NCEAS / arctic-data-outreach

Outreach and engagement activities for the Arctic Data Center
Apache License 2.0
3 stars 4 forks source link

create NSF awardee notification system #4

Closed vlraymond closed 5 years ago

vlraymond commented 6 years ago

create system for automating process of pulling NSF awards from their site currently VR is doing a manual pull from their site.

isteves commented 6 years ago

May be addressed by: https://github.com/NCEAS/datamgmt/blob/master/R/get_awards.R

mbjones commented 6 years ago

This feature request -- to automate frequent interactions with NSF PIs throughout their award lifecycle -- was raised last fall during our site visit with NSF, and was raised again this week by NSF, so it needs to rise to the top of our priority queue. Here's a beginning of the description of what we need to produce, which should be captured in a series of other tickets as well.

This is related to the Arctic Data Center Project Pages feature, which will be described in a separate feature issue and linked here. I need to send a schedule for this to NSF ASAP. Let's discuss who is best to lead and participate in this project.

csjx commented 6 years ago

The conversation I've had with Polar Field Services and gaining access to their ARLSS database programmatically is pertinent here, so I'll add that into the plan.

mbjones commented 6 years ago

Yes, I am about to add another ticket about ARMAP and project pages that relates. Also, issue #21 relates as well.

maier-m commented 6 years ago

What should the proper email address be for sending these notification emails? A. support@arcticdata.io B. new email (notifications@arcticdata.io perhaps?)

laurenwalker commented 6 years ago

I like having a separate email for this - notifications@arcticdata.io sounds good.

On Wed, Jun 13, 2018 at 4:25 PM, Mitchell Maier notifications@github.com wrote:

What should the proper email address be for sending these notification emails? A. support@arcticdata.io B. new email (notifications@arcticdata.io perhaps?)

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/NCEAS/arctic-data-outreach/issues/4#issuecomment-397074143, or mute the thread https://github.com/notifications/unsubscribe-auth/AGVeFqK5Hb8_QF2OjVl0KHxygTyuXNznks5t8XVOgaJpZM4RoyZI .

-- National Center for Ecological Analysis and Synthesis (NCEAS) University of California Santa Barbara (UCSB)

mbjones commented 6 years ago

We will need to respond if the reply to the address, so either way the correspondence should go into RT to ensure we follow through and address the issue. So, I'm not sure there is a benefit of a different address, except if we want to route these to a different RT queue.

jagoldstein commented 6 years ago

Well in that case, I think we should just stick with using support@arcticdata.io. It will be less confusing than having multiple addresses and we can just add a new RT label to those tickets if needed. @maier-m

maier-m commented 6 years ago

I feel like a second queue would be nice right? Otherwise, the arcticdata queue could get confusing no?(with support and notification emails all together)

jagoldstein commented 6 years ago

I feel like these are still a type of ADC support tickets. I'd rather have to check fewer queues than more. Right now the other queue we monitor ('Ecoinfo') is mostly related to KNB support. I don't see this case as being that distinct from our other ADC support tickets. They will be related to pre-processing and will be helpful to engage contributors in the convo early-on. Even if we resolve the tickets, we will be able to dig them up later as needed.

csjx commented 6 years ago

During the ADC call, we discussed the components of a Phase 1 Notification system based on an R package polling the NSF awards database, populating a CSV file with minimal award information geared toward start-of-award and end-of-award milestones, and sending RT API calls to create RT tickets that are in turn sent to the PI for notification. The design notes are at https://hpad.dataone.org/GYTgxgTBDMCGCmBaArMgJmxAWCB2YisWGiuAbAEZkCMa0EZaADLkA===, and will be moved into design documents in this arctic-data-outreach repo. Development will happen in this repository.

We briefly discussed Phase 2 needs, where we support a richer milestone and task system that incorporates information from the Polar Field Services database too, but there was enough to talk about for the initial implementation that we'll schedule more time for later. Given the simplicity of this system, we only wrote out components and sequnces in psuedo-code rather than building UML component and sequence diagrams, which we likely will do for Phase 2.

maier-m commented 6 years ago

Hack down from Arctica meeting 06/15/18

ADC Project Notification Notes (Phase 1)

Information to be stored in awards database

  1. NSF award Number
  2. PI
  3. PI email
  4. Title of project
  5. Initial contacted date
  6. 3 month before due - contacted date
  7. 1 month before due - contacted date
  8. 1 week before due - contacted date
  9. List of ADC pids for project
  10. Award date
  11. Award start date
  12. Award expiration date
  13. Program Type (AON, ARC, ANS, etc. - see below)
  14. PI ORCID ID (used as a search key into the ADC datasets)
  15. RT ticket #

Workflow components

Workflow sequence

Emails

Possibly MIME/multipart emails to add more branding and design elements (https://rt-wiki.bestpractical.com/wiki/REST#Tickets_History_Reply)

  1. Congrats email for 2018 or new awardees. Address timelines on a program specific basis. Ask to create ORCID if PI's don't have one yet.
  2. End of award emails (3 mo, 1 mo, 1 wk)

Testing

Swap out PI emails for test email address

NSF Programs

Search limited by fundProgramName with POLAR|ARCTIC|AON

jagoldstein commented 6 years ago

@dmullen17 will merge the updated email templates from his fork to https://github.com/NCEAS/awards-bot/tree/master/inst/emails

mbjones commented 5 years ago

@csjx @dmullen17 I think this ticket can now be closed, right?

dmullen17 commented 5 years ago

@mbjones @csjx yep it can be closed