I read this post on whether DOI is a good identifier or not. My feeling is that it has the following weaknesses:

It cannot (normally) be generated from citation information (a big disadvantage for an identifier) – you have to resolve them at e.g. CrossRef. This kills it as a way to communicate articles effectively.

If you want to resolve lots of them, you have to pay (there is no real value in this.. except that they have the identifiers and you do not).

It does not replace the URL, it is simply a redirect. This makes it hard to bookmark and those unfamiliar with the system who think they have bookmarked it have in fact bookmarked the URL.

Also, publishers have to pay for it too (though its possible they may receive money from CrossRef too). Essentially, all they are paying for is an unintuitive link that does not break provided they keep the redirect up to date.

Hence OpenURL.

It creates a persistent link as DOI does except it actually exists as a webpage (it is not a redirect) and can therefore be bookmarked easily and it CAN be generated from citation information without permissions. Here is a useful implementation.

A note on the CrossRef website caught my eye. It states that OpenURL is not competitive with DOI. This, of course, is nonsense (since it addresses link permanency). Apparently:

An OpenURL link that contains a DOI is similarly persistent.” [as a link]

Why would an OpenURL pointing to a publisher website not be persistent without a DOI? OpenURL can be created with citation data so it is TOTALLY persistent. With DOI, you need to fill in a form at CrossRef or Doi.org which you do not need to do with OpenURL.

It is DOIs that need third party ‘resolving’, not URLs and especially not OpenURLs which require no link up to a database (a restricted one in the case of CrossRef) for generation.

So, it is a shame that only a few publishers have taken it up. Surely, it is a competitive advantage to use a totally freely available URL structure that anyone can generate? After all, the worst that could happen is that someone might find your articles more easily.

Stumble it!

4 Responses to “OpenURL”

  1. baoilleach says:

    Just to note that OpenURL (unlike the OpenRef system in my blog post or indeed the DOI) is not unique for a particular publication; several different OpenURLs can point to the same paper. This may make it difficult to connect information relating to the same publication by inspecting a bookmarked OpenURL. The OpenURL really describes an agreed system for key-value pairs to describe publication metadata, as well as to describe the context of a request for a paper.

  2. David Bradley says:

    I think this concept could have legs, but maybe we could escape the proprietary closed nature of DOI altogether and create a PaperID that is initiated by the author using standards. I just posted about Open Source PaperID idea on Chemspy.


  3. Mitch says:

    That’s just being silly. On many levels a redirect is much more useful than a hard url. I’m not even that old, and I’ve lived through enough url modifications from publishers to be completely turned off hard URLs. Hell, http://blog.chemicalforums.com now exits as a redirect. ;)


  4. will says:

    If the redirect were the only issue I would agree.

    “There is no charge for most libraries to get a CrossRef account to retrieve DOIs or metadata” – http://crossref.org/03libraries/index.html

    So some libraries will be charged then. Libraries would have zero metadata retrieval costs if OpenURL/similar system were implemented. Charging for metadata (zero value on its own) is what is really silly.

Leave a Reply

Spam protection by WP Captcha-Free