The following are anecdotes contributed by users of iDigBio's data and data portal. They aim to be helpful in several ways:
- Anyone submitting data should read them and make adjustments and improvements in their own data to avoid the issues found by others.
- They can be a springboard for interested parties to address overall data quality issues.
- This is also the place to document portal interaction difficulties.
iDigBio interest groups are aware of this documentation and feed it back to their respective groups. Progress and feedback from the developers is also noted here.
Another useful feature to take note of are iDigBio's data flags - common data quality issues and data corrections that may be performed on recordsets to improve the capabilities of iDigBio Search (see https://github.com/iDigBio/idigbio-search-api/wiki/Data-Quality-Flags). Data quality flags are identified for each of the ingested recordsets on their respective portal webpage.
||K. Schultz, EOL (2014) https://www.idigbio.org/redmine/issues/1393|
||C. Johnson, AEC (2/2015) https://www.idigbio.org/redmine/issues/1394|
||K. Seltmann, R. Rabeler, TTD TCN (2/2015) https://www.idigbio.org/redmine/issues/1395|
One thing we might think of in any attempt to create a national vascular plant portal - how easy is it for the user to acquire data. I had this idea after conducting a simple search for records of one species in each of four portals: iDigBio, SEINet, CPNWH, and the California Consortium.
How many clicks/windows does the user have to navigate to get the results? Ideally it should be minimal. I found in one case (CPNWH: http://www.pnwherbaria.org/data/search.php) it was - after I entered the name and hit "search", I had label data, a map of georeferenced specimens, and any extant images, even records entered under synonyms. I could easily examine label data, sometimes including annotations (!), enlarge images, the map, etc. In other portals, I had to make additional input to get one or more of types of data available.
The Pacific Northwest portal allows, as one option, grouping records, incl. sorting by collection and number. I found this quite useful - it's a way to spot specimens that are likely duplicates that may, for various reasons, be filed under different names. Someone doing monographic work can learn of (or examine if they are imaged) specimens they may not otherwise know about - the commenting/ Filtered Push system where one could send comments to all owners of the duplicates simultaneously would be a wonderful addition to any such portal."
Brent Mishler replied to my original post and pointed out some of the advantages of the data display in the Consortium of California Herbaria (http://ucjeps.berkeley.edu/consortium/) page.
"The table view used by CCH has advantages in that it can be sorted quickly and easily on any column. CPNWH certainly does have an attractive search return pages -- you are looking at images, label data etc. as soon as you search, but note that's only for the first 50 records, as PNW groups into bins of 50, which need to be clicked open separately to be viewed or searched. Also, after you've executed your search, it takes three clicks to sort the results (e.g. by Collector #) in CPNWH while it takes only one click in CCH, so I think the CCH setup ends up being more efficient for certain information needs."
--> To me, one of the things that iDigBio should be concerned about is having the portal be easily usable. If we want it to be the "one stop" for biodiversity data, we need to see what users can get from other portals and provide improvements to that level of info. If it's easier to get the info by using a combination of other sources, folks still might do that. In the examples I sent along on the screen shots, that's what I am trying to show - what we present should be at least as good as what you can get elsewhere. If you compare the results of the "label" view that you get in the iDigBio portal with that in the CPNWH portal, it's clear (at least to me....) that ours is inferior for the reasons that I pointed out. See example.
|R. Rabeler, TTD TCN (2/2015) https://www.idigbio.org/redmine/issues/1396|
Database Search resulted in a rich specimen record dataset (in this case for lichens and bryophytes) for a participant at the IPT workshop. The researcher wants only the distinct taxon names (and count of speicmens per distinct taxon name). The researcher describes that downloading the dataset is then followed by "a lot of work" to get the distinct list of taxon names to share with a colleague.
|Mac Alford, (2015), entered here by Deb.|
|Collections - It would be great if collections could get a sense of their collections uniqueness – what do they have in their collections that no other collection in the portal has – either taxonomically or geographically. It would be great if you could get a uniqueness factor and display a summary of the unique aspects of the collection. Maybe even include preparation information – do I have tissues that nobody else has of a particular species or from a particular area? This would be very handy in grant proposals and in justifying the existence of particularly small collections.||Andy Bentley (3/2015)|
|Researchers – It would be very handy if researchers could subscribe to a portal in order to get an alert when specimens of target species or geographic regions are added to the portal. For research purposes, this would alert them to new material that may warrant their inspection and would facilitate loan traffic from collections that have newly catalog material.||Andy Bentley (3/2015)|