[Taxacom] General call to collaboration

Sat Feb 13 01:26:09 CST 2021

Thanks George,

I think it is a good idea that the subject be discussed calmly, but I
wonder if it is possible now. This topic is not really new, similar themes
pop up every other year, and the disagreements are never resolved.

As others have said, the problem is nuanced. Carlos is focused purely on
the Code-compliance of Sharkey´s and Mieirotto´s papers, but in my limited
understading of the Code, they are compliant, or at least more so than
clear cases of vandalism like Hoser or Pic, and if they are acceptable then
so are theirs. The idea the  Code is the ultimate arbiter of scientific
quality crops frequently (photos as types, Hoser, names and conservation,
etc) enough that you´d think that by now we would have come to grips with
the fact that it is just a set of vaguish rules to standardize nomenclature
and provide a minimum of rigour to descriptions. Since most of the
Taxacomers think the same, I think the subject is mostly settled, though
the topic of what constitutes a type needs updating.

A separate and distinct point is the practical and philosophical
perspective of the papers, in particular if the descriptions are
sufficiently “diagnostic” to be useful, to which I would argue, not
quite.There is no doubt that barcoding is here to stay and that it is an
important tool, but the point that Sharkey et al are making (while not
being true to their message) is that it be the only tool. They make this
point quite clearly when they claim that “In light of the evidence
presented above, and much more like it in the literature, it seems absurd
to us that manyin the taxasphere complain about barcode-based species
concepts, yet readily accept delineations based solely on morphology.”It is
this claim that I take issue with and their article has convinced me
further that their approach is not really what it claims to be.

In my mind a barcode (BC) is just another line of evidence in support of a
species´ hypothesis and it needs to be contrasted within the context of
previous hypotheses. Since the vast majority of descriptions are based on
morphology, it is necessary for the description to be sufficient to allow
others to compare with previous ones as well as with future data. This
doesn´t mean that a description must be exhaustive or written, but it
should provide all the available data to allow the comparison to be made.
Morphological descriptions need not be word descriptions but they need to
at least provide sufficient images of high enough quality for the
comparison to be made. In this regard both papers fall short.

I am not a braconid expert but a quick examination of papers shows that
detailed images of the head(frontal), thorax (side and dorsal), abdomen
(side and dorsal), wings and back legs are the least you need (genitalia
don´t seem to feature in the braconid papers I have looked at but I would
say that they are the most important character in most insect species in
general). Mieirotto et al´s images are both poor and insufficient and
though Sharkey et al´s (2021) first two figures were a big improvement,
alas it was not followed in the rest of the paper. If we are doing away
with worded descriptions (or at least limited to diagnostic features) in
order to stream descriptions, then the onus should be on the images.
Considering the number of authors and the ease with which imaging can be
done nowadays, I can´t understand why they did such a poor job.

Another criticism regarding their “barcode-defined speciesconcept” is that
it is nothing of the sort, but rather a post hoc construct that ejects
phenotype in the final stage and presents the BC as the complete picture.
When they compare the accuracy of BCs against morphology in Alabagrus
(suggesting accuracy of 60% for the latter) it is only because they used
other phenotypic data (host use) to boost the refine the BINs. What is the
accuracy of morphology combined with ecology? We don´t know, but if
morphology and host use had been used in conjunction, it is likely that the
accuracy of the former would have also been significantly boosted and it
only demonstrates that descriptions work best when all data is combined.

They seem to acknowledge as much since all their examples are a combination
of phenotypic and genetic data, yet at the end they claim that “As a final
illustration of how ineffective morphology is to differentiate braconid
species, we ... tested their ability to differentiate the species of
Lytopylus before including COI and hostdata. They had error rates of 62%
and 54%, respectively. ... In contrast, the molecular species concepts
matched with the final species delimitations, which were also based on host
data, at 96.6%.”This simply demonstrates that the best approach is a
combination of all data, including ecology, in descriptions, not that
barcode based descriptions are the way to go. Furthermore, they also fail
to consider that feedback between BCs and morphology is a two-way street.

In the end the problem I have with the paper is that it is purporting to be
something that it isn´t and sending a dangerous and inaccurate message that
BCs are all that you need, in spite of their insistence that “...we are not
proposing that comprehensive revisions that include keys and morphological
diagnoses be abandoned.” Considering the high cost of museum storage,
taxonomist training, etc, and that “...old museum specimens have little
value, because of the enormous cost in time and budgets to examine and
identify them morphologically,…” one has to wonder if they are not buying
the rope or at least chipping in.

Finally, telling other taxonomists that “...we view barcode-based
descriptions as a first pass in an iterative approach to solve the
taxonomic impediment…, if a taxonomist wishes to integrate these elements
to Zelomorpha or any of the taxa that we review below they will have a
great starting point (and resources).” fails to understand that taxonomists
are people and that though they largely work for their own applause, they
are still human.

Considering that the cost of high quality imaging has plunged, that digital
media provides limitless possibilities for incorporating data in the form
of images and video to descriptions, it is strange that the debate should
be based on straw men instead of finding integrative approaches that work
into the future but at the same way are rooted in the past. Disruption
shouldn´t mean burning the library.

Jason

On Sat, 13 Feb 2021 at 05:07, George Sangster via Taxacom <
taxacom at mailman.nhm.ku.edu> wrote:
Two days ago, Carlos wrote:
> Deutsche Entomologische Zeitschriftand and ZooKeys. This lack of logic is
> restricted to Pensoft journals.
> I wonder who could profit the most from a pipeline like this, overriding
> all need to diagnose the new taxa from the previous taxa:
> We have all the elements to streamline an output of thousands of new names
> per month: 1) Environmental samples. 2) Metabarcoding. 3) Machines
> assembling the data as data papers and assigning names as arbitrary
> combinations of letters. 4) Machines being the name authors, so that there
> is no one to blame. 5) Compact, inexpensive storage of type specimens, now
> in Eppendorfs (molecules are "parts of animals"). It is a Code-compliant
> and technologically possible scenario.
> Well, definitely for-profit publishers have the most to win from a steady
> input of hundreds and even thousands of pages per month. I don't think
that
> stopping illogical novelty claims is among their interests.

I have two comments and a suggestion.

First, the editorial process at ZooKeys, including the decision to
accept or reject a manuscript, is carried out by the subject editors.
The decision to publish a manuscript is based on scientific grounds, not
impact factors or (the author's ability to pay) page fees. The editorial
process is entirely independent from the commercial aspects of
publishing the journal. I suppose you agree with me that this is how it
should be.

Second, you might be interested to read the Criteria for publication
of new species descriptions in ZooKeys*.*
https://zookeys.pensoft.net/about#TaxonomicTreatments

It was published in January and was the result of multiple rounds of
consultation among the subject editors of ZooKeys. It was initiated by
the publisher/owner.

I'd be very interested to learn if we have overlooked anything important.

Instead of going to war, and battling (seemingly) different opinions,
perhaps we as a group can work towards a universal set of
recommendations ("best practices") for species descriptions. For
instance, while publishing a new species with an illustration of a live
animal as its type specimen (i.e. without a vouchered type specimen) can
be acceptable under the ICZN, it is not generally a good idea to do so.
So let's specify what other things a taxonomist must do to make such a
description acceptable for us. Same with DNA diagnoses. These are
acceptable (i.e. not explicitly unacceptable) under the ICZN, but they
are far from ideal, so what hoops should a taxonomist jump through to
make such a description acceptable? This might turn into a nice
multi-author paper, useful for authors, reviewers, journal editors and
teachers. In medicine, it is common to have Standard Operating
Procedures, and some are adopted by 100s of journals (e.g.
http://www.icmje.org/recommendations/). I suggest we try this approach
in taxonomy. (I am aware of Judith Winston's 1999 book.)

Have a nice weekend.

All the best,
George

Dr George Sangster
Subject Editor Aves, Zookeys
Naturalis Biodiversity Center
Leiden, the Netherlands

_______________________________________________
Taxacom Mailing List

Send Taxacom mailing list submissions to: taxacom at mailman.nhm.ku.edu
For list information; to subscribe or unsubscribe, visit:
http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
You can reach the person managing the list at:
taxacom-owner at mailman.nhm.ku.edu
The Taxacom email archive back to 1992 can be searched at:
http://taxacom.markmail.org

Nurturing nuance while assaulting ambiguity for about 34 years, 1987-2021.

On Sat, 13 Feb 2021 at 05:07, George Sangster via Taxacom <
taxacom at mailman.nhm.ku.edu> wrote:

> Two days ago, Carlos wrote:
> > Deutsche Entomologische Zeitschriftand and ZooKeys. This lack of logic is
> > restricted to Pensoft journals.
> > I wonder who could profit the most from a pipeline like this, overriding
> > all need to diagnose the new taxa from the previous taxa:
> > We have all the elements to streamline an output of thousands of new
> names
> > per month: 1) Environmental samples. 2) Metabarcoding. 3) Machines
> > assembling the data as data papers and assigning names as arbitrary
> > combinations of letters. 4) Machines being the name authors, so that
> there
> > is no one to blame. 5) Compact, inexpensive storage of type specimens,
> now
> > in Eppendorfs (molecules are "parts of animals"). It is a Code-compliant
> > and technologically possible scenario.
> > Well, definitely for-profit publishers have the most to win from a steady
> > input of hundreds and even thousands of pages per month. I don't think
> that
> > stopping illogical novelty claims is among their interests.
>
> I have two comments and a suggestion.
>
> First, the editorial process at ZooKeys, including the decision to
> accept or reject a manuscript, is carried out by the subject editors.
> The decision to publish a manuscript is based on scientific grounds, not
> impact factors or (the author's ability to pay) page fees. The editorial
> process is entirely independent from the commercial aspects of
> publishing the journal. I suppose you agree with me that this is how it
> should be.
>
> Second, you might be interested to read the Criteria for publication
> of new species descriptions in ZooKeys*.*
> https://zookeys.pensoft.net/about#TaxonomicTreatments
>
> It was published in January and was the result of multiple rounds of
> consultation among the subject editors of ZooKeys. It was initiated by
> the publisher/owner.
>
> I'd be very interested to learn if we have overlooked anything important.
>
> Instead of going to war, and battling (seemingly) different opinions,
> perhaps we as a group can work towards a universal set of
> recommendations ("best practices") for species descriptions. For
> instance, while publishing a new species with an illustration of a live
> animal as its type specimen (i.e. without a vouchered type specimen) can
> be acceptable under the ICZN, it is not generally a good idea to do so.
> So let's specify what other things a taxonomist must do to make such a
> description acceptable for us. Same with DNA diagnoses. These are
> acceptable (i.e. not explicitly unacceptable) under the ICZN, but they
> are far from ideal, so what hoops should a taxonomist jump through to
> make such a description acceptable? This might turn into a nice
> multi-author paper, useful for authors, reviewers, journal editors and
> teachers. In medicine, it is common to have Standard Operating
> Procedures, and some are adopted by 100s of journals (e.g.
> http://www.icmje.org/recommendations/). I suggest we try this approach
> in taxonomy. (I am aware of Judith Winston's 1999 book.)
>
> Have a nice weekend.
>
> All the best,
> George
>
> Dr George Sangster
> Subject Editor Aves, Zookeys
> Naturalis Biodiversity Center
> Leiden, the Netherlands
>
> _______________________________________________
> Taxacom Mailing List
>
> Send Taxacom mailing list submissions to: taxacom at mailman.nhm.ku.edu
> For list information; to subscribe or unsubscribe, visit:
> http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
> You can reach the person managing the list at:
> taxacom-owner at mailman.nhm.ku.edu
> The Taxacom email archive back to 1992 can be searched at:
> http://taxacom.markmail.org
>
> Nurturing nuance while assaulting ambiguity for about 34 years, 1987-2021.
>