Re: Proposal: WoT Profiles 1.0 Registry

On 20/08/25 16:15, Ben Francis wrote:
> Hi,
> 
> Further to the Call for Resolution <https://lists.w3.org/Archives/ 
> Public/public-wot-wg/2025Jul/0024.html> to publish WoT Profiles 1.0 as a 
> Group Note as per the plan agreed in a previous Working Group resolution 
> <https://www.w3.org/2025/05/21-wot-minutes.html#335a>, it appears that 
> there is now some uncertainty about whether the latest W3C Process 
> <https://www.w3.org/policies/process/> will allow us to switch from the 
> Recommendation Track to the Note Track in order to do this.
> 
> We may therefore have a situation where W3C Process may prevent us from 
> publishing as a Group Note, but there may also not be sufficient 
> consensus within the Working Group to proceed along the Recommendation 
> track with the document in its current form. This could force us to 
> publish the Profiles 1.0 specification as a Discontinued Draft, which 
> the group has agreed would not be a very satisfactory outcome since it 
> does not accurately represent the current state of the document.
> 
> I hope to continue to discuss this with W3C Staff to clarify exactly 
> what the issue is with switching tracks, and whether it could be 
> feasible for us to continue with a Group Note as per the consensus of 
> the Working Group.
> 
> In the meantime, I wanted to present a potential "wildcard" option to 
> resolve the deadlock.
> 
> *Profiles 1.0 Registry*
> 
> The idea is to split the individual profiles out of the current WoT 
> Profiles specification as separate documents, and reduce the Profiles 
> specification itself to defining only the profiling /mechanism/ and a 
> registry of profiles (in the form of a table). Each individual profile 
> would become a new Group Note. This would result in four documents:
> 
>   * WoT Profiles 1.0 (Candidate Recommendation)
>       o Profiling Mechanism
>       o Profile Registry
>   * WoT HTTP Basic Profile (Group Note)
>   * WoT HTTP SSE Profile (Group Note)
>   * WoT HTTP Webhook Profile (Group Note)
> 
> To help illustrate this I have mocked up what this cut-down version of 
> WoT Profiles 1.0 could look like, and temporarily hosted it here 
> <https://benfrancis.github.io/wot-profile/>.
> 
> What is neat about this is that it is an approach we have discussed 
> taking for Profiles 2.0 as well.
> 
> An advantage of this approach is that we already seem to have broad 
> consensus on the general need for a profiling mechanism, and multiple 
> existing implementations of the mechanism as currently defined. 
> Publishing only the mechanism as a Candidate Recommendation means we 
> don't necessarily need multiple implementations of every individual 
> profile in order to continue along the Recommendation Track, with the 
> individual profiles being published as Group Notes.
> 
> A disadvantage is that the Profiles task force would likely have to 
> spend the next two to three months working on these modified documents, 
> rather than kicking off work on Use Cases & Requirements for Profiles 2.0.
> 
> ---
> 
> My strong preference is still to publish WoT Profiles 1.0 as a Group 
> Note in its current form, as per the Working Group consensus. But if 
> that turns out not to be feasible then what do people think about this 
> alternative approach? Would you be happy for this cut-down version of 
> WoT Profiles 1.0 to proceed to Candidate Recommendation?

I'd rather devote the available time on Profiles 2.0, but as fallback 
sounds good and probably we could reuse the modified 
1.0-as-start-for-registry for 2.0 nonetheless

lu

Received on Wednesday, 20 August 2025 16:37:42 UTC