[Taxacom] encyclopedia of life

Doug Yanega dyanega at ucr.edu
Mon May 14 18:59:23 CDT 2007


I've been away, and only just returned to find all of this 
discussion. Naturally, as always, I have an opinion. ;-)

Richard Zander wrote:

>Scholarpedia is directly relevant to EoL, Paul.
>
>I'm sure I do not like the idea of an invited "curator" sitting on a
>page like a dog in a manger. I hope the EoL does not follow this model.

There is no one model that will please everyone. Period.

Wikipedia, Wikispecies, Citizendium, Scholarpedia, and such are all 
geared to give a single authoritative answer. While some of them can 
accommodate for a dispute, that is not what they are designed to do. 
In the present case, the wiki that comes closest to the EoL is 
Wikispecies - and Wikispecies does NOT allow for multiple 
classifications to coexist. Any given taxon name must have a Linnaean 
rank, and only one. This is what the general public and science-using 
public needs and desires to see: an authoritative scheme. High school 
teachers are not going to care whether termites have been recently 
re-classified as a family of roaches UNLESS no one in the world is 
ever going to refer to them as "Isoptera" again, and all the 
resources they use change to that scheme as well.

Taxonomists, on the other hand, need to know about all the competing 
hypotheses and alternative classifications. That's going to make 
things VERY messy, if you plan on indicating this for every single 
species in existence.

Consider the following example: look at nearly any North American 
fritillary butterfly and imagine what an EoL species page would look 
like for any of them. Would you expect such a page to list any or all 
of the higher taxa to which a given species belongs? Consider then 
that for fritillaries, literally *every* taxon name and rank between 
genus and order is in dispute, somewhere (e.g., the genus Speyeria is 
no more). Will every fritillary species page list ALL of these higher 
taxon names and the alternative possible rankings? How will the 
different alternative combinations be presented? For example, just 
taking a short segment of the hierarchy, some authorities consider 
fritillaries to be Heliconiines, but there are others who consider 
Heliconiines to be a family, and others who consider Argynnines 
(typically a tribe, sometimes a subfamily) to be a family. That means 
that fritillaries can be in the subfamilies Nymphalinae, 
Heliconiinae, or Argynninae, in either the families Nymphalidae, 
Heliconiidae, or Argynnidae. That's a lot of possible combinations, 
and pretty confusing to try to present all of them in parallel. 
Multiply this out across all of the different rank levels, and I 
estimate that one can come up with at least 60 different alternative 
hierarchies for each species, going up to order. Do we honestly 
expect anyone other than taxonomists to care that there are 60 
different hierarchies possible for the Great Spangled Fritillary 
Argynnis cybele (30 that place it in the genus Speyeria, as most 
people are familiar with, and 30 that place it in Argynnis)?

Yes, I see that the EoL can deal with this, in part, by the use of a 
"novice view" interface versus an "expert view" interface, but I must 
stress that it is indeed only in part.

Further, what happens given that most fritillaries have at least 2 or 
3 subspecies? Does the EoL give a separate page for each subspecies? 
If only full species are given pages, what then happens when there 
are disputes over whether a given subspecies is actually a full 
species? If each subspecies gets a separate page, what happens when 
there are photos, or life history descriptions, or scientific papers, 
that cannot be definitively attributed to one subspecific taxon 
versus another? There is hardly a butterfly species, Carabus, or 
tiger beetle in the world to which these questions will not apply (I 
think the average tiger beetle taxon has ~10 subspecies, for 
example). There is nothing cut and dry about this, unless the choice 
is made to go with a single classification. The approach of having a 
"novice view" interface and an "expert view" interface does not 
resolve the matter of which taxon gets a page.

It's efforts like the EoL that are among the reasons I and others 
have been advocating a formal and mandatory system of taxon name 
registration - and even among this group, I am among a minority 
arguing that we need a single, authoritative classification. There's 
no reason we can't allow names or ranks to change, when there is 
evidence that such change is necessary - but I see no reason or 
benefit to allowing disputes to go unresolved. What we NEED is a 
mechanism to resolve taxonomic disputes! It's like expecting the 
United Nations to accept delegates from every ethnic or cultural 
group in the world, on the off-chance that some day, some of them 
might become sovereign nations, or arbitrarily declare themselves as 
such. Without an authoritative structure, projects such as EoL are 
likely to get bogged down by bickering.

In another vein, Wolfgang Lorenz wrote:

>exclude such species (at least in the first approach) where
>there is nothing but a single publication on type material and the page author
cannot add anything new.

That would exclude at least 1.2 million of the 1.8 million planned 
pages; most species are known only from the original publication and 
subsequent catalogues, keys, and checklists (based ON the original 
pub). That might make the project a lot more practical, but it would 
certainly violate the underlying premise of the project. Heck, even 
E.O. Wilson, whose dreams for the future of taxonomy are embodied in 
the EoL, will be forced to admit that of the 12,000 or so known ant 
species, there's no information other than that given in the original 
description for the majority. Then consider how the species pages 
might look for ant species in the EoL. If one takes Wolfgang's 
concept of "the minimum requirements an acceptable species page must 
meet" then is it fair to assume that will also mean digitized photos 
or illustrations of the taxon? Descriptions of both sexes? Then ask 
(with respect to ants) how many of those 12K species actually have 
copyright-free images available to be digitized? How many of those 
12K species have published descriptions of the males or queens? 
Ultimately, I doubt that there are even 200 ant species for which 
there is presently enough material to put together anything 
approaching a complete species page for the EoL, assuming that 
complete means anything like what it SHOULD mean (i.e., content in 
all of the fields listed in the page template shown on the EoL demo 
pages). Then consider taxa like the superfamily Ichneumonoidea, with 
over 80,000 species in 2 families (ichneumon and braconid wasps), 
maybe about 80 of which are well-known enough to have complete pages 
compiled.

Sure, the idea is wonderful, the goals lofty, and the list of 
supporters and affiliates impressive. But if, five years from now, 
only 5% of the species pages in the EoL are non-butterfly arthropods, 
then I won't be all that impressed. Is it reasonable to expect that 
you are going to be able to "mine" enough original literature and 
automate the process enough to build pages for all the taxa for which 
there are no participating authorities? And if you mine and automate, 
who is going to check to make sure that the mined data are not in 
error? It's inspiring to hear people talk about biodiversity and "all 
life on earth", but I find it hard to get excited if the reality is 
that the taxa that will actually get pages are only about 10% of 
global biodiversity, because there aren't actual human beings who can 
*write* or even *proofread* entries for all the obscure taxa that 
comprise the other 90%. Or is the idea that 90% of the pages in the 
EoL *will* be generated automatically, without any human input? I 
don't see how else it can succeed in the allotted time frame, given 
that it does not supply funding. After all, just looking at the page 
template, I find it hard to imagine that voluntary efforts of 
taxonomists in their spare time can produce more than one or two 
completed species pages a week on average - if we assume that a 
contributor can add 100 species a year, then, we're talking about 
1000 species per taxonomist over 10 years, which means 1800 
taxonomists to get the work done - if there is ZERO overlap between 
their spheres of interest. I really don't think you're going to find 
that many non-overlapping volunteers. You might have 1800 people 
volunteering to work on butterflies and macro-moths, for example, but 
once those 15,000 species are done, that's the end of those 
volunteers' contributions - with some 900,000 more insects left to 
do, and some 150,000 non-insect arthropods.

Ultimately, then, I hope that the EoL can attract enough money to 
start paying for content delivery. Otherwise, it's going to rapidly 
asymptote; the 60K vertebrate species will be done quickly, and about 
15K worth of Lepidoptera won't be too hard, and the 110K flowering 
plants will take much longer, together totalling about 200K species - 
maybe molluscs would also get done, bringing it to 300K. But if you 
don't have specific funds set aside for taxa outside of vertebrates, 
plants, molluscs, and butterflies, I can't imagine who is actually 
going to build all the remaining 1.5 million pages for you. Can you 
honestly say you have significant promises of labor (significant 
meaning 1000 or more species) from expert volunteers outside of those 
areas? How many members of this list plan to contribute 1000 or more 
pages that are NOT vertebrates, plants, molluscs, or butterflies?

Sincerely,
-- 

Doug Yanega        /Dept. of Entomology         /Entomology Research Museum
Univ. of California - Riverside, Riverside, CA 92521-0314
phone: (951) 827-4315 (standard disclaimer: opinions are mine, not UCR's)
    Skype: Dyanega               http://cache.ucr.edu/~heraty/yanega.html
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82


More information about the Taxacom mailing list