finding a (plazi) data set

[myrmoteras]

@mdoering @timrobertson100 @gsautter

I am looking for an article that should be on GBIF, but I can’t find it.

• http://tb.plazi.org/GgServer/summary/0767FFD8082E3F06FFB13B380732F137
• or DOI: https://doi.org/10.11646/zootaxa.4482.2.3

Now there are several questions:

• What is the best way to find an article on GBIF? I thought by entering the DOI that might do it.
• Finding the GBIF version of the article would be easier if we could include the GBIF ID into the Treatmentbank summary page or also in the article deposit in BLR. This might be a quesion also to @gsautter because this would need a feedback mechanism after the DWCA has been submitted.
• Is there a way that we could compare the articles for which we submit a DWCA to GBIF and whether they are actually in GBIF? This could also include some additional information.

Finally, I am not clear whether this is the right place to submit such an issue

[MattBlissett]

1. I'd choose a significant part of the title and search for that: https://www.gbif.org/dataset/search?q=New%20species%20and%20new%20records%20of%20Tingidae ... There is an API call to search datasets by DOI, but the website doesn't use it. A nice idea, @MortenHofft? http://api.gbif.org/v1/dataset/doi/10.11646/zootaxa.4482.2.3 → https://www.gbif.org/dataset/fd8ba24f-4d68-431c-81f5-e924906aed5d

2. Yes, that would be done on the Plazi side — I think Guido will already record the GBIF UUIDs, it will just be a matter of displaying them. (If not, use the DOI search to find them.)

3. We made a tool for IPT users to monitor the state of their datasets, https://management-tools.gbif.org/ipt-sync-state (e.g. try http://ipt.gbif.es). There was something similar for the Atlas of Living Australia, but I can't find it (!). You could use a small Javascript that queries your own APIs for the list of datasets you expect to be in GBIF (including their GBIF UUIDs), then for each one make a search for occurrence/name count for each one. Finally, query GBIF for everything published by Plazi, to see any surprises (datasets we have that you don't know we have).

This is a fine place to ask questions.

[MattBlissett]

3. This is the sync tool: https://timrobertson100.github.io/uk-gbif-sync/ . With Plazi's number of datasets you might need a slightly different approach, e.g. offline rather than refreshed every time someone visits the page.

[mdoering]

@gsautter definitely has the GBIF dataset keys. They are returned when first registered and he needs them to trigger subsequent changes and imports, so it should be straight forward to include them anywhere he likes.

Why don't we add the full DOI to the full text index of the dataset? Seems like an obvious improvement

[MattBlissett]

The DOI is probably in there, but one like 10.11646/zootaxa.4482.2.3 probably ends up split into 10, 11646, zootaxa, 4482, 2, 3 which then matches many other things.

[gsautter]

I've found the dataset: https://www.gbif.org/dataset/fd8ba24f-4d68-431c-81f5-e924906aed5d

This most likely has been some timing issue, i.e., the offset between registering the dataset at the GBIF API and the harvesting results showing up in the website.

[gsautter]

I searched for the article title to find it: https://www.gbif.org/dataset/search?q=New+species+and+new+records+of+Tingidae+(Hemiptera.+Heteroptera)+from+Thailand

[myrmoteras]

Actually, to find a DOI, you get it if you enter it with exclamation marks eg "10.11646/zootaxa.4482.2.3"

[MattBlissett]

Is there any remaining confusion or query with this issue? If not, I shall close it.