• Icon: Story Story
    • Resolution: Done
    • Icon: Minor Minor
    • None
    • 1
    • 20160630-20160707, 20170313-20170320, 20170320-20170326, 20170327-20170403, 20170404-20170410, 20170410-20170417, 20170418-20170424, 20170424-20170501, 20170502-20170508, 20170522-20170528, 20170605-20170611, 20170612-20170618, 20170619-20170625, 20170626-20170709, 20170710-20170716, 20170717-20170723, 20170724-20170730

      Refine the admin capabilities following internal review. Specific topics include:

      Admin side Metadata Entry Form

      I'm stealing a lot of ideas from Zenodo here, might be worth you reviewing their approach again

      • Can we also include the authors affiliations? Instead of just comma separated author lists you could add a row for each author and in each row you have one box for their name and a second box for their affiliation/institution.
      • We should add a license section with 2 questions
      • Access - open, embargoed, or restricted. having another "hidden" option might be useful for when we're prepping a page but aren't ready for it to be listed publicly yet because we haven't finished collecting the data or are waiting on something else to happen.
      • License - CC Zero, etc. I imagine there's a way to import a list of the most common ones, or we could decide on a smaller subset of options we'd allow users to pick from. 99% of the time it should be CC Zero but just in case...?
      • We should add a field that lets you specify whether the DOI is "primary data", e.g. a new Collection, or "secondary analyses" derived from primary data on TCIA. When "secondary analyses" is selected we should provide an extra option to connect the DOI to the collection(s) its derived from.
      • the keyword field needs to tie into some kind of standardized set of terms or else we should just get rid of it. free text will result in a mess of useless tags. Is there a way to tie this to the Pubmed MESH system?
      • the publisher year field should default to the current year. Zenodo defaults to today's date (uses full date, not just year). perhaps we should be using full dates?
      • Is the References field just for listing other papers the author of the DOI wants other people to cite when using the DOI? If so, should this be titled "Required Citations" or something similar? Btw I don't think we should be trying to track every other paper that has ever made use of the DOI itself here. It'd be cool if there was some way to do an automated feed from google scholar or elsewhere that would automatically show all known publications that have cited the data on the end user side so we don't have to enter such a list manually.
      • We should add a field where people can enter in funding support info (e.g. grants, contracts, etc that enabled creation/sharing of the data)
      • Suggest you shorten "Submit Query" button to "Submit".
      • When trying to submit my first DOI nothing appeared to happen. I clicked submit twice. I had intentionally entered bad data into the year field. When I corrected this it worked, but you should try to put some kind of feedback about what is wrong (e.g. data entry error checks of some kind). Also, after correcting the year and the submission going through I now see all 3 entries in the index list, so even though I put in bad data it still got submitted to Datacite.
      • When trying to create a second DOI I left most fields empty. Same thing where nothing seems to happen when I press submit, but the request is sent to Datacite anyhow and the DOI is created.

        Admin side Add Resources form

      • can the "Add resources" workflow be integrated into the main data entry page? i don't see why they're separated. not a huge deal, but if it's easy to put into a single creation workflow that makes more sense to me.
      • On https://pubhub.cancerimagingarchive.net/details?doi=http://dx.doi.org/10.5072/FK2.2001.nrfo8klp&version=1 I added 2 resources. The first one I chose URL and entered an invalid URL (it shouldn't let me do that). The second one I added I chose Image, but for some reason they both link to the broken URL I entered.
      • How is the image resource supposed to work? Do you upload a JNLP or is that JNLP going to be created somehow based off of what you put in the fields for the image resource? It seems like it would make sense to either enter in NBIA collection name (for new primary data/collections) or a shared list name (for secondary analyses DOIs) in order to tell it what data to use to create the JNLP.
      • For adding file resources it would be good to modify the "Drop file here" text to say "Click or drop file here" or something.
      • In https://pubhub.cancerimagingarchive.net/details?doi=http://dx.doi.org/10.5072/FK2..dziebgja I added an exe file as a file attachment. Should we be restricting executable files or other thing that could pose security risks? I realize this is an admin interface that will be password protected but still might be asking for trouble. Also, it seems attaching the exe file broke the page. It won't load if you click version 1 to try and get to the data.
      • When I go back to add more resources to a DOI it offers that I can select older resources at the top rather than adding new resources. I don't understand what that actually does. It doesn't seem to make any difference in the result whether I click on them or not when adding new resources.

            ksmith01 Kirk Smith
            ashish Ashish Sharma
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated:
              Resolved: