Errors = phantom names on the Internet

Tue Mar 28 17:17:44 CST 2006

The problem for me is that deleting the record doesn't always solve  
the problem and may lead to additional confusion.  It is similar to  
the conversations I have had regarding bacterial names that have lost  
their nomenclatural standing.  Deleting them from authoritative  
databases doesn't remove them from existence if they continue to  
exist in content elsewhere.  Indexing these forms is not an  
endorsement of their continued use.  It provides the means to  
annotate the the sort of knowledge that I picked up here today.

When we  (uBio and AMNH library)  indexed all the bacterial names  
that exist in Medline we get all names, the good, the bad, and the  
ugly (including misspellings that only exist in Medline).  We get 60  
years worth of names from across the taxonomic spectrum.  Indexing  
allows me to add an annotation layer that can contain qualifiers.   
There are those forms addressed by the bacterial nomenclators.  Go  
there for more information.  There are those addressed by the BDWD.   
Go over there for more information.   Some forms are not addressed in  
these expert sites, maybe never will be.  Some have been expunged  
from them.

  A negative answer is more useful than a null answer.  If I or  
someone else comes across one of these forms in the medical  
literature or online or in an article, an index provides me with a  
place to hang an answer even if that answer is "immediately put this  
name down, wash your hands, and never refer to it again."  If the  
only place the "phantom record" existed is in our index then I would  
remove it.  Until then, it remains.  If we have to add additional  
attributes to reconcile it then we will.

David Remsen

On Mar 28, 2006, at 4:09 PM, Richard Pyle wrote:

> If you do not maintain *some* record of the phantom name --  
> especially names
> which might have been picked up and used outside of the original  
> dataset in
> which they were erroneously created -- then you are forcing  
> *everyone* in
> the future who encounters the name again to figure out, on their  
> own, what
> the story is with this name.
>
> Issues of making it "clear" that the name is bogus (i.e., forcing  
> someone to
> see right up front that the name is phantom -- not making them  
> scroll down
> page to find this out) is a question of data presentation --  
> altogether
> different from the question posed here (i.e., keep it or delete it).
>
> I see the semantics problems of this conversation already getting  
> thick.
> Does "discard" mean delete from the data management system, or does  
> it mean
> conceal from the main user interface?
>
> If someone has knowledge about a name that has been used outside  
> the context
> of a single database (e.g., Chris Thompson's knowledge about the  
> status of
> "Pollenia pseudobscura"), then it seems silly to me to fail to  
> record that
> knowledge.
>
> Having said that, YES! of course we should focus our limited time and
> energies on sorting out the names that are likely to be meaningful  
> to us.
> But when information is already available to us, we should capture  
> that
> information for the benefit of future generations.
>
> Rich
>
>> -----Original Message-----
>> From: Taxacom Discussion List [mailto:TAXACOM at LISTSERV.NHM.KU.EDU]On
>> Behalf Of Pape, Thomas
>> Sent: Tuesday, March 28, 2006 10:55 AM
>> To: TAXACOM at LISTSERV.NHM.KU.EDU
>> Subject: Errors = phantom names on the Internet
>>
>>
>> Yes, the web is "public", and much information may 'hang on' for
>> considerable time through various online sources. We only need to
>> decide on what information we WANT to deal with.
>>
>> If we decide to keep phantom names like "Pollenia pseudobscura
>> Rognes, 1985", much as we database unavailable names from the
>> scientific literature, would we even want to annotate phantom
>> misspellings? I prefer to keep nomenclatural information that are
>> 1) meant to be permanent, scientific records and 2) have some
>> (potential of) permanency. Chris mentions that the record was not
>> peer reviewed but a working record. I would guess that the
>> Biosystematic Database of World Diptera contains many more
>> phantom names than "Pollenia pseudobscura", and most probably
>> some of these are already finding their way into other online
>> products. I am not sure the 'permanent' registration of such
>> phantom names would serve the taxonomic community - nor the
>> biological sciences at large - any good.
>>
>> At least for Diptera, we are currently seeing what appears to be
>> an exclusively digital and certainly NOT peer reviewed journal
>> proposing several new names.
>>
>> So, my recommendation is to discard these phantom names as errors
>> or working records or dummies, etc. - and concentrate our efforts
>> of designing a mandatory registry system including peer review.
>>
>> Thomas Pape
>> Natural History Museum of Denmark
>>
>> -----Oprindelig meddelelse-----
>> Fra: Taxacom Discussion List [mailto:TAXACOM at LISTSERV.NHM.KU.EDU]
>> På vegne af christian thompson
>> Sendt: 28. marts 2006 20:43
>> Til: TAXACOM at LISTSERV.NHM.KU.EDU
>> Emne: [TAXACOM] Errors = phantom names on the Internet
>>
>> We all make mistakes. And today an old one of mine was discovered
>> by a colleague.
>>
>> I do not know how I did it or whether some one else on my team
>> did it, but we created a phantom name. Pollenia pseudobscura
>> Rognes, 1985. That was done back in the late 1980s when we were
>> building the Nearctic Diptera Database. That dataset was passed
>> along to ITIS via NODC in 1992. From ITIS the name passed to
>> Species2000 and GBIF ECAT, and also to uBio. Today Knut Rognes
>> called our attention to the error.
>>
>> My reaction was just to DELETE the bogus data record. It was a
>> mistake. The name has never existed in the printed literature***.
>> AND our record is clearly marked as a WORKING record, not
>> peer-reviewed, not available for public use, etc.
>>
>> However, my colleague, David Remsen, has said we should annotate
>> the name record and retain this error.
>>
>> So, I wonder what others think? Should we delete database errors?
>> or leave them in with appropriate annotations?
>>
>> When thinking about that, I should note that I did GOOGLE this
>> "phantom" name and got 10 hits. There was 1 for ITIS, a couple of
>> uBio, but most interesting is the name is now in WikkiPedia under
>> Pollenia as well as at the University of Michigan Animal
>> Diversity Web site.
>>
>> *** the BDWD is interested only in indexing SCIENTIFIC
>> nomenclature and since the International Code of Zoological
>> Nomenclature remains largely restricted to PRINTed medium and the
>> Internet is excluded, we do not index the Internet / WWW.
>>
>> F. Christian Thompson
>> Systematic Entomology Lab., USDA
>> c/o Smithsonian Institution
>> MRC-0169 NHB
>> PO Box 37012
>> Washington, DC 20013-7012
>> (202) 382-1800 voice
>> (202) 786-9422 FAX
>> cthompso at sel.barc.usda.gov e-mail
>> www.diptera.org  web site

_______________________________________________
David Remsen
uBio Project Manager
Marine Biological Laboratory
Woods Hole, MA 02543
508-289-7632