search

plazi / arcadia-project

2 stars 1 forks source link

treatment metadata: zookeys example: error in scientific name authorship, link to article deposit in BLR #159

Open myrmoteras opened 3 years ago

myrmoteras commented 3 years ago

@teodorgeorgiev I just looked at this treatment https://doi.org/10.5281/zenodo.4056468 in this article https://zenodo.org/record/3555661 with 136 treatments

  1. The article deposit in BLR does not list the treatments in the metadata
  2. The article BLR deposit is not listed in the treatment deposits
  3. The figure link in the article metadata does not resolve to image in Pensoft, but from there to the article XML
  4. The Scientific name authorship in the custom keywords is wrong. Ashmead should be Arias-Penna,
teodorgeorgiev commented 3 years ago

Hi Donat,

See my reply inline:

On 14.12.2020 г. 10:47, Donat Agosti wrote:

@teodorgeorgiev https://github.com/teodorgeorgiev I just looked at this treatment https://doi.org/10.5281/zenodo.4056468 https://doi.org/10.5281/zenodo.4056468 in this article https://zenodo.org/record/3555661 https://zenodo.org/record/3555661 with 136 treatments

  1. The article deposit in BLR does not list the treatments in the metadata

that is correct, the treatments are deposited via GGI, and it usually happens after the article has been deposited (directly from us)

  1. The article BLR deposit is not listed in the treatment deposits

same reason, see above

  1. The figure link in the article metadata does not resolve to image in Pensoft, but from there to the article XML

not sure for which links you are talking about.

I see only these, but they works

Related identifiers: Has part

  * 10.3897/zookeys.890.35786.figure173
    <https://doi.org/10.3897/zookeys.890.35786.figure173>
  * 10.3897/zookeys.890.35786.figure1
    <https://doi.org/10.3897/zookeys.890.35786.figure1>
  * 10.3897/zookeys.890.35786.figure5
    <https://doi.org/10.3897/zookeys.890.35786.figure5>
  1. The Scientific name authorship in the custom keywords is wrong. Ashmead should be Arias-Penna,

Taxonomic concept label is generate by GGI.

In the source TaxPub XML the taxon-authority for "gunnarbrehmi" is:                 Arias-Penna,  sp. nov.</tp:taxon-authority> - which is not enterally correct, having the status in it, but still no idea how "Ashmead" happened.

All the best,

Teodor

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/plazi/arcadia-project/issues/159, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABCUA5Y75BWJDA67MMWOXXTSUXGLBANCNFSM4U2MO4KQ.

--

Teodor Georgiev Chief Technology Officer Pensoft Publishers https://www.pensoft.net/ ARPHA Journal Publishing Platform https://arphahub.com/ 12 Prof. Georgi Zlatarski Street 1700 Sofia, Bulgaria Tel +359-2-8704281 Fax +359-2-8704282

Publishing services for journals: https://arphahub.com/about/services https://arphahub.com/about/services Books published by Pensoft: https://pensoft.net/books https://pensoft.net/books Services for scientific projects: https://pensoft.net/projects https://pensoft.net/projects Find us on: Facebook https://www.facebook.com/Pensoft/, Twitter https://twitter.com/Pensoft

gsautter commented 3 years ago

Just to clarify ... it's not GGI that uploads treatments and generates taxon concept labels ... that's up to our server, specifically the Zenodo uploader component.

Regarding the figures, the aforementioned uploader component only loops through what it fiends in the documents, and on TaxPub import, this is the exact URL found in the source.

Regarding the taxon authority, I think the taxon status being included in the authority (end hence the name proper after import) is triggering a cascading effect: not recognizing the status as such (for which it would have to be marked separately) has the TaxPub import check GBIF and CoL for an authority (mainly in search for a year), and apparently came back with "Ashmead" (who is, far as I read the article, the author of the genus).

The TaxPub import relies upon treatment taxa being marked correctly, including their status.

CoL returning an HTML page on what used to be API calls isn't helping, either ... not sure since when this has been broken or changed.