University of Glasgow

Roderic Page

This week saw the release of two tools from the Biodiversity Heritage
Library, [CiteBank](http://citebank.org/) and the [BHL-Europe
portal](http://prototype.bhle.eu/). Both have actually been quietly
around for a while, but were only publicly announced last week.  
  
In developing a new tool there are several questions to ask. Does
something already exist that meets my needs? If it doesn\'t exist, can I
build it using an existing framework, or do I need to start from
scratch? As a developer it\'s awfully tempting sometimes to build
something from scratch (I\'m certainly guilty of this). Sometimes a more
sensible approach is to build on something that already exists,
particularly if what you are building upon is well supported. This is
one of the attractions of [Drupal](http://drupal.org), which underlies
CiteBank and Scratchpads. In my own work I\'ve used Semantic Mediawiki
to support editable, versioned databases, rather than roll my own.
Perhaps the more difficult question for a developer is whether they need
to build anything at all. What if there are tools already out there
that, if not *exacty* what you want, are close enough (or most likely
will be by the time you finish your own tool).  
  
**CiteBank**  
![bhlsquare_reasonably_small.png](http://lh4.ggpht.com/_Gct8lVAxKqQ/TP8UnImUYbI/AAAAAAAAAzQ/yVscmVVG7Gc/bhlsquare_reasonably_small.png?imgmax=800){border="0"
width="128" height="128" align="right"}  

> CiteBank is an open access platform to aggregate citations for
> biodiversity publications and deliver access to biodiversity related
> articles. CiteBank aggregates links to content from digital libraries,
> publishers, and other bibliographic systems in order to provide a
> single point of access to the world's biodiversity literature,
> including content created by its community of users. CiteBank is a
> project of the Biodiversity Heritage Library (BHL).

  
I have two reactions to CiteBank. Firstly, Drupal\'s bibliographic tools
really suck, and secondly, why do we need this? As I\'ve argued earlier
(see [Mendeley, BHL, and the \"Bibliography of
Life\"](http://iphylo.blogspot.com/2010/10/mendeley-bhl-and-of-life.html)),
I can\'t see the rationale for having CiteBank separate from an existing
bibliographic database such as [Mendeley](http://www.mendeley.com) or
[Zotero](){http:="" www.zotero.org\"=""}. These tools are more mature,
better supported, and address user needs beyond simply building lists of
papers (e.g., citing papers when writing manuscripts).  
  
For me, one of BHL\'s goals should be integrating the literature they
have scanned into mainstream scientific literature, which means finding
articles, assigning DOIs, and becoming in effect a digital publishing
platform (like [BioOne](http://www.bioone.org) or
[JSTOR](http://www.jstor.org)). Getting to this point will require
managing and cleaning metadata for many thousands of articles and books.
It seems to me that you want to gather this metadata from as many
sources as possible, and expose it to as many eyes (and algorithms) as
possible to help tidy it up. I think this is a clear case of it being
better to use an existing tool (such as Mendeley), rather than build a
new one. If a good fraction of the world\'s taxonomists shared their
person bibliographies on Mendeley we\'d pretty much have the world\'s
taxonomic literature in one place, without really trying.  
  
**BHL-Europe**  
![logo.jpg](http://lh4.ggpht.com/_Gct8lVAxKqQ/TP8O5qEeyCI/AAAAAAAAAzI/kbRRwiRdbKg/logo.jpg?imgmax=800){border="0"
width="200" align="right"}  
It\'s early days for BHL-Europe, and they\'ve taken the \"lets use an
existing framework\" approach, basing the BHL-Europe portal on
[DISMARC](http://www.dismarc.org/), the later being a EU-funded project
to \"encourage and support the interoperability of music related
data\".  
  
BHL-Europe is the kind of web site only its developers could love. It\'s
spectacularly ugly, and a classic example of what digital libraries came
up with while Google was quietly eating their lunch. Here\'s the web
site showing search results for \"Zonosaurus\":  
  

::: {style="text-align:center;"}
![bhleu.png](http://lh5.ggpht.com/_Gct8lVAxKqQ/TP8ZVEnxFcI/AAAAAAAAAzY/nHZ5zCLU0Pc/bhleu.png?imgmax=800){border="0"
width="400"}
:::

  
  
Yuck! [Why do these things have to be so
ugly?](http://www.slideshare.net/andraz/semtech2010-do-semanticwebuserinterfaceshavetobeugly).
DISMARC was designed to store metadata about digital objects,
specifically music. Look at commercial music interfaces such as iTunes,
[Spotify](http://spotfy.com), and [Last.fm](http://last.fm). Or even
academic projects such as [mSpace](http://research.mspace.fm/).  
  
To be useful BHL-Europe really needs to provide an interface that
reflects what its users care about, for example taxonomic names,
classification, and geography. It can\'t treat scientific literature as
a bunch of lifeless metadata objects (but then again, DISMARC managed to
do this for music).  
  
**Where next?**  
CiteBank and BHL-Europe seem further additions to the worthy but
ultimately deeply unsatisfying attempts to improve access biodiversity
literature. To date our field has failed to get to grips with
aggregating metadata (outside of the library setting), creating social
networks around that aggregation, and providing intuitive interfaces
that enable users to search and browse productively. These are big
challenges. I\'d like to see the resources that we have put to better
use, rather than being used to build tools where suitable alternatives
already exist (CiteBank), or used to shoe horn data into generic tools
that are unspeakably ugly (BHL-Europe portal) and not fit for purpose.
Let\'s not reinvent the wheel, and let\'s not try and convince ourselves
that squares make perfectly good wheels.


This week saw the release of two tools from the Biodiversity Heritage Library, CiteBank and the BHL-Europe portal. Both have actually been quietly around for a while, but were only publicly announced last week.In developing a new tool there are several questions to ask. Does something already exist that meets my needs? If it doesn't exist, can I build it using an existing framework, or do I need to start from scratch?

Rogue Scholar Beiträge

First thoughts on CiteBank and BHL-Europe