Implement SWORD 2.0 Profile 6.4. - Retrieving the content

[eaquigley]

---

Author Name: Philip Durbin (@pdurbin)
Original Redmine Issue: 3595, https://redmine.hmdc.harvard.edu/issues/3595
Original Date: 2014-02-27
Original Assignee: Philip Durbin

---

http://swordapp.github.io/SWORDv2-Profile/SWORDProfile.html#protocoloperations_retrievingcontent says "the response to a request for the Media Resource should be considered a representation of the entire content of the object on the server, and it is not permissable to attempt to content negotiate for a particular part of the Media Resource."

Since we require Data Deposit API users to zip up their files before adding them to a study/dataset, it stand to reason that when they "retrieve content from the server" that it would also be a zip file.

Please note that we much consider terms of use discussions that have taken place in #3277, for example, which is about the Data Sharing API.

This feature was requested by the Open Science Framework (OSF) team at http://irclog.iq.harvard.edu/dvn/2014-02-27#i_6612

---

Related issue(s): #338
Redmine related issue(s): 3753

---

[pdurbin]

Please note that we much consider terms of use discussions that have taken place in #3277, for example, which is about the Data Sharing API.

@eaquigley can you please take a look at this ticket as well as 3277 from Redmine? It was never migrated to GitHub issues but has some good discussion: Feature #3277: Data Sharing API: Allow data owner to download without TOU - https://redmine.hmdc.harvard.edu/issues/3277

I'm asking because @augustfly is also interested in downloading files via the API and I'm not sure what the plan is for Terms of Use.

[eaquigley]

@pdurbin I'm a bit confused by this ticket. A few questions:

1. Does this relate to the Data Deposit API, Data Sharing API, or both?
2. Is this ticket only about allowing the owner of data to download the data without having to agree to Terms of Use?
   I'm not really sure what exactly this ticket is about based off the title and comments.

[pdurbin]

@eaquigley it might help to read through the "Direct access to data via URL?" discussion at https://groups.google.com/forum/#!topic/dataverse-community/Uz-_PoyxvY0

In that thread I summarize it this way:

"Why can't I use the API to download something I just uploaded via the API?"

[eaquigley]

If a dataset creator/admin/whatever has assigned CC0 to their dataset, they will not get hit with the Terms of Use and having to agree to that via the API. If the dataset creator/admin/whatever has assigned their own Terms of Use to their dataset, then an end user will have to agree to it via the API. If we have time to develop logic for the dataset creator/admin/whatever to not get the Terms of Use if its their own dataset they are downloading via API, then great. Allowing those with better knowledge of these things to decide.

[pdurbin]

If a dataset creator/admin/whatever has assigned CC0 to their dataset

There is not currently a way to specify a license (such as CC0) for a dataset, as far as I know. I believe this ticket is blocked on #22.

[pdurbin]

This ticket continues to be blocked on #22.

[scolapasta]

@pdurbin Is this something we still need or have we handled it with the download API?

[pdurbin]

I'd say to be more fully SWORD-compliant we still need this. My understanding is that it would be a representation of a dataset as a zip file.

[kcondon]

@pdurbin Would you rephrase this ticket in terms of specific user-oriented behavior? It's hard to prioritize otherwise. It seems to relate to 1. SWORD compliance, 2. Downloading a Dataset as a zip file, 3. Downloading files as a zip file, 4. Also allowing downloading of individual files (see irclog), 5. Requiring an file/dataset owner to complete TOU before downloading their own file.

[pdurbin]

I'm closing this issue. If someone out there wants us to implement section 6.4 "Retrieving the content" from the SWORD spec, please get in touch!