Talk:Digital object identifier/Archive 1

Archive 1

Difference from URNs

It would be great if the article could discuss the difference between DOIs and URNs, since at first glance they seem to do much the same thing. —Psychonaut 01:59, 1 January 2006 (UTC)

implement in wikipedia

Recently I've raised the possibility (on the Wikipedia-L list) of implementing DOI into Wikipedia in the same way as ISBNs are: the wiki will simply fashion a link out of anything that follows a DOI code. Problems are the parsing (how long is the longest DOI string and what are the stopping characters). Another solution would be an interwiki identifier ([doi:etc.etc./94809324]). Any ideas on the matter? Who does one approach to have this implemented? JFW | T@lk 13:46, 15 Apr 2004 (UTC)

I created a template doi (see Template talk:doi) that simulates this:

{{doi|10.1000/1}} produces: doi:10.1000/1.

--Lexor|Talk 10:09, 10 Sep 2004 (UTC)

The main example 10.1002/ISBNJ0-471-58064-3 (doi:10.1002/ISBNJ0-471-58064-3) seems to be invalid. Is this a made up example, or should this be looked into? --Alex 02:03, 2004 Oct 3 (UTC)

The ISBN example is not only invalid, but the explanation is also wrong. The "-3" does not refer to a specific part of the book, but is part of the ISBN itself (in fact, it is its checksum). --Qlmatrix 14:17, 2004 Oct 22 (UTC)

Would it be possible to make something like the PMID WP:PMID that seems to be well integrated in wikipedia? KristianMolhave 20:08, 10 September 2006 (UTC)

DOIs are not specifically a legal concept, so term 'Intellectual property' may not be the best

The article does not document that DOIs were originally invented to specify ownership of written work, so I am hoping that the more neutral wording of what they identify might be restored.

Since I chanced to hit upon this article, I read the opening sentence and found that, wow, intellectual property is key to the definition of DOIs. Wanting to know if this was a well-settled conclusion of the editors working here, I found that from January 2004 until April 2006 this article had basically this opening sentence:

A digital object identifier (or DOI) is a permanent identifier (permalink) given to a World Wide Web file or other Internet document so that if its Internet address changes, users will be redirected to its new address.

Then, after an edit on 3 April 2006 by User:Cogitabondo who never again appeared on Wikipedia, suddenly this was the opening sentence:

A digital object identifier (or DOI) is a standard for persistently identifying a piece of intellectual property on a digital network and associating it with related current data, the metadata, in a structured extensible way.

This mixes together an information retrieval term (DOI) with a controversial legal concept. If a DOI pointed to a Wikipedia article, would that make the WP article intellectual property? The catch phrase about the DOI as a "bar code for intellectual property" is repeated in the article, but with no speaker identified, and none was easily findable with Google. The DOI Handbook more neutrally states "A DOI® (Digital Object Identifier) name is an identifier (not a location) for an entity on digital networks." Please reply if you have opinions on whether I should restore the original wording, or something similar. EdJohnston 00:59, 9 December 2006 (UTC)

In my opinion the term "intellectual property" should be replaced, either with proper wording from an earlier version or something else that accurately describes what a DOI is. 18 May 2007 —The preceding unsigned comment was added by 66.46.103.18 (talk • contribs).

I've made a stab at this. It is a somewhat common practice in some circles (semiconductor manufacture springs to mind) to say "intellectual property" where most of the word would say "document", "file", "resource", "object", etc. But I agree that it tends to drag in (mostly) irrelevant legal meanings, and isn't usually the most natural wording. Kingdon 14:06, 16 August 2007 (UTC)

Thanks for your change to the article. It seems to cure the problem I noted above. EdJohnston (talk) 18:49, 4 January 2008 (UTC)

Sorry to put a wrench in this discussion, but on the DOI.org web site on their FAQs page (http://www.doi.org/faq.html#1), a DOI® Name is specifically defined as "a Digital identifier for any object of intellectual property...A DOI name can apply to any form of intellectual property expressed in any digital environment."

The way this reads then, the "I" in DOI doesn't stand for Identifier (as I thought it did) but for Intellectual property. However, it's a little confusing at first because on their Home page (http://www.doi.org/index.html) the first sentence is "The DOI System is for identifying content objects in the digital environment" - but notice that this is not actually a definition. The first statement (the one with Intellectual property) is their stated definition.

Ironically, since DOI® is a Registered Trademark of the International DOI Foundation, then they really do have the final say on this definition since it is their own intellectual property. Take a look at the IDF Staff page for more info (http://www.doi.org/foundation/bios.html). There's only one person listed (does that mean they have a staff of one?), but there's contact information and even a telephone number if you feel like calling the UK for clarification :).

I don't know if DOI constitutes a "legal" term or not (i.e. recognized by the legal community as something that is enforceable just like a copyright or trademark would be). But it definitely has to do with intellectual property. At any rate, that's the official definition, even if it may not always be the "practical" definition.

Betsy R. (talk) 00:47, 31 January 2008 (UTC)

We should be grateful they don't use that terminology throughout. Their web site incorporates part of an ANSI standards document, http://www.doi.org/handbook_2000/appendix_1.html, which makes clear that DOI stands for 'digital object identifier,' and only uses the term 'property' once. The catch phrase that was used in the WP article for a while as part of the DOI definition, "persistently identifying a piece of intellectual property", is fortunately no longer found anywhere on the web by a Google search. EdJohnston (talk) 02:03, 31 January 2008 (UTC)

I suggest that Betsy's update of the article lead is overkill. Wikipedia does not usually go out of its way to announce who owns a trademark, and the naked link to the IDF's web site looks like bad style. I propose that we restore the former version of the lead, or something close to it. The reference to 'intellectual property' can probably be worked in later, since the IDF tries to persuade publishers that the system is useful to them, and publishers usually want to be paid for their work. Since the DOI identifier is regulated as an international standard, the IDF is not the only authority on the meaning of the term. EdJohnston (talk) 06:38, 31 January 2008 (UTC)

Privacy protection?

Recently 213.188.227.119 (talk · contribs) asked a question under Privacy Protection about why DOI providers need to collect IP addresses and domain names for people looking up articles. While this may be a valid point, I think it needs to be neutrally phrased (per WP:NPOV). If we can quote someone as asking that question (and cite a reliable source for them asking it), it is clearly OK. I'm not sure if we can ask it directly. I suggest this section needs to be rephrased. EdJohnston 19:48, 2 April 2007 (UTC)

Besides that, I'm not sure that that's really a valid question. The DOI foundation states only that they do collect domain names or IP addresses, not that they actually need to. Maybe it should be rephrased as a note about privacy concerns, instead. 208.101.144.199 03:30, 9 April 2007 (UTC)

I had email contact with CrossRef and DOI. They collect IP data to determine whether the system is abused. They store the IPs indefinitely which is somewhat strange (you may ask them using the published email addresses to verify that IPs are stored indefinitely). I believe that a systematic collection of IP addresses needs to be questioned. The amount of data collected is available here: http://www.doi.org/privacy.html "Our logs collect and store only domain names or IP addresses, dates and times of visits, and the pages visited" So the data is not anonymized. Next: "Data from the logs may be used to measure the number of visitors to the site". In my point of view, for single usage stats, they collect too much data.

Just a thought. 213.188.227.119 21:11, 5 May 2007 (UTC)

Unless a published source has questioned their data collection practices, I don't see why we would get involved in it. That would be WP:OR. We would not normally publish the results of an email inquiry; we need the info to be in published form. EdJohnston 16:24, 6 May 2007 (UTC)

Seems a little strange to get particular about publised sources on this point, when so far the article cites no independent sources at all. (Not saying that shouldn't use published sources, but should start by demonstrating notability and providing outside sources for what is here.) Zodon (talk) 17:53, 2 May 2008 (UTC)

'Disadvantages' section is POV

I believe that the Disadvantages section reflects the personal opinion of the editor who created it. Any reflection on the lack of security or privacy about the DOI system ought to come from a reliable source. You can't just cite the DOI system's own policies and assert "privacy issues... still unclear." In whose opinion? DOI themselves did not say they were unclear. Since there are no sources, I suggest that the section should be removed. Since that would then give an 'Advantages' section but no 'Disadvantages', I suggest that 'Advantages' be changed to 'Intended Benefits'. Please comment on this possible change. EdJohnston (talk) 19:46, 2 May 2008 (UTC)

I concur, and I went ahead and did it. It is true that the article is generally lacking references, but the other claims in the article are not generally controversial or likely to be challenged. If references can be found for the disadvantages (or a good argument for keeping them temporarily pending the improvment of the article's references), revert my edit. ASHill (talk) 20:19, 2 May 2008 (UTC)

I oppose the change. No idea about the verifiability of the disadvantages section, but looked like they are likely to be disadvantages of such a system. There have been examples of similar problems with similar systems, such as Westlaw attempting to claim copyright on the citation numbering system for public laws. So since the disadvantages listed are reasonable, I think they should be retained and verification/improvement sought. Since I am not an expert on DOI, I brought them to a section here for further work/contributions.

As far as privacy issues - the phrasing of the item could have been improved, but certainly the privacy statement leaves a lot to be desired compared to standards for information handling practice (e.g. the Code of Fair Information Practice). Zodon (talk) 05:48, 3 May 2008 (UTC)

Digital Object Identifier or digital object identifier

The article is more-or-less totally about the Digital Object Identifier, rather than the generic digital object identifier.

"The uncapitalised term 'digital object identifier' may be used non-specifically to describe a number of varied technologies concerned with the identification of entities in a digital environment. The capitalised term 'Digital Object Identifier' refers to one specific system defined and managed by the International DOI Foundation...". Norman Paskin (2008). "Digital Object Identifier (DOI®) System". p. 2.

The article should be moved to Digital Object Identifier, unless it is going to be expanded to cover digital object identifiers in general. I favour the move. Nurg (talk) 02:48, 22 December 2008 (UTC)

Persistence of metadata?

Article says "metadata are persistent", but is this correct? Is it not the case that the metadata is not persistent, but the DOI is persistent, and that this is precisely the value of having a DOI? Nurg (talk) 01:20, 6 January 2009 (UTC)

Jan 2010 flag additions

Three flags were added to this article by user DGG in Jan 2010. As the contributor who has added most on this article over the last year, I would be happy to help work to solve these problems, with some guidance if necessary. I have edited to remove these flags pending a response to the following queries: “1. This article may be confusing or unclear to readers. Please help clarify the article; suggestions may be found on the talk page. (January 2010).” Reviewing the talk page here, these comments are quite old and seem to relate to earlier versions of the article. I don't see any specific suggestions meriting this flag. I believe most of the issues/problems have been addressed in subsequent 2009 edits, but am happy to help fix any that remain. Please help by drawing specific questions and issues to attention. “2. This article is in need of attention from an expert on the subject. WikiProject Computing or the Computing Portal may be able to help recruit one. (January 2010)”. The article was mainly put together by updating earlier material and adding new material by the Managing Agent of the International DOI Foundation with input from the IDF which created the system and the DOI System users and registration agencies. Consequently it has been authored by experts in the subject. In what further way is it “in need of attention from an expert?”: please provide specific questions and issues to be fixed. “3. This article includes a list of references or external links, but its sources remain unclear because it has insufficient inline citations. Please help to improve this article by introducing more precise citations where appropriate. (January 2010)”. I endeavoured (a) to follow the Wikipedia guidelines, which say that Citations can be presented within articles in one of five ways, the first being “By placing the citation in a list at the end of an article….” and that Editors are free to use any method; (b) to provide a clear narrative, with links to the main sources. If there are specific references where the link is not clear, please mention these specifically. Thanks Npaskin (talk) 12:53, 2 February 2010 (UTC)

Feb 2010 flag additions

User David Eppstein observed “try another tag since the last batch was just reverted without any improvements”. Whilst no changes to the article were made as a result of the earlier (Jan 2010) tags, a response was added to Discussion (see “Jan 2010 flag additions” above) explaining why, and requesting further specific suggestions for change. Please respond with any further specific suggestions.

The tag “technical” [“This article may be too technical for most readers to understand. Please improve this article to make it accessible to non-experts, without removing the technical details”] was added. In response, I have added an introductory section with a brief non-technical overview, followed by a Business Week quote with a reference to the 2001 Business Week article on the DOI System. I have, as requested, not removed any technical details. Npaskin (talk) 08:27, 3 February 2010 (UTC)

An example of the sort of problem that exists in the current article (still): the first sentence has the phrase "managed system". I have no idea what a managed system is supposed to mean. Undefined and confusing jargon is used like this throughout the article. It makes it very difficult to read. —David Eppstein (talk) 16:09, 3 February 2010 (UTC)

More stable than URLs

The 1st para states "naming a document by its DOI provides a more stable mechanism than URLs for linking to online content", and cited http://www.businessweek.com/magazine/content/01_30/b3742032.htm. What this ref actually says is "Assuming the publishers do their job of maintaining the databases, these centralized references, unlike current Web links, should never become outdated or broken." However publishers do not always keep the databases up to date, so the assumption and the conclusion are incorrect, and do not adequately support Wikipedia's statement. Nurg (talk) 04:00, 23 April 2010 (UTC)

Please note: more stable is very far from the same thing as perfectly stable. Your objection would make sense if the lede said that DOIs were infallable, but all it says is that they're not as bad as URLs. —David Eppstein (talk) 05:35, 23 April 2010 (UTC)

Some questions this article could answer

1. Who pays a DOI assignment fee? (author of a book, editor of a journal, librarian, company, what?)
2. Who do they pay it to?
3. What is the order of magnitude of a DOI assignment fee? ($10, $1000, $1 million)
4. What advantages will be enjoyed due to the paying of the fee?
5. What are some examples of countries where the DOI system is actively used? (i.e. places where new publications are being put into the DOI system)
6. What are the the DOI Registration Agencies for the United States, England, Canada, Germany and China?
7. Do DOI references get resolved through a central computer? If so, where is it located?

EdJohnston (talk) 22:21, 3 February 2010 (UTC)

And why do so many DOIs refer to paywalled articles in scholarly journals? --Damian Yerrick (talk | stalk) 02:14, 19 October 2011 (UTC)

Because they refer to journals, and the journals cost money. It mostly is because journals seem to decided to use DOI's.

Pronunciation?

What is the official pronunciation of the acronym? Do-ee, doy, D-O-I? Gareth Jones (talk) 22:07, 8 June 2013 (UTC)

As far as I know, the official pronunciation is D-O-I. See http://lists.ibiblio.org/pipermail/cc-metadata/2004-April/000365.html Robert Nürnberg (talk) 09:59, 25 October 2013 (UTC)

Thanks, that's a good source. Gareth Jones (talk)

Section heading from "DOI names" to "Nomenclature"

I disagree that "DOI names" is actually clearer than "Nomenclature" because the section talks about how the Digital Object Identifier system names objects, where as "DOI names" is ambiguous and somewhat contradictory as a reader would understand the that the system has several names for itself.174.3.125.23 (talk) 09:25, 1 February 2014 (UTC)

My feeling is that the section talks about strings that within the DOI system are called "DOI names", and explains clearly what a DOI name is, so that section title is simple and makes sense. "Nomenclature" is an unnecessarily sesquipedalian circumlocution that contravenes WP:TECHNICAL — by using longer and more obscure words than necessary it makes the article harder to understand. Not to mind that by hinting at the Nomenklatura it brings to mind all the bureaucracy and tendentiousness of the old Soviet Union. —David Eppstein (talk) 16:56, 1 February 2014 (UTC)

I have to disagree with you here. "DOI name" is a term, but a system of terminology is nomenclature. I feel the quality of the article is improved, also encyclopedically, with this section name.174.3.125.23 (talk) 14:27, 4 February 2014 (UTC)

Prose remains difficult

The DOI system offers persistent, semantically-interoperable resolution to related current data and is best suited to material that will be used in services outside the direct control of the issuing assigner (e.g., public citation or managing content of value).

This is hardly grade eight reading level. Plus it persists in confusing the DOI Foundation's aspirational definition with its reality on the ground (no org. ever overpromised and underdelivered).

One could begin by writing "A fully maintained DOI system offers ..." and then move on to tackle "interoperable resolution to related current data" which could be a phrase lifted straight out of a techno-babble satire pastiche. — MaxEnt 21:19, 21 April 2014 (UTC)

Entity Registry System (ERS)

I don't know if this is going anywhere, but see also the Draft:Entity Registry System ★NealMcB★ (talk) 13:57, 4 August 2015 (UTC)

oadoi dot org = risky URL?

While I'm very much in favour of the idea of DOI resolvers that prefer open access versions of articles over paywalled/free-registration-required version, the URL https: slash slash oadoi.org has a quite risky javascript requirement. I've replaced the less than and greater than symbols by LT and GT to avoid triggering off any automatic mediawiki filters. Right now the bottom of the http response I got includes:

LT!-- our actual app --GT
LTscript type="text/javascript" src="/static/ti.js"GTLT/scriptGT

LTscript src="//localhost:35720/livereload.js"GTLT/scriptGT
LT/bodyGT

which looks quite risky - requiring a javascript script to be installed on the user's local machine... I'm not sure if we should show this URL directly...

Boud (talk) 14:07, 27 November 2017 (UTC)

Why is this risky? Person54 (talk) 14:12, 27 November 2017 (UTC)

requiring a javascript script to be installed on the user's local machine is asking for malware.--Quisqualis (talk) 16:15, 27 November 2017 (UTC)

if we started removing links on Wikipedia to pages on other sites which include javascript, we wouldn't have many links left. Javascript is ubiquitous and the days when people disabled Javascript in their browsers or objected to pages with Javascript are ancient history. In this specific case, if you are objecting to livereload.js being loaded from localhost, it is probably the open source developer tool available at https://github.com/livereload/livereload-js. This is a system which allows edits by a developer to HTML files, CSS stylesheets, etc on his or her personal development system ("localhost") to be reflected "live" (i.e. immediately) on the page without a manual reload in the browser. It uses Web Sockets, and is a developer convenience. Probably someone involved in creating the page in question forgot (or didn't bother) to remove some development-time code. The only people affected by this would be other developers who were already running a livereload-js server of their own on their local systems, and it would be harmless because developer A's livereload-js server wouldn't know anything about this HTML page from developer B, the developer of the page in question. Only developer B himself would have a livereload server that would do anything with this page. Anyway, if you aren't a developer and you have some server you don't know about and not under your control running on port 35720 of your local system, serving up malware javascript files when you load random pages on the internet into your browser, your system was already compromised before you ever clicked on a reference link in this Wikipedia article. Person54 (talk) 01:50, 29 November 2017 (UTC)

Links to DOI on Wikipedia

It would be nice if this article (or the talk page) had some kind of links to how DOIs are used in Wikipedia. Some of the cite templates seem to have some way of using them, there is a bot . Still newbie on Wikipedia, so I don't know where such things are documented and how you find them, but if there is some clean way to do it, it would be nice to have a pointer someplace here. Thanks. Zodon (talk) 06:07, 3 May 2008 (UTC)

For example: Interwiki map[1] includes DOI Template:Cite journal/doc has DOI field Bot: User:DOI_bot

I came here after finding a doi: link in a reference in a Wikipedia article, Adhesive_capsulitis_of_shoulder, Dierks and Stevens article, currently ref#9). The link was broken; the reference also contained a broken pubmed URL, but the reference contained a URL that did work; and the target of the working link included the doi number, as given in the reference. Now the claim is made that doi's are permanent, unchanging identifiers. How, then, is it possible for a doi: link to be broken?

So I examined the crossref (an RA) fee structure, at [2]. It says "Content registration (metadata deposit) fees are billed quarterly in arrears." It goes on to say that these are one-time fees, which confused me; but I suspect that these registration fees are recurring fees. So I wonder what happens if you fail to pay your fees at the end of the quarter? Does your 'permanent' doi number get de-registered?

Whatever the reason that doi link was broken, I would consider the fact that a doi link can be broken to be a major disadvantage - sufficiently serious that Wikipedia should refrain from using doi links in references, in favour of archive.org URLs. MrDemeanour (talk) 11:55, 29 May 2018 (UTC)

DOIs can be broken for a few reasons, one of them being the doi resolver being temporarily down, the other one being that there's an issue in a database somewhere. If you click on a broken DOI, there's the option to report it, and that usually fixes the issue (However, don't reporting broken DOI links from Wikipedia, report them from the original database you got the broken DOI from). Headbomb {t · c · p · b} 12:32, 29 May 2018 (UTC)

As it says in the article (third paragraph): "Every time a URL changes, the publisher has to update the metadata for the DOI to link to the new URL.[4][5][6] It is the publisher's responsibility to update the DOI database. By failing to do so, the DOI resolves to a dead link leaving the DOI useless."

Hinnerk R (talk) 19:23, 30 May 2018 (UTC)

Disadvantages section needs work

Though the disadvantages section was removed from the article (see #'Disadvantages' section is POV ). I think it has some potentially valid points, but I don't know the literature in this area to come up with verification. So brought it here for people to work on. Would appreciate help with references, improvement suggestions, etc. Thanks. Zodon (talk) 05:26, 3 May 2008 (UTC)

I think the Wikipedia policies are clear. This is no place for us to insert our own personal opinions. Unless a published source has criticized DOI's privacy policies, we have no business adding criticism here. WP:NOR, WP:SYNTH and all that. EdJohnston (talk) 17:56, 4 May 2008 (UTC)

As I understand it, items who's verifiability has been questioned may be moved to the talk page to be worked on (discussed, sources gathered, etc.) Wikipedia:Verifiability I don't see anything on WP:NOR or WP:SYNTH that obviously contradicts this practice. The opinions I expressed (that this is not unlikely to be verifiable, and that it seems worth pursuing) seem germane to improving the article. If working on/discussing this material on the articles talk page is not accepted practice, please indicate where that information is given in the documentation (so I can better comply with accepted practice), and where such discussion is appropriate. Thanks. Zodon (talk) 18:40, 4 May 2008 (UTC)

That's fair; I think the comment above refers to including the material in the article, not the talk page. If a reference is found, then we can certainly include the criticism in the article, and keeping the material here until that happens is the right thing to do. However, I've never heard the privacy concerns about DOIs except in this discussion. That doesn't mean verifiable criticism doesn't exist; I just haven't seen it.

Preemptive warning: It will need to be a WP:Reliable source, which excludes a self-published source expressing these concerns. ASHill (talk | contribs) 19:05, 4 May 2008 (UTC)

Disadvantages

There are some issues to consider before adopting or using the DOI:

• Neutrality of code issuing or resolving organization : code-issuing or resolution restrictions inherits the dangers of censorship. IDF-FAQ-#22
• Privacy issues : ownership, security, data retention, and future usage plans for its resolution server logs still unclear. DOI-Privacy-Statement
• Lack of recognition and adoption compared with ISBN, or even Amazon's ASIN.
• Additional point of failure: additional resolution mechanism means another potential point of failure.
• Identifiers are numeric, so more difficult for people to use than alphanumeric systems.

x6 ?TFXZ0auf 97.81.98.1 (talk) 16:47, 24 August 2021 (UTC)

contentdirections.com

contentdirections.com going out of business and abandoning https://dx.doi.org/10.2277 might be worth mentioning. — Preceding unsigned comment added by AManWithNoPlan (talk • contribs) 02:57, 4 June 2022 (UTC)

I can find sources for contentdirections commercializing doi registration [3] but not to their leaving the market. Do you have any? Without sources we cannot mention it. —David Eppstein (talk) 05:30, 4 June 2022 (UTC)

https://dx.doi.org/10.2277 has been blocked from adding new ones. Also, they seem to all fail to resolve since they all point to the contentdirections.com website. It looks like they have changed names, rather than being bankrupt. https://www.linkedin.com/in/david-sidman-39110b https://www.buzzfile.com/business/Linkstorm-212-792-1847_no:4506533 AManWithNoPlan (talk) 15:37, 4 June 2022 (UTC)