ThierryBourgoin
commented
7 years ago Hi Markus,
Management Classification is an old story in CoL that so far I remember started around 2005 during a post GB GBIF evening meeting with Frank and perhaps also Yuri. The original idea was quite different of what we use to call now days a management classification in the our GT.
The idea was very practical: to easier manage and follow all the changes proposed by the GSDs I suggested to build what I called a management-classification ( 'M-classification’). It should have been the core reference of CoL = an fixed structured network of taxa (a backbone!) providing anchor points to which one attaches the updated "boxes" of the taxonomic data provided by the GSDs (via their higher parent taxa) but without losing the original schema at each updates. More or less the schema provided in the link to Species 2000 website you join.
The purpose was to start using this process to follow and try to measure - very ambitious! ;-) - what was going on in the CoL (not only in terms of how many taxa or updated GSDs we delivered). It was also a first step to start managing (in a very simplistic way!) multi-classifications by being at any time able to provide different updated views based on this M-classification (for instance presenting locally concurrent taxonomies). In other terms the idea was to deal with a starting point and actualized one (to manage) and not only having the single last updated vision (as we still have now).
During the preparation of the PESI project I suggested a WP to build this M-Classification and to produce an authoritatively relatively stabilized classification that could be populated to answer the needs of stability for many users (society, conservation issues, administrative communities, …).
The M-classification would have acted as a buffer helping managing between the dynamic and evolving expert taxonomic knowledge, and the more classical and stabilized one of the external users (who don’t like changes!). And therefore allowing managing more smoothly the changes provided by taxonomic advances.
As I defined it, the characteristics of the S-Classification were: 1) relaxing some level of ranking (e.g. super-families, sub-families, tribes, …) to the next upper rank 2) allowing polytomy for adapting paraphyly at nodes but excluding polyphyly.
(more or less your suggestion of yesterday with your “pro bono" synonyms as I would call them)
Species 2000 did not joined the PESI project but the idea remained and it ended with our "Higher Level Classification" paper.
In other terms the idea was to have 3 simultaneous classifications:
- a fixed M-classification for technical issues, used only as an anchor reference. Probably only machines readable and for sure non accessible.
- a slowly evolving and less precise (but not wrong) S-classification to satisfy the non-taxonomic community (updated every 3-5 years only)
- A continuously updated CoL for the taxonomic community.
In my idea, the M-classification would have been updated regularly (only every 10 years minimum) with a more recent then fixed one (issued from the S-classification) while the other two would had continued to moved respectively slowly and quickly.
The management classification we have currently corresponds to a S-classification. For me it is normal that it is different from the CoL itself, which reflect more closely teh expertise of the taxonomic community. We never got a real Management classification one.
I've the feeling that we still need an anchored and fixed classification. The Ruggiero published one could play this role for the next years if adopted. New web-versions has already be produced but I'm not sure hat we could compare/evaluate the different versions and for sure certainely not with the current CoL. How could we evaluate and manage correctly the future tasks, maintaining the system, etc if we don't have a fixed reference?
So for your 3 questions: Yes we should adopt Ruggiero paper as a fixed M-classification. Yes we need to continue to manage it to propose a more stabilized taxonomic framework to users and Yes we should go down to family for the S-classification but we need leave GSDs to provide their own classification inside their products. We different classifications enough structured are in concurrence and we should be able to propose the various alternatives.
Hope it is also here what you wanted to discuss ;-) BW, Thierry
Le 16 oct. 2017 à 18:15, Markus Döring notifications@github.com a écrit :
The current CoL uses a higher classification that evolved over time and is in many groups defined by the respective GSD. This causes for example all algae classes and even phyla to disappear when Algaebase was removed.
It is also unclear what relationship the classification in use in the CoL to the officially published classification papers by Ruggiero et al have. They do not match up.
Discuss about the desired future management classification of the CoL.
• Should it follow the Ruggiero paper? • Should it be managed continuously by some advisor group similar to a GSD? • Should it go down to family/order level in all groups or should GSD be able to define the classification? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.
mdoering
The current CoL uses a higher classification that evolved over time and is in many groups defined by the respective GSD. This causes for example all algae classes and even phyla to disappear when Algaebase was removed.
It is also unclear what relationship the classification in use in the CoL to the officially published classification papers by Ruggiero et al have. They do not match up.
Discuss about the desired future management classification of the CoL.
See also the official information from CoL: http://www.catalogueoflife.org/annual-checklist/2017/info/hierarchy
And the related issue #30 about which (higher) ranks the CoL should include.