[Taxacom] Wikipedia classification

[Tony.Rees at csiro.au]

Re my last remark

----
> Regarding the more recent discussion concerning the amount of detail and functions for a disambiguation page - in my view it should be at least a way to differentiate between a current valid name (or "name in current use" maybe) and a name that is a known synonym (of xxxxx...) - then probably you would point the latter to the current page for xxxxxx instead of making another page for the junior homonym in question??
----

My apologies, you or relevant persons are already doing this, at least in the Latreillia (fly) instance.

Regards - Tony


-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Tony.Rees at csiro.au
Sent: Monday, 6 July 2009 10:07 AM
To: una.smith at att.net; taxacom at mailman.nhm.ku.edu
Subject: [ExternalEmail] Re: [Taxacom] Wikipedia classification

Dear Una, all,

Noting that this discussion has already moved on a little, but to answer Una's question:

<snip>
> > There are LOTS of these at genus level - in IRMNG my current count of
> > non-unique genus names is currently 68730 and rising.

> Could you do a data analysis for Wikipedia?  Identify all the names
> on your list of 68730 that on Wikipedia are not disambiguation pages?
> The result would be a project for Wikipedia's "disambiguous" editors
> to tackle.
</snip>

If Wikipedia supplies me a list of current genus pages that are not disambiguation pages, yes I can do that.

However there are some issues regarding the status of some of these duplicates in my present holdings, some of which are as follows:

- In a few cases (probably less than 50-100) there may be instances of the same genus name being used e.g. in different phyla that do represent the same name - particularly for some eighteenth century names, that I have not yet cleaned up in my own system (for example fungal species attributed to a genus name in algae or vice versa), where I may have the name listed twice and not yet fixed it.

- Also in my list are nomina nuda which can be homonyms, i.e. the author published the name without fulfilling all of the criteria for valid publication (a initial nomenclatural event) then later published it again validly (or it was republished validly in another authors work). I have both instances in my system, contributing to maybe 5-10% of homonyms.

- I include family and higher taxonomic attributions for my genera names, but these are preliminary and subject to modification as they are reviewed further

- Not all of my genera have associated authority information at this time, but the majority do (and for the homonym genera, only 885 lack an authority)

- I am sure that my list of homonyms will not be totally complete, although at a guess it is probably around 80% there or better

- Some of the listed homonyms may be very obscure, e.g. perhaps never used since their original publication over 100 years ago, thus perhaps not very relevant to the "general reader"

- Only a subset (I would guess perhaps 20-30%) of known junior homonyms in my system currently point to their current valid name (IRMNG being a work in progress, many things to do!!).

There is also some good news: for the 68730 homonym genera currently in my list, these only represent around 29000 distinct names i.e. the number of disambiguation pages that would be required (this is the equivalent to the difference between having 2 persons named John Smith, but only 1 name shared between them).

However...  I do have some concerns about the degree of manual effort that would be required for someone to create and maintain such pages, considering that they are based on free text manual construction - to me a database approach is the only one which is tractable - e.g. if another Latreillia appears in my system, the page http://www.marine.csiro.au/mirrorsearch/ir_search.go?searchtxt=Latreillia&hlevel=genus automatically updates to show three instances, and if one disappears, well fine, it just shows a single one... Also if I change something at a higher taxonomic level (for example moving a family to a different order), there is nothing else to do, all the pages that cite that family automatically inherit and display the new information as appropriate. However that's one of the prices Wikipedia pays for being a general encyclopedia as opposed to a specialist taxonomic one, I guess.

Regarding the more recent discussion concerning the amount of detail and functions for a disambiguation page - in my view it should be at least a way to differentiate between a current valid name (or "name in current use" maybe) and a name that is a known synonym (of xxxxx...) - then probably you would point the latter to the current page for xxxxxx instead of making another page for the junior homonym in question??

Anyway, I can certainly assist with the initial request if you still think it is worthwhile bearing in mind the above.

Regards - Tony