Closed turbomam closed 2 years ago
These notes aren't really about this issue. Move to another existing issue?
why are there different tooltips for PI and Contrib ORCiD etc. ... couldn't we just use person value's slots in both places on the study submission pages?
box around PI fields could give a sense of encapsulation?
or could have tooltip on principal investigator slot
we don't have any instances of principal investigator or contributor... weak grounds for creating classes
@turbomam ... not sure what you need from me here? Anything?
@turbomam ... not sure what you need from me here? Anything?
Thanks. I think I'll be able to resolve this in the process of switching from the static MIxS 5 import to a dynamic MIxS 6 import. I'll keep you posted.
I'm going to move this one to April as it appears to be in process. I will remove @mslarae13 as an assignee.
@dehays provided some useful feedback about other classes that could receive this information, esp. 'Data Types'. Omics processing?
I have also started to think of representing the DOM location for each display hint annotation, so that an external lookup table wouldn't be required. If we did this, @subdavis, what kind of notation would you want me to use? In Chrome's developer console, there are a couple of different copy as options.
I don't expect to close this before April 28th or possible even the month's end.
I don't expect to close this before April 28th or possible even the month's end.
would you like me to move this to May or put it back in the backlog? @turbomam
@turbomam I will move this to May but if it should go in the backlog let me know and we can remove it from the May sprint
Moving to June sprint per @turbomam on stand up
We need better overall alignment between the labels used on the Submission Portal form elements, the field names returned by the API, and the real slots names defined by the NMDC Schema.
Then, we need to determine what element in the NMDC Schema should bear the display hints annotations, and what that formally corresponds to in the Portal's DOM. (@subdavis ?)
There's no reason for a submitter to enter NCBI BioProject Title
in the Submission Portal. It can be looked up from the NCBI BioProject Accession, either during entry into the Submission Portal, or in the downstream metadata repair service (e.g. submissions_as_studies.py
). Then there's no chance of typographic errors.
Mark and I had a conversation about this recently, and there were two outcomes
We could schedule this work for the July sprint.
Thanks, @subdavis
I would clarify that the use case for the first bullet point is validating soemthing like an NCBI BioProject accession.
The ID for the EMP500 BioProject is 686344 and the accession is PRJEB42019. THE NMDC assumption is that all identifiers will be prefixed, so the user should enter something like NCBI:PRJEB42019 or BIOPROJECT:PRJEB42019. @cmungall , which do you prefer?
I'll think about other use cases for validation and lookup.
In-portal validation and lookup would be very valuable, but maybe we should move those two ideas to a separate issue like #354 ?
My intent for this issue was to document the fact that there are elements on the study portion of the submission forms that don't relate directly to any slot or class in the schema, which makes it hard to assert what display hint goes where.
I'll be providing more documentation on this soon. @subdavis are you coming to the NMDC Sync meeting this afternoon?
Study form elements to NMDC slots shows the relationships between the submission portal form element labels, the corresponding fields in the JSON returned byhttps://data.dev.microbiomedata.org/api/metadata_submission, and the corresponding NMDC Schema slots.
There's some overlap between this new sheet and the earlier SubmissionPortal_TermUpdates
@ssarrafan : the schema still doesn't authoritatively say what display hint goes where on the NMDC portals, and I don't think the portals are getting their hints from the schema either. This needs work.
I'll move to July but let me know if it's not high priority so I can move it to the backlog.
@turbomam @cmungall based on the meeting today with Kitwware does it make sense to move this to August or to backlog? Or would you like me to keep it in for July?
@ssarrafan : the schema still doesn't authoritatively say what display hint goes where on the NMDC portals, and I don't think the portals are getting their hints from the schema either. This needs work.
I think this needs to be another / separate ticket if there isn't already one.
@ssarrafan : the schema still doesn't authoritatively say what display hint goes where on the NMDC portals, and I don't think the portals are getting their hints from the schema either. This needs work.
I think this needs to be another / separate ticket if there isn't already one.
@mslarae13 @turbomam so the idea is to close this issue and open a new one for the proper fix? interim fix label?
I'll close this and tag it with interim fix. When you're ready to work on the change we can add another ticket. @mslarae13 @turbomam
See #267 and SubmissionPortal_TermUpdates
I have added display hint annotations to the NMDC schema for most of the rows in the
SubmissionPortal_TermUpdates
sheet. In those cases, I have provided jq paths to the hints.However, I don't see anything related to studies in the schema that correlates strongly with "bioproject title" or "data type"
I guess you could argue that alternative titles could be used for "bioproject title", but the semantics are pretty weak.
Also, why would somebody enter a title when it could be looked up from the accession?
For "Data Types", is that supposed to correlate with analysis_type from our DataHarmonizer interface? I don't think that term is defined in either the NMDC or MIxS schemas yet.