We need markup for science blogs

Like many other blogging platforms, Nature Network uses Textile to format blog entries. Textile is a markup language, and Andrew Sun has nicely put together the most important tags in this blog post.

Unfortunately, Textile doesn't much help with formatting that is specific to science blogging. Specifically, there is no standard way to link to journal articles. Should we use the DOI, should we link to Pubmed, or should we link to the journal homepage? If we do the latter, should we link to the fulltext (that might not be accessible to everybody) or the abstract? And why is there no standard markup for this?

MediaWiki, the platform behind Wikipedia, uses a different markup language. Here, linking to PubMed is semiautomated. Linking to DOIs is also supported. And the Biblio extension, also used by OpenWetWare, makes linking to papers even easier.

We have often talked about standard data formats. And the small things also matter. One result of standard blog linking to papers: it would become much easier to find all blog posts that link to a particular paper.

This entry was posted in Snippets. Bookmark the permalink.

15 Responses to We need markup for science blogs

  1. Richard P. Grant says:

    What’s wrong with DOI? I love it.

  2. Massimo Pinto says:

    I guess there may be many users of this Network who cannot find their journals on PubMed. Is DOI more general to all science disciplines?

  3. Andrew Sun says:

    “The Chem Blog”:http://www.thechemblog.com/ has a fancy feature that recognizes the source host of your hyperlink text and automatically applies the corresponding favicon after the link.

  4. Mark Tummers says:

    I hardly ever click on links in blog entries. Unless the link promises to be more interesting than the blog entry itself.

  5. Euan Adie says:

    I’m pretty sure that the Network team is working on better tools for markup… but for the record best practice is definitely linking to papers using their DOI using *http://dx.doi.org/[ your doi goes here, slash n’ all ]*.
    On that note we’ve got some cool (IMHO ;)) new services for bloggers coming up that track the papers you mention in your posts – no launch date yet but will happily give a demo to anybody at the science blogging conference in a couple of weeks!

  6. Martin Fenner says:

    Always using the DOI is obviously the first step, and something I still have to get used to. The next step would be best practices of how to reference papers in blog posts. Every journal has a style for references, so why not science blogs[1]? Textile footnotes work great for this.
    fn1. *Diemand J et al.* “doi:10.1038/nature07153″:http://dx.doi.org/10.1038/nature07153
    A little extension to the Textile markup would make citing papers a little easier. Everything that starts with *doi:* should automatically turn into a link, e.g. “doi:10.1038/nature07153″:http://dx.doi.org/10.1038/nature07153.
    Euan, I’m looking forward to this paper tracking tool.

  7. Maxine Clarke says:

    To Massimo: The doi is a unique identifier and is applied to all articles published by journals in the CrossRef consortium, which is most of them. It is a journal equivalent to the book’s ISBN. So yes, it is common to all scientific disciplines and is in wide use.

  8. Frank Norman says:

    There is a big problem with DOIs for very new articles in my experience – the database does not seem to updated quickly when new articles appear online. Many times I have written a news item and put a DOI link to an article only to find that users get a message “DOI does not exist” or somesuch. Mostly this has been for journals from Elsevier and Nature Pub Group, but I suspect it is a general problem.

  9. Richard P. Grant says:

    I was going to blog about that – Nature is particularly badly affected, as you say.

  10. Christian Hauschke says:

    FYI: I’m just working on an extension for WordPress to include and cite bibliographic data in blog postings. It’s currently in alpha phase.
    But that won’t help you here, I guess.

  11. Martin Fenner says:

    Christian, this is exactly what I had in mind. Please keep me posted.

  12. Christian Hauschke says:

    Until now only ISBN is supported, but it shouldn’t be to complicated to add support for DOI and URN, too. I’ll keep you informed.

  13. Christian Hauschke says:

    Here we go: ISBN2WP (alpha), a wordpress plugin to integrate bibliographic data.
    Please don’t raise to high expactations. Until now it’s more a proof of concept than anything else.

  14. Martin Fenner says:

    Christian, thanks a lot for the link to the plugin. Now we need something similar for Nature Network…