[Taxacom] the hurdle for all biodiv informatics initiatives

Sat Feb 20 19:36:53 CST 2010

Hi Tony et al.,

Just to flesh out your comments a bit, and respond a little:

>- WikiXXX are highly responsive (anyone can update or add anything immediately), but potentially chaotic for the same reason, also do not have what "database" persons (including myself) would consider a robust, hierarchical and efficient/internally consistent structure

potentially chaotic?? Not so fast! A good Wikispecies page is written in such a way that it is verifiable against the cited sources. It is the responsibility of the user to verify the information*. If someone edits a page to make it less verifiable (which can be seen by comparing versions in the page history), then their edit can be justifiably reverted, and if they continue to try to redo it, then they can be justifiably blocked from editing. Where is the chaos?
* in practice, the user won't want to have to do this, but IMHO they have just as much reason to verify it from a closed source database - trusting authority (or even trusting "peer review") is in practice not a robust guarantee of reliability.
>also do not have what "database" persons (including myself) would consider a robust, hierarchical and efficient/internally consistent structure
I'm not sure why Wikispecies isn't "robust" or "hierarchical", but at the end of the day it mostly comes down to the quality of the information - if the info you are getting from a robust, hierarchical and efficient/internally consistent structure is less reliable/up-to-date, then it would be highly irrational not to make cautious use of Wikispecies. "Suck it and see", as they say! I bet, if there was someone who was an "oracle of all knowledge", scientists wouldn't just ignore them out of hand as an "unreliable source". After the oracle had correctly predicted the first dozen or so winning lottery tickets, their talent would not be ignored! So, it is irrational to dismiss Wikispecies as a priori unreliable or chaotic - you have to try it out and see ...

>So maybe the task is to build a sweet spot of some sort between these two approaches. The EOL's etc. have talked about "annotation" facilities to do this\
Unfortunately, from what I can gather, the "annotation" facility for EoL would amount to little more than a placebo sugar pill for the annotater. The annotation would be kept well out of obvious sight of relevant page, which would still be explicitly or implicitly deemed to be the "officially endorsed" version. In practice, I don't think such annotations would be treated seriously or even noticed ...

I have just been looking at CoL for the beetle group Scarabaeoidea, and I am gobsmacked to see that they recognise the dynastines as a separate family, Dynastidae! This goes against the prevailing taxonomic/phylogenetic opinion of all but a few dynastine only specialists. Specialists in one group tend to elevate the taxonomic status of their speciality group. I don't see how CoL's mission to provide a single classification could possibly be a sensible one. Surely, the best single overall classification depends on ongoing phylogenetic studies, and is the domain of ToL, if anyone? Surely, we don't want just some arbitrarily chosen consistent classification? Having Dynastidae as a family suggests that they are not paying much attention to phylogenetic studies ...

By the way, I have just added to Wikispecies 15 dryinid names from a Chinese book published in 2002 that have not been picked up by any other database (to my knowledge), not even Zoological Record ...

Cheers,

Stephen

________________________________________
From: taxacom-bounces at mailman.nhm.ku.edu [taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Tony.Rees at csiro.au [Tony.Rees at csiro.au]
Sent: Sunday, 21 February 2010 9:14 a.m.
To: tas27 at schweich.com; taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] the hurdle for all biodiv informatics initiatives

Dear all,

I believe Tom (and others who have contributed on this topic - thanks Stephen) has put the finger on an important point. Essentially (in my view, anyway):
- WikiXXX are highly responsive (anyone can update or add anything immediately), but potentially chaotic for the same reason, also do not have what "database" persons (including myself) would consider a robust, hierarchical and efficient/internally consistent structure
- The more "official" (what Stephen refers to as "closed") projects tend to have (at least notionally) more controlled content and a better database structure, but can be less responsive to the man in the street (or read non-authorised contributor if you prefer), i.e. are missing a lot of potential contributions, spotting and correcting of errors, etc.

So maybe the task is to build a sweet spot of some sort between these two approaches. The EOL's etc. have talked about "annotation" facilities to do this. The wikiXXX *may* be going in the direction of more taxonomic scrutiny/database rigour (or then again, not.).

Perhaps this is an opportunity for some further debate and looking for ways to leverage both aspects for the common benefit.

Regards - Tony
________________________________________
From: taxacom-bounces at mailman.nhm.ku.edu [taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Tom Schweich [tas27 at schweich.com]
Sent: Sunday, 21 February 2010 3:46 AM
To: taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] the hurdle for all biodiv informatics initiatives

I dunno, guys.  I've been reading this "debate" and trying to make sense
of it, then looking at the biodiversity data bases in light of my very
small area of knowledge, the desert Frasera (Gentianaceae).

I agree that the data is rather sparse. I also would say that what data
exists is a little on the disappointing side, and the prospects for
correcting things are dim.

As a small example, GBIF and EOL will tell you that Frasera paniculata
occurs in Lincoln County, Nevada.  They will even give us geographic
coordinates to six digits to the right of the decimal point (37.641515N,
114.875416W).  They do not, however, give us a clue that those
coordinates just happen to be the geographic center of Lincoln County,
Nevada.

I happen to believe that this observation is the result of a single
mis-identified specimen (Marie Gentry #41, Frasera utahensis, Panaca
Valley and vicinity, Summer 1941. Annotated: Frasera paniculata, D. M.
Post, 1954).  I have seen this collection at UC and CAS, and it is, in
fact, Frasera albomarginata.  Any idiot can see that, especially me.

What's happened here is that a unknown location of "Panaca Valley and
vicinity" became  a low-accuracy location of "somewhere in Lincoln
County" and was then translated into extremely high-precision
coordinates. Not good.  So I would say the geographers behind the data
bases have not adequately determined a method to display the geographic
uncertainty behind an observation.

Thinking about fixing this mistake:  when I try to trace back the
observation to its source, in GBIF and EOL, I get an un-followable
reference to a USDA observation data base OR a reference to an
unpublished document.  Well, that's not too helpful. A couple of hours
to learn that the error could not be traced was more than I wanted to
spend.

So, looking at EOL, GBIF, and Wikispecies, I think I have an easier and
greater chance of correcting the mistaken impression that Frasera
paniculata occurs in Nevada using Wikispecies rather than EOL or GBIF.
So I might be tempted to put effort into Wikispecies rather than the
others.

Tom

--
Tom Schweich KJ6BIT tas27 at schweich.com
http://www.schweich.com
http://twitter.com/schweich

dipteryx at freeler.nl wrote:
> Van: Stephen Thorpe [mailto:s.thorpe at auckland.ac.nz]
> Verzonden: za 20-2-2010 0:14
>
> Hi Paul,
> A frequent complaint you seem to voice about all biodiversity databases, including Wikispecies, is that they don't, by your estimation, contain much in the way of "useful" information. To my mind, however, they function to organise vast numbers references (preferably with links of some kind to those references) in a taxonomic way. The "useful information" is contained in the references, and not in the database per se. This is certainly how I view Wikispecies - a vast taxonomically organised library/bibliography, supplemented where possible with images...
> Cheers,
> Stephen
>
> ***
> Yes, among the concerns I have voiced (consistently, I hope, rather
> than "frequently") is that the biodiversity databases appear to be
> mostly empty infrastructure, waiting for content to (magically?)
> manifest itself. (The only aspect where content is 'magically'
> manifesting itself is in the form of pictures: a surprising amount
> of pictures is available on the web, in great part from books
> that by their age have entered the public domain, but also from
> enthusiastic photographers)
>
> In how far references can yield this "useful information" is an open
> question. For some species the original description or the most recent
> monograph will yield state-of-the-art information. However, this will
> not universally, or even generally, be the case: often a lot more
> information will exist. In extreme cases so much information exists for
> a single species that even a mere bibliography can be daunting.
>
> But certainly it is very desirable for Wikispecies (or any biodiversity
> initiative) to list (taxon by taxon) what taxonomic treatment is being
> followed (backed up by a reference where necessary). Most parts of
> Wikispecies that I have seen fall short of this standard.
>
> Paul
>
>
> _______________________________________________
>
> Taxacom Mailing List
> Taxacom at mailman.nhm.ku.edu
> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
>
> The Taxacom archive going back to 1992 may be searched with either of these methods:
>
> (1) http://taxacom.markmail.org
>
> Or (2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here
>
>

--
Tom Schweich KJ6BIT tas27 at schweich.com
http://www.schweich.com
http://twitter.com/schweich

_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of these methods:

(1) http://taxacom.markmail.org

Or (2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here
_______________________________________________

Taxacom Mailing List
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

The Taxacom archive going back to 1992 may be searched with either of these methods:

(1) http://taxacom.markmail.org

Or (2) a Google search specified as:  site:mailman.nhm.ku.edu/pipermail/taxacom  your search terms here