Re: dwbp-ISSUE-46 (PIDs): How should we handle the issue of persistent URI design? [Use Cases & Requirements Document]

Makx, Christophe,

How about:

R-PersistentIdentification

An identifier for a particular resource should be resolvable on the Web 
and associated for the foreseeable future with a single resource or with 
information about why the resource is no longer available.

This combines both of your suggestions (and makes the Web aspect 
explicit). It doesn't say anything here, in the use case doc, about how 
to do that - we can do that in the BP doc where DanBri's sage advice can 
be mixed with others'.

WDYT?

Phil


On 01/10/2014 11:45, Makx Dekkers wrote:
> Phil,
>
>>>
>>> The identifier that is assigned to a particular resource should
>> resolve, at least for the foreseeable future, to that same resource or
>> to information why the resource is no longer there.
>>
>> I like the simple approach for the reasons you give.
>>
>> What worries me is that we risk getting into a long debate about
>> whether
>> a DOI resolves or not (IMO of course it doesn't - it only resolves if
>> you stick it on the end of a URL in which case it is a different
>> identifier) and what the semantics of a DOI may be (or ORCID or any of
>> the other similar schemes).
>>
>
> I thought of a different formulation:
>
>
> The identifier that is assigned to a particular resource should be associated, at least for the foreseeable future, with that same resource or with information why the resource is no longer there.
>
> Would that be less controversial?
>
>>
>>>
>>> The actual syntax of URIs can vary widely, because an organisation
>> will choose a design that reflects the way they can make and maintain
>> the commitment to persistence. Some organisations will opt for a
>> semantically rich syntax (such as the UK Gov approach); others will
>> follow an (almost) semantics-free design like Tomas' COMURI proposal.
>>>
>>> As organisations might have good arguments for selecting a
>> particular design, this group may not be able to declare one approach
>> 'best practice' beyond saying that (as someone once said) "persistence
>> is non-negotiable".
>>
>> Again, I like that. I'd be in favour of encouraging a more structured
>> approach, as in the UK examples, but agree we need to recognise,
>> always,
>> that people can only build on top of what they have, technically and
>> institutionally.
>>
>
> We could have a long discussion about good reasons to strive for as little semantics as possible in URIs (in line with DanBri's 'first rule of URI design': "you're more likely to regret things you included, than things you omitted" https://twitter.com/danbri/status/508981586738814976), but I'd say that is out of scope for the group.
>
> Makx.
>
>
>
>
>

-- 


Phil Archer
W3C Data Activity Lead
http://www.w3.org/2013/data/

http://philarcher.org
+44 (0)7887 767755
@philarcher1

Received on Wednesday, 1 October 2014 11:22:05 UTC