How many species are there, and why do we get two very different answers from same data?

Two papers estimating the total number of species have recently been published, one in the open access journal PLoS Biology:

Camilo Mora, Derek P. Tittensor, Sina Adl, Alastair G. B. Simpson, Boris Worm. How Many Species Are There on Earth and in the Ocean?. PLoS Biol 9(8): e1001127. doi:10.1371/journal.pbio.1001127

the second in Systematic Biology (which has an open access option but the authors didn't use it for this article):

Mark J. Costello, Simon Wilson and Brett Houlding. Predicting total global species richness using rates of species description and estimates of taxonomic effort. Syst Biol (2011) doi:10.1093/sysbio/syr080

The first paper has gained a lot of attention, in part because Jonathan Eisen Bacteria & archaea don't get no respect from interesting but flawed #PLoSBio paper on # of species on the planet was mightily pissed off about the estimates of the number:

Their estimates of ~ 10,000 or so bacteria and archaea on the planet are so completely out of touch in my opinion that this calls into question the validity of their method for bacteria and archaea at all.

The fuss over the number of bacteria and archaea seems to me to be largely a misunderstanding of how taxonomic databases count taxa. Databases like Catalogue of Life record described species, and most bacteria aren't formally described because they can't be cultured. Hence there will always be a disparity between the extent of diversity revealed by phylogenetics and by classical taxonomy.

The PLoS Biology paper has garnered a lot more reaction than the Systematic Biology paper (e.g., the commentary by Carl Zimmer in the New York TimesHow Many Species? A Study Says 8.7 Million, but It’s Tricky), which arguably has the more dramatic conclusion.

How many species, 8.7 million, or 1.8 to 2.0 million?

Whereas the Mora et al. in PLoS Biology concluded that there are some 8.7 million (±1.3 million SE) species on the planet, Costello et al. in Systematic Biology arrive at a much more conservative figure (1.8 to 2.0 million). The implications of these two studies are very different, one implies there's a lot of work to do, the other leads to headlines such as 'Every species on Earth could be discovered within 50 years'.

What is intriguing is that both studies use the same databases, Catalogue of Life and the World's Register of Marine Species, and yet arrive at very different results.

So, the question is, how did we arrive at two very different answers from the same data?

Taylor and Francis Online breaks DOIs - lots of DOIs

DOIs are meant to be the gold standard in bibliographic identifier for article. They are not supposed to break. Yet some publishers seem to struggle to get them to work. In the past I've grumbled about BioOne, Wiley, and others as cuplrits with broken or duplicate or disappearing DOIs.

Today's source of frustration is Taylor and Francis Online. T&F Online is powered by (Atypon), which recently issued this glowing press release:

SANTA CLARA, Calif.—20 September 2011—Atypon®, a leading provider of software to the professional and scholarly publishing industry, today announced that its Literatum™ software is powering the new Taylor & Francis Online platform (www.TandFOnline.com). Taylor & Francis Online hosts 1.7 million articles.
...
"The performance of Taylor & Francis Online has been excellent," said Matthew Jay, Chief Technology Officer for the Taylor & Francis Group. "Atypon has proven that it can deliver on schedule and achieve tremendous scale. We're thrilled to expand the scope of our relationship to include new products and developments."

Great, except that lots of T&F DOIs are broken. I've come across two kinds of fail.

DOI resolves to server that doesn't exist
The first is where a DOI resolves to a phantom web address. For example, the DOI doi:10.1080/00288300809509849 resolves to http://tandfprod.literatumonline.com/doi/abs/10.1080/00288300809509849. But the domain tandfprod.literatumonline.com doesn't exist, so the DOI is a dead end.

DOI doesn't resolve
Taylor and Francis have digitised the complete Annals and Magazine of Natural History, a massive journal comprising nearly 20,000 articles from 1841 to 1966, and which has published some seminal papers, including A. R. Wallace's "On the law which has regulated the introduction of new species" doi"10.1080/037454809495509 which forced Darwin's hand (see the Wikipedia page for the successor journal Journal of Natural History. Taylor and Francis are to be congratulated for putting such a great resource online.

Problem is, I've not found a single DOI for any article in Annals and Magazine of Natural History that actually works. If you try and resolve the DOI for Wallace's paper, doi"10.1080/037454809495509, you get the dreaded "Error - DOI not found" web page. So something like 20,000 DOIs simply don't work. The only way to make the DOI work is append it to "http://www.tandfonline.com/doi/abs/", e.g. http://www.tandfonline.com/doi/abs/10.1080/037454809495509. This gets us to the article, but rather defeats the purpose of DOIs.

Why?
Something is seriously wrong with CrossRef's quality control. It can't be too hard to screen all domains to see if they actually exist (this would catch the first error). It can't be too hard to take a random sample of DOIs and check that they work, or automatically check DOIs that are reported as missing. In the case the Annals and Magazine of Natural History the web page for the Wallace article states that it has been available online since 16 December 2009. That's a long time for a DOI to be dead.

There is a wealth of great content that is being made hard to find by some pretty basic screw ups. So CrossRef, Atypon and Taylor and Francis, can we please sort this out?

Linked data that isn't: the failings of RDF

OK, a bit of hyperbole in the morning. One of the goals of RDF is to create the Semantic Web, an interwoven network of data seamlessly linked by shared identifiers and shared vocabularies. Everyone uses the same identifiers for the same things, and when they describe these things they use the same terms. Simples.

Of course, the reality is somewhat different. Typically people don't reuse identifiers, and there are usually several competing vocabularies we can chose from. To give a concrete example, consider two RDF documents describing the same article, one provided by CiNii, the other by CrossRef. The article is:

Astuti, D., Azuma, N., Suzuki, H., & Higashi, S. (2006). Phylogenetic Relationships Within Parrots (Psittacidae) Inferred from Mitochondrial Cytochrome-b Gene Sequences(Phylogeny). Zoological science, 23(2), 191-198. doi:10.2108/zsj.23.191

You can get RDF for a CiNii record by appending ".rdf" to the URL for the article, in this case http://ci.nii.ac.jp/naid/130000017049. For CrossRef you need a Linked Data compliant client, or you can do something like this:

curl -D - -L -H "Accept: application/rdf+xml" "http://dx.doi.org/10.2108/zsj.23.191"

You can view the RDF from these two sources here and here.

No shared identifiers
The two RDF documents have no shared identifiers, or at least, any identifiers they do share aren't described in a way that is easily discovered. The CrossRef record knows nothing about the CiNii record, but the CiNii document includes this statement:

<rdfs:seeAlso rdf:resource="http://ci.nii.ac.jp/lognavi?name=crossref
&amp;id=info:doi/10.2108/zsj.23.191" dc:title="CrossRef" />

So, CiNii knows about the DOI, but this doesn't help much as the CrossRef document has the URI "http://dx.doi.org/10.2108/zsj.23.191", so we don't have an explicit statement that the two documents refer to the same article.

The other shared identifier the documents could share is the ISSN for the journal (0289-0003), but CiNii writes this without the "-", and uses the PRISM term "prism:issn", so we have:

<prism:issn>02890003</prism:issn>

whereas CrossRef writes the ISSN like this:

<ns0:issn xmlns:ns0="http://prismstandard.org/namespaces/basic/2.1/">
0289-0003</ns0:issn>

Unless we have a linked data client that normalises ISSNs before it does a SPARQL query we will miss the fact that these two articles are in the same journal.

Inconsistent vocabularies
Both CiNii use the PRISM vocabulary to describe the article, but they use different versions. CrossRef uses "http://prismstandard.org/namespaces/basic/2.1/" whereas CiNii uses "http://prismstandard.org/namespaces/basic/2.0/". Version 2.1 versus version 2.0 is a minor difference, but the URIs are different and hence they are different vocabularies (having version numbers in vocabulary URIs is asking for trouble). Hence, even if CiNii and CrossRef wrote ISSNs in the same way, we'd still not be able to assert that the articles come from the same journal.
Inconsistent use of vocabularies
Both CiNii use FOAF for author names, but they write the names differently:

<foaf:name xml:lang="en">Suzuki Hitoshi</foaf:name>

<ns0:name xmlns:ns0="http://xmlns.com/foaf/0.1/">Hitoshi Suzuki</ns0:name>

So, another missed opportunity to link the documents. One could argue this would be solved if we had consistent identifiers for authors, but we don't. In this case CiNii have their own local identifiers (e.g. http://ci.nii.ac.jp/nrid/1000040179239), and CrossRef has a rather hideous looking Skolemisation: http://id.crossref.org/contributor/hitoshi-suzuki-2gypi8bnqk7yy.

In summary, it's a mess. Both CiNii and CrossRef organisations are whose core business is bibliographic metadata. It's great that both are serving RDF, but if we think this is anything more than providing metadata in a useful format I think we may be deceiving ourselves.