Add missing nomenclatural code

[mdoering]

Some sources/sectors are missing a nomenclatural code which makes it difficult for rendering, but more importantly for querying for bad homonyms like accepted duplicate genera within the same code:

https://www.checklistbank.org/dataset/3/duplicates?category=uninomial&codeDifferent=false&limit=500&minSize=2&mode=STRICT&offset=0&rank=genus&sourceDatasetKey=1141&status=accepted

There are 42 sources with missing nomenclatural code:

clb=> select s.subject_dataset_key, count(*) as cnt from name n join sector s on s.dataset_key=n.dataset_key and s.id=n.sector_key where n.dataset_key=3 and n.code is null group by 1 order by 1;
 subject_dataset_key |  cnt  
---------------------+-------
                1010 |  5759
                1011 |  2607
                1026 |  1065
                1031 |   307
                1033 |    28
                1037 |   323
                1042 |   502
                1046 |   118
                1052 |   377
                1054 |   286
                1055 |  1315
                1068 |  2597
                1070 |    86
                1078 |    34
                1080 |    12
                1082 |    76
                1104 |     4
                1113 |  1262
                1118 |     8
                1119 |    23
                1120 |   491
                1138 |   133
                1139 |    18
                1142 |   160
                1143 |    66
                1144 |   305
                1146 |     1
                1161 |   303
                1171 |  1532
                1172 |    33
                1190 |   562
                1201 |    21
                2007 |   673
                2073 |   230
                2141 |   284
               54170 |     1
              124661 |  8184
              125101 |     6
              268676 | 24620
              279229 |     8
              296427 |    33
              298081 |  4016
(42 rows)

With the following 136 sectors:

clb=> select s.subject_dataset_key, s.id, count(*) as cnt from name n join sector s on s.dataset_key=n.dataset_key and s.id=n.sector_key where n.dataset_key=3 and n.code is null group by 1,2 order by 1,2;
 subject_dataset_key |  id  |  cnt  
---------------------+------+-------
                1010 | 1543 |   272
                1010 | 1544 |    11
                1010 | 1545 |     9
                1010 | 1595 |    14
                1010 | 2013 |  5443
                1010 | 2014 |     3
                1010 | 2015 |     3
                1010 | 2016 |     2
                1010 | 2017 |     2
                1011 | 2139 |  2607
                1026 | 2106 |     6
                1026 | 2107 |     4
                1026 | 2108 |     5
                1026 | 2109 |     4
                1026 | 2110 |   160
                1026 | 2111 |     5
                1026 | 2112 |     2
                1026 | 2113 |   387
                1026 | 2114 |    69
                1026 | 2115 |     6
                1026 | 2116 |    11
                1026 | 2117 |    11
                1026 | 2118 |    13
                1026 | 2119 |    77
                1026 | 2120 |     3
                1026 | 2121 |    21
                1026 | 2122 |     2
                1026 | 2123 |     2
                1026 | 2124 |   273
                1026 | 2125 |     4
                1031 | 1761 |   307
                1033 |  584 |     1
                1033 |  613 |    27
                1037 |   56 |    52
                1037 |  295 |   246
                1037 |  296 |    18
                1037 |  297 |     1
                1037 |  298 |     1
                1037 |  299 |     1
                1037 |  300 |     1
                1037 |  301 |     1
                1037 |  302 |     1
                1037 |  303 |     1
                1042 |  260 |   502
                1046 | 1745 |    86
                1046 | 1746 |    32
                1052 | 2137 |   377
                1054 | 2092 |   286
                1055 |  255 |  1103
                1055 |  256 |    61
                1055 |  257 |   123
                1055 |  292 |    26
                1055 |  293 |     2
                1068 |  291 |  2597
                1070 |  730 |    86
                1078 |  736 |    34
                1080 |  733 |    12
                1082 |  735 |    76
                1104 |  521 |     4
                1113 |  472 |  1262
                1118 |  280 |     8
                1119 |  471 |    23
                1120 |  646 |   491
                1138 | 1979 |     8
                1138 | 1980 |     2
                1138 | 1981 |     2
                1138 | 1982 |    73
                1138 | 1983 |     5
                1138 | 1984 |     3
                1138 | 1985 |    12
                1138 | 1986 |     2
                1138 | 1987 |     2
                1138 | 1988 |     2
                1138 | 1989 |     2
                1138 | 1990 |     8
                1138 | 1991 |     3
                1138 | 1992 |     2
                1138 | 1993 |     2
                1138 | 1994 |     2
                1138 | 1995 |     3
                1139 |  738 |    18
                1142 | 2127 |   160
                1143 | 2126 |    66
                1144 | 2089 |   305
                1146 | 1778 |     1
                1161 | 1367 |   303
                1171 | 2091 |  1532
                1172 | 1753 |    33
                1190 |  532 |   562
                1201 |  619 |    21
                2007 |   25 |     2
                2007 |   59 |     3
                2007 |   60 |     5
                2007 |   61 |     2
                2007 |   62 |     8
                2007 |   63 |     4
                2007 |   64 |     2
                2007 |   73 |     2
                2007 |  443 |     2
                2007 |  444 |     2
                2007 |  445 |   292
                2007 |  446 |    21
                2007 |  448 |     6
                2007 |  449 |    16
                2007 |  450 |     4
                2007 |  451 |    88
                2007 |  452 |    60
                2007 |  453 |    31
                2007 |  454 |     8
                2007 |  455 |    12
                2007 |  460 |     9
                2007 |  461 |     8
                2007 |  462 |     5
                2007 |  463 |     2
                2007 |  576 |    36
                2007 |  577 |     1
                2007 |  578 |     2
                2007 |  579 |     2
                2007 |  580 |    38
                2073 |  669 |    30
                2073 |  670 |   182
                2073 |  671 |    18
                2141 |   55 |   284
               54170 | 1842 |     1
              124661 | 2164 |  8182
              124661 | 2167 |     2
              125101 | 1808 |     6
              268676 | 2007 |   221
              268676 | 2008 | 16733
              268676 | 2159 |  7666
              279229 | 2018 |     2
              279229 | 2019 |     2
              279229 | 2020 |     2
              279229 | 2021 |     2
              296427 | 2248 |    33
              298081 | 2274 |  4016
(136 rows)

As all names from a given sector usually are following the same code, the sector code setting can be used to specify a default code which will be applied during a sync.

[mdoering]

@yroskov if you could give me a list of sector keys per code I could update their settings in the database directly to save manual config work

[yroskov]

@mdoering, unfortunately, I cannot recognize neither GSD names nor sector Latin names. I can't do anything about it. Could you please give me a list of GSDs & sectors with missing Codes? Direct links to appropriate CLB locations would be even better.

Most importantly, how do these missing nomenclatural code cases reappear? As far as I remember, we applied Codes to all spotted cases a year ago. How do we have "non-Coded" sectors again? Where is the problem and how to avoid it?

[mdoering]

you would need to lookup the dataset or sector by its key.
For example the list of 136 sectors take the middle id column and use it in this link instead of the 1543:

https://www.checklistbank.org/catalogue/3/sector/sync?sectorKey=1543

[mdoering]

Are you sure you did change all sectors before? I don't think sector settings have changed, they simply were never there I bet.

[yroskov]

I am sure, we fixed Code assignment for all GSDs (well, not for sectors).

[yroskov]

Code application in the CoL should depend on the kingdom:

kingdom Animalia = ICZN
kingdom Archaea = ICNP
kingdom Bacteria = ICNP
kingdom Chromista = ICN
kingdom Fungi = ICN
kingdom Plantae = ICN
kingdom Protozoa = ICZN
Viruses = ICVCN

If you implement this feature in the CLB, we'll have no problem with Code assignment in the present and in the future, I guess.

[mdoering]

Chromista and Protozoa is probably not that simple. We have over 12k zoological Chromista names already and 4,3k botanical Protozoan algaes.

[mdoering]

I am sure, we fixed Code assignment for all GSDs (well, not for sectors).

GSDs might be, but then you clearly miss out the larger mixed ones like ITIS. And it requires a new import after that setting has been applied, not sure if that was done for all?

[mdoering]

maybe we can split the task and I will try to solve all sectors with clear code from kingdom assignment and you can focus on the Chromista and Protozoa sectors?

[yroskov]

My point is simple, it doesn't matter in which Code chromistean and protozoan taxa were originally described. Once they are placed in CoL under the kingdom Chromista they are regulated by ICN, and if taxa placed in the kingdom Protozoa, they are governed by ICZN.

Would you agree, there is a logic behind such simplification? This pragmatic approach will make your and my tasks much easier, and also will not confuse CoL users.

[mdoering]

If that reflects reality, yes. It would be a lot simpler. When is a name published according to a specific code? Can all names be governed by any code or do they have to be explicitly published with one code in mind? What about the authorship style? I am still not entirely sure how to best deal with ambiregnal names.

[DaveNicolson]

Not to further add to the complexity, but the bacterial group cyanobacteria (which we are adding as GSD in ITIS this month) follows the ICN ('botanical' Code) rather than the ICNP, despite its placement within Bacteria. For some time it was treated under both Codes re rules for availability, effective/valid publication, and so on. Hopefully those days are over, as it made everything much messier & harder to deal with, which led us to add this comment to the group a decade ago:

Cyanobacteria are now considered to be bacteria, but their nomenclature has traditionally been treated under the 'botanical' Code of nomenclature, rather than under the separate bacterial Code (see Preamble item 8 of http://www.iapt-taxon.org/nomen/main.php?page=pre&emph=cyanobacteria). However, there are also indications that recently the 'bacteriological' Code has begun to ALSO cover the cyanobacteria (see 'General Consideration 5' at http://www.bacterio.net/-code.html ). In any case, there are only a handful of cyanobacterial names that have any formal standing under the 'bacteriological' Code. Further work will be required before ITIS can consider this group complete

[yroskov]

Our task is to ensure the integrity of the final product. If you look through different journals, Floras/Faunas, monographs/books, you'll find that the style of presentation of data (including scientific names, synonymy, nomenclatural comments & citations) may vary and and be defined by the editorial rules. Thus, the provisions of the Code are always interpreted by editorial practice.

If CoL decides that presentation style of scientific names is actual Kingdom=Code related, then so be it.

Can all names be governed by any code or do they have to be explicitly published with one code in mind?

At the time of the nomenclatural act, the name might be regulated by one Code, and later by another Code, after their actual placement in the classification.

What about the authorship style?

I'd like to see CoL embrace innovation and add the year in the authorstrings of botanical names: Amoria hybrida (L., 1753) C.Presl, 1831.
Paul Kirk introduced this style in the Species Fungorum many years ago: http://www.catalogueoflife.org/annual-checklist/2019/search/all/key/boletus/fossil/1/match/1.
Technically, you can do this now for most (if not all) botanical names in the CoL. We and users will get a lot of benefits from management and usage of enhanced names (priory, homonymy, etc.).

It would be more difficult to convince zoologists to add the author of the subsequent combination in zoological names. Millions of combinations in Zoology do not keep this important information. However, we have such inspiring example from Erik van Nieukerken http://www.catalogueoflife.org/annual-checklist/2019/browse/classification/kingdom/Animalia/phylum/Arthropoda/class/Insecta/order/Lepidoptera/superfamily/Nepticuloidea/fossil/1/match/1

[mdoering]

I don't think COL should apply the simple approach code by kingdom.
Instead we should follow what the specialists in those groups do and what the codes tell us.
To me that clearly means mixing the codes for the names under Protozoa and Chromista.

Citing from ICZN:

The ICZN only applies to animal names, and not to names of plants, fungi, bacteria or viruses, which are covered by separate codes of nomenclature.

Animals include metazoans and protists that have been historically considered in the Kingdom Animalia (i.e., protists that do not primarily use photosynthesis as an energy source, if so they are generally considered to be plants and will fall under the botanical Code, ICN). Although higher-level classifications have changed with modern research, which Code (Zoological, Botanical or Bacterial) covers a particular taxon generally remains constant, as it is agreed that the ultimate goal of nomenclatural rules is to maintain stability in names and not to reflect perspectives on phylogenetic relationships. Thus, for example, fungi remain covered by the Botanical Code, and there is little interest in changing this, despite modern consensus that they have a sister relationship with animals and not plants. Protists that have characteristics of both animals and plants are considered ‘ambiregnal’ and are treated following the rules of both the ICZN and ICN.

Citing ICN

The provisions of this Code apply to all organisms traditionally treated as algae, fungi, or plants, whether fossil or non-fossil, including blue-green algae (Cyanobacteria)3, chytrids, oomycetes, slime moulds, and photosynthetic protists with their taxonomically related non-photosynthetic groups (but excluding Microsporidia).

Other resources:

• Corliss, 1995. The ambiregnal protists and the codes of nomenclature: A brief review of the problem and of proposed solutions.
• Blackwell & Powell. Reconciling Kingdoms with Codes of Nomenclature: Is It Necessary?
• Turland, 2019, p. 113
• Proposals to amend the Code to harmonize botanical and zoological nomenclature used for the so-called “ambiregnal” microorganisms (as exemplified by fossils)