[Taxacom] taxonomic names databases

Tony Rees tonyrees49 at gmail.com
Thu Sep 1 19:58:26 CDT 2016

Hi Nico,

I do see where you are coming from, it's just one point along a spectrum
where I sit closer to the other end for pragmatic reasons ( :) ). Again,
following the example of (let us say, Eschmeyer's Catalog of Fishes or
similar), while generally accepting his work as a "point of truth" if you
like, that would not stop me from modifying a record obtained from there
for my own system if I believed it was necessary (citing appropriate
alternative sources etc.).

Look, one can make (and we are making perhaps) a separate argument for a
system which natively incorporates either a single, or a variety of
alternative taxonomic viewpoints (the former is obviously easier to
implement than the latter). Perhaps the latter more closely reflects the
practices of active taxonomists, however there is also (I would contend) a
clear desire for the latter - e.g. a "consensus classification", even
though this may change through time and its exact form be argued about -
for a whole class of users of taxonomic data (clients of the system) who
need e.g. a management hierarchy by which to organise their information,
preferably also one that will be taken up by others (such as let us say,
APGIII for angiosperms). OK, APGIII may not be perfect - and has recently
been superseded by APGIV anyway - but at least you say if your [extant]
angiosperm data are arranged according to APGIII, others will know what you
are talking about.

>From the "about WoRMS" page you cited:

The classification used is a ‘compromise’ between established systems and
recent changes. Its aim is to aid data management, rather than suggest any
taxonomic or phylogenetic opinion on species relationships.

Perhaps this is selling the system a little short - in my experience the
various sector editors do try to incorporate recent changes, at least where
these seem to be evidence-based - but you will see a tacit acknowledgement
here of the practical value of a single management hierarchy here, that
many users appreciate.

Also note that the individual sub-compilations within WoRMS are all
individually citable (with appropriate authorship), it is merely for the
entire collection that authorship is ascribed to the "WoRMS Editorial

Best - Tony

Tony Rees, New South Wales, Australia

On 2 September 2016 at 09:58, Nico Franz <nico.franz at asu.edu> wrote:

> Thank you, Tony.
>    I do think that I could have spoken more clearly, but also think that
> we look at things a bit differently here. In building bigger and bigger
> "backbones" (which go all the way to the species-level tips, right?), I
> think necessarily the lines between author and aggregator get blurred. But,
> design can model the distinction, and the lack thereof.
>    On the author-to-aggregator spectrum, Eschmeyer (
> http://www.calacademy.org/scientists/projects/catalog-of-fishes) is
> evidently more on the author side of the spectrum than others "sources" of
> similar or even smaller scope. I assume though that Dr. Eschmeyer (sorry
> for this appearing to get personal - absolutely not intended but I believe
> needed to make the case), might not personally claim equally profound
> expertise regarding the systematics of all fish lineages, in the sense of
> all lineages equally being part of his active revisionary fish systematics
> research program, so to speak. Whenever we speak of biodiversity, largely
> reliably, we do draw on past and current expertise that is in effect partly
> borrowed (from past authors) and distributed (in various sources). Blurry.
> But it does matter immensely, I believe, that Dr. Eschmeyer is a person,
> with a personal and internationally valued reputation, a personal record of
> commitment to "his" domain. Someone that one can disagree with, combing
> through the Eschmeyer catalog, and presumably a signal will come back,
> either reconciliatory or resisting. Those features, to me, are features of
> authorship.
>    IRMNG is somewhere on that spectrum, to be sure, and likely not so far
> from Eschmeyer (as evidenced by your objection). Though note the less
> personalized citation practice: http://www.marinespecies.org/about.php
>    Poor design, to my mind, is the kind of design where - qua aggregation
> - a sense of authorship is weakened, obscured, to the point of "this
> backbone just is". I believe that is a design that necessarily must result
> in lowered trust, because I believe at a fundamental level that good
> taxonomy has an individual expert-driven "flavor". You might call it
> subjective..I think the design has to honor the notion that expertise is
> personalized.
>    The reason why I would not defer to Dr. Eschmeyer's expertise by fiat
> is that in certain cases (fish experts please help me or let me dangle in
> eternal shame and agony), I may well think that he is mistaken in his
> personally researched or editorially chosen preferred classificatory
> representation. Maybe I disagree with his filtering of the latest
> phylogenetic inferences into the catalog.
>    And I disagree on the "they won't appreciate much" issue too - again,
> to me that points to design. For any given group, the aggregating
> environment can in principle store multiple conflicting views, and flag
> these as such. That takes nothing away from an author's unique contribution
> or motivation, it just means designing for multiple views and offering
> choices in cases of conflict. That is how taxonomy operates throughout the
> entire "primary" literature (hard-to-define term, see above), except
> apparently in the aggregation domain where conflict and persistent
> disagreement tends to get designed away (https://bmcbioinformatics.
> biomedcentral.com/articles/10.1186/1471-2105-6-208).
>    Is it likely much harder to build consistently expert-identifying,
> conflict-embracing, but also scalable systems? Of course. But that does not
>  make the decision not do so any less of a choice (pragmatic,
> understandable), and one that has trust-related consequences.
> Best, Nico

More information about the Taxacom mailing list