1

Re: Reporting and cross list synonymy - your input please

In response to a problem with the report wizard in Recorder 6 – 693-015 - Report about a place for selected taxa does not do cross list synonymy, Dorset Software have proposed a solution that will mean that the ‘Use Full translation to preferred term’ option (under Tools – Options in Recorder 6) will be removed and Recorder 6 will always use full translation to preferred term.

I’m trying to get a picture of how many users deselect this option, and what the potential implications will be if this option was removed?

In terms of using the Report Wizard, it would mean that users could pull out all records for a particular taxon, irrespective of which checklist was used to input the record. However, this would also mean that users could not select to report on records only inputted from a particular list.

I would very much appreciate any comments users have, so that we can make the best decision on your behalf.

Many thanks,

Sarah

Sarah Shaw
Biodiversity Information Assistant
JNCC

Sarah Shaw
Biodiversity Information Assistant
JNCC

2

Re: Reporting and cross list synonymy - your input please

I didn't even realise that option was there. It terms of feeling the effects of its removal, we wouldn't miss it here. However, we definitely would miss being able to report on records entered against a certain list. We frequently use the report wizard to report on, say, all Odonata by choosing the Odonata list and selecting All taxa from list.

But I suspect that's not quite what you mean when you say, "this would also mean that users could not select to report on records only inputted from a particular list."

Charles

Charles Roper
Digital Development Manager | Field Studies Council
http://www.field-studies-council.org | https://twitter.com/charlesroper | https://twitter.com/fsc_digital

3

Re: Reporting and cross list synonymy - your input please

No, that's not quite what I originally meant at least. You will still be able to select to report all taxa from a list. What you would not be able to do is say 'all taxa from a list, but only if they were recorded against that specific list'. Having said that, you can't really do that now anyway!

The real difference between having this option checked and unchecked is as follows. With full translation unchecked, you will only find synonyms in other lists if the taxon is listed at least once on both lists with exactly the same term and author. With it checked (the proposed behaviour all the time) it won't matter if there are any shared terms between lists or not - they will all get detected as long as the dictionary data is right.

The example we looked at for this was Arvicola terrestris. On the Recorder 3.3 list this is listed as "Arvicola terrestris, Linnaeus 1758" but on Mammalia its "Arvicola terrestris, (Linnaeus 1758)" which I believe is more accurate. But, because neither list actually shares an identical term with the other list, mapping or reporting the Mammalia version will not find the occurrences entered against the Recorder 3.3 version and vice versa, unless the full translation option is checked. As this behaviour sounds undesirable to me anyway I have proposed the option is removed, and the more accurate full translation behaviuor is always used. I would prefer to simplify things in Recorder where possible, especially where an option is only likely to cause confusion and never likely to have a positive effect.

John van Breda
Biodiverse IT

4

Re: Reporting and cross list synonymy - your input please

Hi there,

Not having access to R6 at the moment makes this slightly confusing, however reading between the lines here, how much of this is a work-round to Nameserver functionality not being fully operational?

The way I set up a few reports that included data which may have been inputted using different lists was to to do this:

Firstly, run a few test cases to find most common-used lists, then design a report which included everything on those lists.
Then export the lot into Excel & use a lookup function to narrow down duplicated species. Crude, but it worked!

Although I'd seen the 'Full Translation" option in the Tools menu, I'd never really pondered on it's implications.

5

Re: Reporting and cross list synonymy - your input please

Hi

Sorry for the confusion - I did find this difficult to word.

As I understand it (and John will probably be able to confirm this), this solution will involve using Nameserver instead of Index_Taxon_Synonym in the query.

The Nameserver table is still a work in progress, but the plan is that by the time this fix is released then the Namserver table should be ready.

Regards,

Sarah

Sarah Shaw
Biodiversity Information Assistant
JNCC

Sarah Shaw
Biodiversity Information Assistant
JNCC

6

Re: Reporting and cross list synonymy - your input please

Yes, that's right, with one clarification. The full translation option actually means "treat items as synonyms when they have the same recommended_taxon_list_item_key in index_taxon_name". This is slightly different than saying it just uses Nameserver, because the recommended_taxon_list_item_key population algorithms (that run when you rebuild indexes) actually revert to the old methods of identifying synonyms when taxa are not handled in the nameserver.

So, having full translation unchecked means:
"treat taxa as synonyms if they are identified as such using Index_Taxon_Synonym"

and checked means:
"treat taxa as synonyms if they are identified as such using the nameserver. If not in the nameserver, treat taxa as synonyms if they are identified as such using Index_Taxon_Synonym"

So, personally, I can't see the point of allowing this option to be turned off. Its even been turned on in Luxembourg when the nameserver table wasn't in use and there were no detrimental effects.

John van Breda
Biodiverse IT

7

Re: Reporting and cross list synonymy - your input please

Sounds fine to me - I'm in favour of removing it.

Charles

Charles Roper
Digital Development Manager | Field Studies Council
http://www.field-studies-council.org | https://twitter.com/charlesroper | https://twitter.com/fsc_digital

8

Re: Reporting and cross list synonymy - your input please

I have just discovered to that to extract all of our Arvicola terrestris records which for historic reasons were entered using a mixture of the preferred Mammalia checklist and the Recorder 3.3 checklist I had to select the species from both checklists in the report wizard, otherwise I only got the records from the one specified. I had been living in blissful ignorance of this issue until I spotted an anomaly in the records being reported, thinking up until now that I was getting all the records.

Having found this thread I ticked "Use full translation to Preferred Term" under Options->General and now I get all the records regardless of which list is specified.

I just thought I'd highlight this for anyone else that might think they are getting all the records for a species when they may only be getting a sub-set....

Regards, Keith

9

Re: Reporting and cross list synonymy - your input please

This problem of reporting from different lists was highlighted quite early on after our upgrade from Rec3.
Since then I have been using reporting methods that ensure I dont rely on the dictionary.

Part of this process is repointing rec3 taxons to the Reccommended taxon via SQL scripts.

So very happy to see this particular proposal going forward.


It does link in with the wider point of the 'robustness' of the taxon dictionary, which remains an issue.

We want to have faith in it, but the problems tell us that we cant.

If for example we still have to create our own designation lists to ensure that 'chiroptera' are identified as protected species:
http://forums.nbn.org.uk/viewtopic.php?id=1540

Or cannot rely on the taxon hierarchy actually finding what its supposed to:
http://forums.nbn.org.uk/viewtopic.php?id=1598

then I regard the value of the seemingly comprehensive dictionary, diminished.  There is much work still to be done here!

M

Cumbria Biodiversity Data Centre
Tullie House Museum