Further exploration reveals that  http://doai.io ,
http://doi2oa.erambler.co.uk/ and https://oadoi.org/ (now
http://unpaywall.org/ ) don't resolve handles in the way that
https://doi.org/ does.

Having said that, at least some are open source (
https://github.com/jezcope/doi2oa) so it shouldn't be too hard to add.

cheers
stuart
--
...let us be heard from red core to black sky


On Fri, 17 Aug 2018 at 14:20, Stuart A. Yeates <[log in to unmask]> wrote:

> Interesting insight Conal, I wasn't aware of that service.
>
> https://doi.org/10063/1710 redirects to
> http://researcharchive.vuw.ac.nz/handle/10063/1710 using a 302 redirect,
> implying that the server knows where the DOI resides by RFC 7231
> https://tools.ietf.org/html/rfc7231
>
> If  10063/1710 were not a valid DOI, the DOI server should use 303 (if it
> redirects) and  a 400 or 404 if it doesn't.
>
> cheers
> stuart
> --
> ...let us be heard from red core to black sky
>
>
> On Fri, 17 Aug 2018 at 13:27, Conal Tuohy <[log in to unmask]> wrote:
>
>> Kia ora Stuart!
>>
>> I think the answer to your question is "no, the identifier is not a valid
>> DOI".
>>
>> As evidence, I offer this URI which is supposed return information about
>> the Registration Agency which registered that DOI:
>> https://doi.org/doiRA/10063/1710
>>
>> As you know, DOIs are a proper subset of Handles; and functionally, the
>> DOI
>> system relies on the Handle system as its infrastructure for URI
>> resolution. I believe that when you resolve the URI <
>> https://doi.org/10063/1710>, the DOI resolver is simply resolving the
>> identifier as a Handle, and not first validating that the Handle is
>> actually a valid DOI. I'd regard that as a bug in the DOI's resolver,
>> personally.
>>
>> Cheers!
>>
>> Conal
>>
>>
>> On Fri, 17 Aug 2018 at 09:37, Stuart A. Yeates <[log in to unmask]> wrote:
>>
>> > We have a DSpace instance that is configured to issue handle.net
>> > identifiers to all items, so links such as:
>> >
>> > https://researcharchive.vuw.ac.nz/handle/10063/1710
>> > http://researcharchive.vuw.ac.nz/handle/10063/1710
>> > https://hdl.handle.net/10063/1710
>> > http://hdl.handle.net/10063/1710
>> >
>> > all take a web browser to the same content. The following URLs also take
>> > web
>> > browsers to the same content:
>> >
>> > https://doi.org/10063/1710
>> > http://doi.org/10063/1710
>> > https://dx.doi.org/10063/1710
>> > http://dx.doi.org/10063/1710
>> >
>> > The lookup at https://www.doi.org/index.html resolves the doi
>> "10063/1710"
>> > to the same content.
>> >
>> > I have two questions:
>> >
>> > (a) is 10063/1710 a valid/legal doi for this item ?
>> > (b) are the doi.org URLs above valid/legal for this item?
>> >
>> > The documentation on the https://www.doi.org/ and https://handle.net/
>> > websites is surprisingly quiet on these issues...
>> >
>> > [We've been assuming the answer to these questions is 'yes' but
>> yesterday
>> > this was questioned by a colleague, so I'm looking for definitive
>> answers]
>> >
>> > cheers
>> > stuart
>> > --
>> > ...let us be heard from red core to black sky
>> >
>>
>>
>> --
>> Conal Tuohy
>> http://conaltuohy.com/
>> @conal_tuohy
>> +61-466-324297
>>
>