Hi all,

Thanks for clarifying this, we didn't had the whole picture here at BBIF and were feeling quite lost with this request.

I also think that this is a rather suboptimal use of the standard. Anyone who has managed databases (in a broad sense) knows that it's a bad idea to attach properties to the wrong entity in order to circumvent a "local" problem. And when dealing with standards, I guess it's a very bad idea to change the standard (making things less clear/more complex for everyone) only in order to legitimate this unusual need.

Saying "no" to user requests is always hard, but sometimes necessary to keep the rest of the world sane. We also have to keep in mind that with standards, once we add a unappropriate field (or a wrongly placed field), we'll have to live with it forever.

I can imagine a few better ways for this publisher to solve its problem:

1) In parallel to these occurences, derive a checklist that use this extension and publish both in their IPT instance
2) Publish this data in a custom form in one of the Remarks fields
3) Create a very "provider-specific" extension to publish these two fields only. In that case, I'd also document properly this extension to make sure that the rest of the world understand it's a very specific extension that they probably should not use.
4) Review their source data so "species" attribute are linked to a species (rather than occurrence) entity.

Hope I haven't offended anyone, this is intended to be constructive criticism. But IMHO we have to stay firm with such decisions: imagine how the standard would look like in a few years after accepting 100 similar requests from 100 different users ?

Best,

Nicolas

On 27/11/12 17:03, "Markus Döring (GBIF)" wrote:
Hi all,
I agree with all that Peter said and it doesn´t seem appropriate to use the species distribution extension for occurrences. Its the distribution of a species, not of a single occurrence.

Ideally I would also think sharing the threat or Cites status of a species should be done separately in a checklist instead of pushing it all into flat occurrences. But then again Peter pointed out rightly that we do that already for some terms. I would rather remove those taxonomic terms from the occurrence core instead of adding new ones, but Im happily convinced of the opposite :)

Can someone explain the use case a bit more to understand the needs?

best,
Markus

PS: And yes, it sounds like a tdwg content discussion, but then again dwca extensions could be a bit too implementation specific, dont know.


On 26.11.2012, at 13:14, DESMET, Peter wrote:


Hi all,

Here are my remarks about this request:

As mentioned in the original email, the reason for this request is the need to use threatStatus and appendixCITES for occurrence records.

1) These are properties of taxa, not occurrences. They *could* be used for the taxa referred in an occurrence (and shared in an occurrence dataset), but we have to ask the question if that information cannot be better published as a checklist (using the taxon core) and/or if it should be aggregated/published by that data publisher (the data publisher is often not the source for this kind of data in an occurrence database, it was derived from somewhere else).

Mind you, this is not limited to these terms: we already have several terms from the taxon "class" available in the occurrence core where we have the same issue, e.g. does it make sense to provide information regarding the originalNameUsage or acceptedNameUsage in an occurrence dataset?

So the main question is: do we want to introduce more terms like this?

2) Assuming we accept these terms for occurrence and we use an extension to do so. Do we allow a one-to-many mapping between the occurrence core and the distribution extension? Does it make sense? Isn't that (again) better suited in a checklist?

3) Assuming we only allow a one-to-one mapping, do we need an extension? All terms in the distribution extension are available in the occurrence core, except: threatStatus, appendixCITES and source (this last term should be covered by http://rs.tdwg.org/dwc/terms/index.htm#dcterms:references).

Maybe (if remark 1 is answered) they should be added to the occurrence core and officially added to the list of Darwin Core terms?

4) Although the question was raised by an IPT user, it might be more useful to have this discussion on the TDWG list? To me this is a conceptual issue first and an implementation issue later.

5) Technical question: why do we need a version increment to the distribution extension if we allow it for occurrences?

Cheers,

Peter

Van: ipt-bounces@lists.gbif.org [ipt-bounces@lists.gbif.org] namens Tim Robertson [GBIF] [trobertson@gbif.org]
Verzonden: donderdag 22 november 2012 9:41
Aan: ipt@lists.gbif.org mailing list
CC: Danny Vélez
Onderwerp: [IPT] Fwd: "Species Distribution" extension for Occurrence Core in IPT

Dear IPT community,

Please see the request below for using the species distribution extension [1] with occurrence core [2].  This request comes from Colombia, who are setting up an IPT network in the country.
I don't see any issue with this, but wanted to put this through the correct forum (this list) for any comments.

Does anyone have any concerns they would like to raise?  

Please note, that extensions are considered immutable in concept, although we allow for minor editorial changes.  Therefore this request would warrant a version increment of the distribution extension; reflected in the namespace.  IPTs would therefore see 2 species distribution extensions (the second having the new version).  The current version cannot be removed as datasets already exist using it, but rather would be marked as "Deprecated" in the description so new installations and updates would show this.

Thanks all,
Tim

[1] http://rs.gbif.org/extension/gbif/1.0/distribution.xml
[2] http://rs.gbif.org/core/dwc_occurrence.xml
* * * * * * * * * * * * * D I S C L A I M E R * * * * * * * * * * * * * 
Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document. 
The views expressed in this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document. _______________________________________________
IPT mailing list
IPT@lists.gbif.org
http://lists.gbif.org/mailman/listinfo/ipt

_______________________________________________
IPT mailing list
IPT@lists.gbif.org
http://lists.gbif.org/mailman/listinfo/ipt