Re: [Tools-discuss] Expiry Doctrine (Re: Expired draft on the w.g. status pages [was Re: disappearing IDs on www.ietf.org])

[Joseph Touch <[e-mail]>]

> On Sep 23, 2020, at 9:52 AM, Carsten Bormann <[e-mail]> wrote:
> 
>> Drafts are drafts unless they’re not. 
> 
> We have learned for a few decades that there are multiple perspectives on that, none of which can be ignored.  A patent lawyer has a different perspective than a software archeologist (a.k.a. implementer) than a standards developer.

Yes, but patent lawyers don’t necessarily need stable links; they need access to an archive, which they already have.

>> Drafts expire unless they don’t. 
> 
> Again, there are different perspectives.
> Saying your’s is the only one we should care about because that is how we SHOULD work, while we DO work differently, is the classic case of process confabulation.
> 
> Having labels 
> 
>> Treating them as artifacts in need of a stable reference means they’re not drafts anymore. They’re just versions. 
> 
> They are versions of drafts!
> 
>> So assign RFC numbers when they’re initially submitted and allow rev numbers or don’t. 
> 
> Nonsense.  The label “RFC” is meaningful and should only be used for documents that are no longer “drafts”.

By making links to drafts stable, the label RFC becomes less meaningful as a distinction. As you and Brian claim, then becomes “just a label”.

> 
>> But if they’re not drafts, again we should just publish them to arXiv and skip all the irrelevant IETF mechanism. 
> 
> Nonsense.  The IETF process is useful in producing labels such as “RFC”, “Standards Track”, “obsoletes”, “Updates”, etc.
> 
> What’s the point of drawing these strawmen?

I am actually trying to figure out whether to post updates to pending drafts via a process that is increasingly diverging into its own, idiosyncratic full publication process or stop wasting my time and just go straight to arXiv.

You and Brian make a very strong case for the latter.

Joe