s.thorpe at auckland.ac.nz
Wed Sep 16 19:07:01 CDT 2009
I congratulate Richard on putting the issue into words with excellent clarity - I can probably infer from this that he is not a machine! There is a human being attached to his computer! I would think that "same name" recognition is only one of the things than humans can do better than computers. Others include being able to "see" non-obvious errors (inferences based on experience), and being able to judge how important/worthwhile/relevant something is. The result of entirely automated bioinformatics would be chaos - we need humans to at least take the initial computer generated info and do the hard work of interpreting it all properly. The problem is that this is a slow and delicate process, which is what I think Paul may have been alluding to when he suggested that the outcome of a bioinformatics project is likely just to be an insurmountable mass of raw data of little use! I suspect bioinformatics people are behind the current trend to regiment the format of primary taxonomic literature from now on, thereby making it easier for computers to digest. Trouble is that the foundations of modern taxonomy are in old and unregimented literature spanning 4 centuries! Realistically, it is "pie in the sky" to track every single use of a taxon name - it is akin to counting every grain of sand on a beach. We need to be able to focus on what is important, and move on from there in a constructive way, rather than being dragged back by a virtual infinity of tiny little problems of no consequence. And this is coming from someone (me), who has often been branded a "mere details person"! One option is to concentrate less on providing an answer to everything anyone might want to know, but rather provide them with a means of finding it out for themselves. To this end, money might be well spent on digitising more and more primary taxonomic literature, and making it free on the web, along with good images of taxa, etc. Do we really want a world in which the answer to everything is just one mouse click away?* I for one like tracking things down for myself ...
*Amusing to imagine the reaction you would get from people 30 years ago, if you had said that in the future, the answer to everything would be just a mouse click away!
From: taxacom-bounces at mailman.nhm.ku.edu [taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Richard Pyle [deepreef at bishopmuseum.org]
Sent: Thursday, 17 September 2009 11:32 a.m.
To: taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] globalnames?
> Giving the poster a very quick male goose(!), two thoughts strike me:
> (1) Isn't this just a fancy way of saying that in order to
> keep track of information about a (recognised) species, we
> need to track its synonymy? Clearly, this has been done in
> non-electronic taxonomic literature since its infancy!
No -- the problem that GNI is attacking is specific to digital information.
GNUB is effectively a way to track not just synonymies, but all usages of
names. But GNI is about untangling the mess of text strings purported to
represent taxon names. It's easy for a human to look at a taxon name
represented by 20 slightly different textual orthographies and resolve them
as the "same name" quickly and easily. It's not quite so easy for a
computer to do this. So to help the computers build links among electronic
biodiversity information, we need tools and services like this. Ultimately,
the point is to reduce the time it used to take for a taxonomists to track
down an old publication, into the time it takes to click a mouse button. We
all *know* the potential the internet has to make information important to
taxonomists (or anyone else interested in biodiversity) ENORMOUSLY more
accessible. We also know that this potential is nowhere near being fully
realized yet. Tools and services like GNI and GNUB are specifically
designed to help get us there.
> I guess I would like to see bioinformatics projects being
> designed for the real world, with real world limitations and
> problems ...
It sounds like all of us are in full agreement on this point, at least!
Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
The Taxacom archive going back to 1992 may be searched with either of these methods:
Or (2) a Google search specified as: site:mailman.nhm.ku.edu/pipermail/taxacom your search terms here
More information about the Taxacom