do not have the spare cycles to be a co-editor on the
>>     > specification, we do expect to be heavily involved in discussion
>> related
>>     > to the specification as we expect that the outcome will have a
>> direct
>>     > impact on our product offerings.
>>     >
>>     > Finally, thank you to Orie, Margo, and Guillaume from Transmute for
>>     > putting this together and kicking off the work. It's a critical
>> piece
>>     > that the industry will need in order to interoperate at the
>> application
>>     > layer.
>>     >
>>     > -- manu
>>     >
>>     > --
>>     > Manu Sporny -
>> https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.linkedin.com%2Fin%2Fmanusporny%2F&amp;data=02%7C01%7Clrosenth%40adobe.com%7C54db72b34f034e36a6cb08d8187fbe4f%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C637286284216262630&amp;sdata=nGVOqadvTWe01uCzQIDa715%2BpTZeqEouewEEF9aXYoA%3D&amp;reserved=0
>>     > Founder/CEO - Digital Bazaar, Inc.
>>     > blog: Veres One Decentralized Identifier Blockchain Launches
>>     >
>> https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Ftinyurl.com%2Fveres-one-launches&amp;data=02%7C01%7Clrosenth%40adobe.com%7C54db72b34f034e36a6cb08d8187fbe4f%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C637286284216262630&amp;sdata=u%2F66e1Uy8HbOvk58Sf3i8RsUKFzdpltLbtSxsNaXnVI%3D&amp;reserved=0
>>     >
>>     >
>>
>>
>>
>
> --
> *ORIE STEELE*
> Chief Technical Officer
> www.transmute.industries
>
> <https://www.transmute.industries>
>