Digital object identifiers are an international standard for document identification:
https://en.wikipedia.org/wiki/Digital_object_identifier
The WMF could be a DOI registrant, and resolve DOIs in the form 10.NNNN.Qnnnnn for Wikidata items, or, say, 10.NNNN.en:609232908 for:
https://en.wikipedia.org/w/index.php?title=The_King_of_Rome&oldid=609232...
Where 's the best on-wiki (Meta?) place to propose this?
Andy Mabbett, 30/12/2014 22:53:
Where 's the best on-wiki (Meta?) place to propose this?
https://meta.wikimedia.org/wiki/RfC looks ok *if* you have enough information already. For instance, AFAIK DOI has a non-negligible cost and Internet Archive uses ARK instead for this reason. Information on the financial cost would certainly inform the discussion significantly, e.g. to decide whether it's worth it at all or whether to use curid instead, in your example https://en.wikipedia.org/?curid=31445754
Nemo
Il giorno mer, 31/12/2014 alle 09.25 +0100, Federico Leva (Nemo) ha scritto:
Andy Mabbett, 30/12/2014 22:53:
Where 's the best on-wiki (Meta?) place to propose this?
https://meta.wikimedia.org/wiki/RfC looks ok *if* you have enough information already. For instance, AFAIK DOI has a non-negligible cost and Internet Archive uses ARK instead for this reason. Information on the financial cost would certainly inform the discussion significantly, e.g. to decide whether it's worth it at all or whether to use curid instead, in your example https://en.wikipedia.org/?curid=31445754
It looks quite expensive indeed: http://www.medra.org/en/terms.htm
Laurentius
On 31 December 2014 at 10:10, Laurentius laurentius.wiki@gmail.com wrote:
The WMF could be a DOI registrant
It looks quite expensive indeed:
Yes, that would seem to be prohibitive (and unreasonably so).
Pity.
Does ARK:
https://en.wikipedia.org/wiki/Archival_Resource_Key
have any benefits for us or our users?
Aside: I've reqeusted a "Name Assigning Authority Number" property for Wikidata:
https://www.wikidata.org/wiki/Wikidata:Property_proposal/Organization#Name_A...
My experience is that to create a DOI you need to provide a basic level of metadata for each item rather than simply registering a target URL - I'm not sure how curated this needs to be, and it can probably be autogenerated, but there might be problems scaling it and doing it on demand. There is also a short delay before they become active at the central registry. (I've certainly seen cases where a publisher has issued a DOI then announced it to the world before CrossRef are able to resolve it, and it takes a day or two before the DOI works...)
As a result, I don't think we could generate these on the fly and use a URL-shortener type approach - there might be problems with generating that many of them, and they would not reliably work at the moment they're generated.
Andrew.
On 30 December 2014 at 21:53, Andy Mabbett andy@pigsonthewing.org.uk wrote:
Digital object identifiers are an international standard for document identification:
https://en.wikipedia.org/wiki/Digital_object_identifier
The WMF could be a DOI registrant, and resolve DOIs in the form 10.NNNN.Qnnnnn for Wikidata items, or, say, 10.NNNN.en:609232908 for:
https://en.wikipedia.org/w/index.php?title=The_King_of_Rome&oldid=609232...
Where 's the best on-wiki (Meta?) place to propose this?
-- Andy Mabbett @pigsonthewing http://pigsonthewing.org.uk
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
(sort of) related to this old thread... the DOI resolver site went down today because they apparently forgot to renew the domain, and the author of this blog post from CrossRef (who runs it) suggests relying on *us* for persistent identifier stability: http://crosstech.crossref.org/2015/01/problems-with-dx-doi-org-on-january-20...
He notes "the “persistence” [of persistent identifiers] is the result of a social contract" -- indeed.
best, Phoebe
On Wed, Dec 31, 2014 at 9:20 AM, Andrew Gray andrew.gray@dunelm.org.uk wrote:
My experience is that to create a DOI you need to provide a basic level of metadata for each item rather than simply registering a target URL - I'm not sure how curated this needs to be, and it can probably be autogenerated, but there might be problems scaling it and doing it on demand. There is also a short delay before they become active at the central registry. (I've certainly seen cases where a publisher has issued a DOI then announced it to the world before CrossRef are able to resolve it, and it takes a day or two before the DOI works...)
As a result, I don't think we could generate these on the fly and use a URL-shortener type approach - there might be problems with generating that many of them, and they would not reliably work at the moment they're generated.
Andrew.
On 30 December 2014 at 21:53, Andy Mabbett andy@pigsonthewing.org.uk wrote:
Digital object identifiers are an international standard for document identification:
https://en.wikipedia.org/wiki/Digital_object_identifier
The WMF could be a DOI registrant, and resolve DOIs in the form 10.NNNN.Qnnnnn for Wikidata items, or, say, 10.NNNN.en:609232908 for:
https://en.wikipedia.org/w/index.php?title=The_King_of_Rome&oldid=609232...
Where 's the best on-wiki (Meta?) place to propose this?
-- Andy Mabbett @pigsonthewing http://pigsonthewing.org.uk
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
--
- Andrew Gray andrew.gray@dunelm.org.uk
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
phoebe ayers, 20/01/2015 23:42:
suggests relying on*us* for persistent identifier stability:
Hmm I'm not sure that's what it's written there.
However, relatedly, also today: http://www.newyorker.com/magazine/2015/01/26/cobweb «The footnote, a landmark in the history of civilization, took centuries to invent and to spread. It has taken mere years nearly to destroy. [...] The footnote problem, though, stands a good chance of being fixed. Last year, a tool called Perma.cc was launched.» I looked into perma.cc some time ago but I had never read such an emphatic supporter yet. (Their stats also seem rather flat lately.)
The two articles combined make me wonder: if I cite a Wikimedia projects page in a long-term document, should I link something like perma.cc or to the oldid? I prefer the oldid, because I think it's every website's responsibility to offer really permanent links. But if such a permalink/archival service was offered by a national library with the guarantees of legal deposit... then I wouldn't be sure.
Nemo
On 21 January 2015 at 00:01, Federico Leva (Nemo) nemowiki@gmail.com wrote:
The two articles combined make me wonder: if I cite a Wikimedia projects page in a long-term document, should I link something like perma.cc or to the oldid? I prefer the oldid, because I think it's every website's responsibility to offer really permanent links.
Mediaviewer does not in fact offer a permanent link. See:
https://www.mediawiki.org/wiki/Extension_talk:Media_Viewer/About#Non_predomi...
2015-01-21 13:40 GMT+02:00 geni geniice@gmail.com:
On 21 January 2015 at 00:01, Federico Leva (Nemo) nemowiki@gmail.com wrote:
The two articles combined make me wonder: if I cite a Wikimedia projects page in a long-term document, should I link something like perma.cc or to the oldid? I prefer the oldid, because I think it's every website's responsibility to offer really permanent links.
Mediaviewer does not in fact offer a permanent link.
Mediawiever makes it harder to reach the permanent link (at least 1 extra click), it does not get rid of it. The problem you describe also occurs for articles moved without redirect, that's why the oldid exist.
I too believe that it's every site's responsibility to offer permalinks and 3rd party services are risky, because out of both the site's and reuser's control.
Strainu
The oldid. At the moment, I trust our long-term viability more than a 2014 web-archiving startup, even one with praiseworthy names attached ;-)
(Foolish question: can oldids be reconstituted from dumps?)
Andrew.
On 21 January 2015 at 00:01, Federico Leva (Nemo) nemowiki@gmail.com wrote:
phoebe ayers, 20/01/2015 23:42:
suggests relying on*us* for persistent identifier stability:
Hmm I'm not sure that's what it's written there.
However, relatedly, also today: http://www.newyorker.com/magazine/2015/01/26/cobweb «The footnote, a landmark in the history of civilization, took centuries to invent and to spread. It has taken mere years nearly to destroy. [...] The footnote problem, though, stands a good chance of being fixed. Last year, a tool called Perma.cc was launched.» I looked into perma.cc some time ago but I had never read such an emphatic supporter yet. (Their stats also seem rather flat lately.)
The two articles combined make me wonder: if I cite a Wikimedia projects page in a long-term document, should I link something like perma.cc or to the oldid? I prefer the oldid, because I think it's every website's responsibility to offer really permanent links. But if such a permalink/archival service was offered by a national library with the guarantees of legal deposit... then I wouldn't be sure.
Nemo
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Andrew Gray, 21/01/2015 13:35:
(Foolish question: can oldids be reconstituted from dumps?)
Yes, grep any XML dump for "<revision>" to see yourself. However, the standard import doesn't preserve this information.
Nemo
On Wed, Jan 21, 2015 at 1:35 PM, Andrew Gray andrew.gray@dunelm.org.uk wrote:
The oldid. At the moment, I trust our long-term viability more than a 2014 web-archiving startup, even one with praiseworthy names attached ;-)
Possibly an obvious remark: an oldid-based permalink gets you an old version of the source text of the specific page, but does not give you the original visible _result_: The page is reparsed and displayed using the _current_ version of all used templates, images, data from Wikidata etc. We do not store the complete state anywhere, so perma.cc does provide extra value here.
-- [[cs:User:Mormegil | Petr Kadlec]]
That said, there was an attempt in 2013 to provide Memento support, which would (partially) mitigate this problem by using appropriate templates:
http://mementoweb.org/wikipedia/
Of course, this would still be vulnerable to deleted images etc. And we weren't even thinking about Wikidata transclusions then :-)
Andrew.
On 23 January 2015 at 16:09, Petr Kadlec petr.kadlec@gmail.com wrote:
On Wed, Jan 21, 2015 at 1:35 PM, Andrew Gray andrew.gray@dunelm.org.uk wrote:
The oldid. At the moment, I trust our long-term viability more than a 2014 web-archiving startup, even one with praiseworthy names attached ;-)
Possibly an obvious remark: an oldid-based permalink gets you an old version of the source text of the specific page, but does not give you the original visible _result_: The page is reparsed and displayed using the _current_ version of all used templates, images, data from Wikidata etc. We do not store the complete state anywhere, so perma.cc does provide extra value here.
-- [[cs:User:Mormegil | Petr Kadlec]] _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
wikimedia-l@lists.wikimedia.org