[Taxacom] Digitizing Science Collections

Thu May 17 00:05:51 CDT 2012

Last comments for today (1.00 am)

On Thu, May 17, 2012 at 12:51 AM, Stephen Thorpe <stephen_thorpe at yahoo.co.nz
> wrote:

> but I am not using my opinion to make a definite conclusion (that the
> record is bogus), for that, as you rightly point out, would not be
> scientific. Rather I am using my opinion to highlight that there could well be
> a problem with the record. That seems to me to be a perfectly good use
> of an "opinion"! Opinions can be very useful, for the right purposes...
>

ok - to say that there could be a problem is different from saying that
there was a problem with the record which is what I erroneously thought you
were saying.

>
> Consider a medical analogy, diagnoses can seldom be straightforwardly
> derived from symptoms by objective protocols. That is why people prefer to
> go to experienced doctors, i.e., because their opinions are more likely to
> be correct, all other things being equal (which they never are, but that's
> not the point!)
>

Sure, but in this case I had the incorrect impression that you had a basis
for the record being considered false.

John Grehan

>
> Stephen
>
>   *From:* John Grehan <calabar.john at gmail.com>
> *To:* Stephen Thorpe <stephen_thorpe at yahoo.co.nz>; taxacom <
> taxacom at mailman.nhm.ku.edu>
> *Sent:* Thursday, 17 May 2012 4:43 PM
>
> *Subject:* Re: [Taxacom] Digitizing Science Collections
>
> More comments below.
>
> On Thu, May 17, 2012 at 12:25 AM, Stephen Thorpe <
> stephen_thorpe at yahoo.co.nz> wrote:
>
>  John, perhaps you are trying to be disagreeable, perhaps not, but any
> rate:
>
> the situation here is entirely analogous to the Wikipedia Munididae/WoRMS
> issue. Sometimes, there is no absolute objective proof that something is
> incorrect, but by stating it without comment, one is implicitly endorsing
> it as reliable (unless you explicitly state otherwise, but then that is
> not without comment!) In these cases, you must either flag it as dodgy, or
> leave it out altogether.
>
>
> Sure, if one has reason to do so. No problem.
>
>
> in the present case, the Australian author may not have known enough about
> N.Z. or the degree to which the N.Z. has been sampled for tenebrionids to
> see a problem at all. If so, then I am not blaming him for anything.
>
>
> ok
>
>
>  Nonetheless, it would have been better for that record to carry a red
> flag or be left out.
>
>
> Only if the author had reason to do so as you said above.
>
>
> Scientists are not robots, processing data as input and output. They have
> real human intelligence and insight, which must not of course be permitted
> to be entirely unrestrained, but rather used with discretion. Given what I
> know about N.Z. tenebrionids, and even what has been published, and other
> groups in N.Z., it is in one of those wonderfully insightful human
> intelligence ways overwhelmingly obvious that the Whangarei record is a
> mislabelling.
>
>
> Well, perhaps for you, but not necessarily the author.
>
>
>
>  Actually, it is based quite heavily just on hard empirical data that
> such disjunctions just don't happen here for big flightless beetles like
> this one.
>
>
> Well of course if the record were true then your statement would not be
> correct.
>
>
>  There is a REMOTE possibility that this is the first, but science isn't
> about giving equal weight to all possibilities, no matter how remote!
>
>
> But something is not remote just because one says so.
>
>
>  If the author knew this, which he may well not have, then IMHO it was
> his duty to make some comment to that effect.
>
>
> Since you do not know either way this is not informative.
>
>
>  Crap data is like DDT in the ecosystem, it tends to accumulate in nasty
> places, so we must always try to purge ourselves of it at the first
> opportunity ...
>
>
> In this case such status appears to be a matter of opinion rather than
> empirical fact.
>
> John Grehan
>
>
> Stephen
>
>   *From:* John Grehan <calabar.john at gmail.com>
> *To:* Stephen Thorpe <stephen_thorpe at yahoo.co.nz>
> *Cc:* "TAXACOM at mailman.nhm.ku.edu" <TAXACOM at mailman.nhm.ku.edu>
> *Sent:* Thursday, 17 May 2012 4:02 PM
>
> *Subject:* Re: [Taxacom] Digitizing Science Collections
>
> Comments inserted below.
>
> On Wed, May 16, 2012 at 10:34 PM, Stephen Thorpe <
> stephen_thorpe at yahoo.co.nz> wrote:
>
>  John,
>
> This issue is quite important, at least "academically", so I will once
> again try to clarify my views, relating them to this specific example, and
> your comments below. Nobody can prove the negative, even if the species
> doesn't occur in Whangarei now,
>
>
> How does one know that the species does not occur in Whangarei now?
>
>
>  and left no trace, it might have done so when that single specimen was
> collected. It is a REMOTE possibility.
>
>
> Why is the possibility remote?
>
>
>  But that is surely not how science works? If you indicate something
> without comment, such as the presence of this species in that surprising
> place, then it should be more than just a remote possibility.
>
>
> But what is surprising about the place, what makes it remote as
> a possibility?
>
>
>  If it is just a remote possibility, then you should say so.
>
>
> Then that would be true of any locality, but one does not see taxonomists
> making such comments about every locality for every taxon in each
> publication.
>
>
>  This may require appeal to all sorts of evidence, including experience
> and local knowledge *of the group*. Scientists should not ignore good
> evidence!
>
>
> So what is the "good evidence" that the author ignored?
>
>
> It was actually my point and not yours that if it were some other kind of
> organism, like a mite for example, then the Whangarei record could well be
> genuine, and the apparent disjunction due to uneven and low sampling
> effort.
>
>
> ok. Good for you.
>
>
>  But, given what I know about actual sampling effort of N.Z.
> tenebrionids, this is overwhelmingly unlikely in this case. There is a
> remote possibility that the disjunction could be real, but there are so few
> other N.Z. examples of this, that it is also highly unlikely.
>
>
> What is your empirical evidence for this being unlikely?
>
>
> At any rate, I am not suggesting that the author ought to have stated with
> certainty that it is a bogus record,
>
>
> or even mentioned the possibility as that possibility applies to all
> records.
>
>
>  they should just have either left it out until confirmed by further
> material, or at least made some comment to the effect that it is an outlier
> and may be bogus.
>
>
> Again, what empirical evidence is there that it is an 'outlier' any more
> than any other record, or why should the author make any reference to
> the possibility of it being a bogus record?
>
>
>  Otherwise, the reader may just accept it without a second thought ...
>
>
> Well, then that is true of all published records. But in practice all
> records are just hypotheses of location at a time and place, and all may be
> in question. In the end for practical purposes, we are given a set of
> records and it is what we make of them rather than what may not exist that
> seems to matter for practical purposes.
>
> John Grehan
>
>
> Cheers,
>
> Stephen
>
>   *From:* John Grehan <calabar.john at gmail.com>
> *To:* Stephen Thorpe <stephen_thorpe at yahoo.co.nz>
> *Cc:* "TAXACOM at mailman.nhm.ku.edu" <TAXACOM at mailman.nhm.ku.edu>
> *Sent:* Thursday, 17 May 2012 1:54 PM
>
> *Subject:* Re: [Taxacom] Digitizing Science Collections
>
> Stephen,
>
> I take it that it was obvious to you that there was mislabeling because
> you have proved that this species was not present at the locality or area
> indicated?
>
> And if that is the case it does not preclude what I have said, that the
> distributional relationship is found in other taxa and so not beyond a
> reasonable possibility.
>
> Of course ultimately all records are subject to corroboration or
> refutation, but if we were to take that approach with everything nothing
> would be accepted unless we individually checked the location ourselves for
> every living entity on this planet.
>
> As for ignoring local knowledge and experience you will need to provide
> specifics for that to be evaluated.
>
> John Grehan
>
>
> On Wed, May 16, 2012 at 6:30 PM, Stephen Thorpe <
> stephen_thorpe at yahoo.co.nz> wrote:
>
>  I have to disagree John! I was saying that the mislabelling is obvious
> to someone who has local knowledge *of the group* (Tenebrionidae) in N.Z.,
> but not obvious to others. You have just proved me correct on the last
> point. "Objectivity" does *not* mean ignoring local knowledge and
> experience with a taxon...
>
> Stephen
>
>   *From:* John Grehan <calabar.john at gmail.com>
> *To:* Stephen Thorpe <stephen_thorpe at yahoo.co.nz>
> *Cc:* "TAXACOM at mailman.nhm.ku.edu" <TAXACOM at mailman.nhm.ku.edu>
> *Sent:* Thursday, 17 May 2012 3:07 AM
>
> *Subject:* Re: [Taxacom] Digitizing Science Collections
>
> While geographically distant disjunctions certainly attract interest,
> accepting them at face value may be no more uncritical than for any record.
> In this case there appears to be no information to make any judgement and I
> do not see the need for the author to make any necessary author. Further
> corroboration might be desirable if one wan't to question the record. There
> is the case of Metrosideros parkinsoni on offshore islands near Whangarei
> and also in NW South Island, or the Sticherus and Asplenium in the northern
> North Island and South Island, or the moth Sabantica calliarcha in the
> Coromandel Peninsula (single location) and South Island. If anything, the
> author might be critiqued for not recognizing the pattern in the work, but
> now the question might be, how common is the occurrence and what is its
> biogeographic significance?
>
> John Grehan
>
> On Wed, May 16, 2012 at 12:48 AM, Stephen Thorpe <
> stephen_thorpe at yahoo.co.nz> wrote:
>
>     note that I am primarily arguing against the *uncritical* use of
> label data (which happens), and suggesting that in cases of doubt, the only
> way to settle it is to go out into the field and verify presence/absence
>
> for an example of how not to do it, see p. 11 in:
> Matthews, E.G. 2003: Ulomotypus Broun a member of the new subfamily
> Palorinae, with remarks on Aphtora Bates and Demtrius Broun (Coleoptera,
> Tenebrionidae). New Zealand entomologist, 26: 7?14.
>
> http://www.ento.org.nz/nzentomologist/abstract.php?volume_issue=j26_1&first_page=7
>
> not really the fault of the Australian author, but the peer review
> editorial process should have picked it up. Yes, a single specimen of
> Ulomotypus laevigatus (a fairly large flightless beetle) is labelled as
> being from Whangarei (north of the North Island), while all others are from
> the South Island, but to simply list it without comment is hardly helpful!
> For some other taxa in Hymenoptera or Diptera, for example, such a
> distribution pattern could be a genuine consequence of small sample size
> (so it is not "objectively" a red flag), but this is most unlikely in the
> present case, and simple mislabelling will be to blame. An N.Z.
> entomologist might be able to see this as an obvious case of mislabelling,
> but not so others, so at least some sort of comment was warranted. This is
> a case of "too much objectivity", instead of utilising genuine local
> knowledge and judgement with the fauna...
>
> Stephen
>
>
>   *From:* John Grehan <calabar.john at gmail.com>
> *To:* Stephen Thorpe <stephen_thorpe at yahoo.co.nz>
> *Cc:* "TAXACOM at mailman.nhm.ku.edu" <TAXACOM at mailman.nhm.ku.edu>
> *Sent:* Wednesday, 16 May 2012 4:23 PM
> *Subject:* Re: [Taxacom] Digitizing Science Collections
>
> Going out an personally recording the distribution is not always feasible.
> Hence using label data from specimens is sometimes the only practical
> option. In a current revision of a ghost moth genus in Latin America I do
> not have the $ to collect specimens from the wild and so have to rely on
> specimens collected by others.
>
> John Grehan
>
> On Tue, May 15, 2012 at 6:24 PM, Stephen Thorpe <
> stephen_thorpe at yahoo.co.nz> wrote:
>
> label data is a huge can of worms for all sorts of reasons! After having
> thought the issue through at some length, I have come to the probably
> rather unpopular conclusion that it is a bad idea to think that the
> totality of label data for any given taxon is a useful thing to have! It is
> somewhat analogous to the example of using WoRMS as a source for that squat
> lobster article on WP ... given the way the WoRMS page was at the time, it
> was best to just not mention it at all! There is a certain irony that it is
> common nowadays for scientists to spend money going on collecting tours to
> get fresh material of taxa for use in DNA analysis. Actually, it would be
> better to go on a collecting tour to gather reliable distributional
> information for the taxon! Any distribution map based on the uncritical
> acceptance of label data is likely to be problematic, unles the map is so
> coarse that points cover a significant proportion of the planet! Even worse
> is displaying a
>  distribution map of a tiny subsample of available label data! If you want
> to know the geographical distribution of a taxon, you gotta go out and
> measure it ... carefully!
>
> Stephen
>
>
> ________________________________
> From: David Campbell <pleuronaia at gmail.com>
> To: TAXACOM at mailman.nhm.ku.edu
> Sent: Wednesday, 16 May 2012 10:01 AM
> Subject: Re: [Taxacom] Digitizing Science Collections
>
> > Unfortunately, original data labels are often extremely misleading,
> > especially regarding distributional information. After georeferencing
> > over 200,000 insect specimens from multiple institutions, the figures
> > work out to over 20% of all original labels have significant
> > omissions or ambiguities (the latter category being primarily things
> > like abbreviations, uncertain reference points, uncertain date
> > formats, or handwriting problems), and some 5% of all original labels
> > have genuine errors of one sort or another (mostly typographical
> > errors). I doubt such problems are unique to insect labels.
>
> I can confirm that mollusk labels have similar issues with similarly
> high frequency, though I have not tallied percentages. In fact,
> insects may have better than average quality of average data because
> fewer specimens survive from the "I picked these up somewhere when my
> ship sailed around the world" era.  Having a misidentified specimen
> with good locality data is far more useful than a correctly identified
> specimen with bad locality data, as the identification can be fixed
> from the specimen.  Someone knowledgeable about the taxon or locality
> may be able to recognize that a record is supect, even without
> examining the specimen.  But all data in a database needs competent
> checking to be reliable.  And many larger collections will include
> items from various sources, of varying reliability for their
> localities and identifications.
>
> The problem is compounded when online databases grab information from
> varied sources without careful verification or good harmonization
> protocols.  In particular, homonyms can lead to perfectly good
> information from one database being presented wrongly in another one,
> which claims that corrections should be referred back to the first
> database.
>
> --
> Dr. David Campbell
> Collections Assistant
> The Paleontological Research Institution
> 1259 Trumansburg Road
> Ithaca NY 14850
>
> _______________________________________________
>
> 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) by visiting http://taxacom.markmail.org
>
> (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) by visiting http://taxacom.markmail.org
>
> (2) a Google search specified as:  site:
> mailman.nhm.ku.edu/pipermail/taxacom  your search terms here
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>