Wednesday, 3/20/2019
- ID
- 9c90d762-0950-4494-a2bf-a39e0d29fe24
#+BEGIN: clocktable :scope file :maxlevel 2
#+CAPTION: Clock summary at
| Headline | Time | |
|----------------------+---------+-------|
| *Total time* | *11:49* | |
|----------------------+---------+-------|
| Wednesday, 3/20/2019 | 11:49 | |
| \_ Tasks [3/3] | | 11:49 |
#+END:
Tasks [3/3]
Today's meetings
:LOGBOOK:
CLOCK: -- => 0:29
CLOCK: -- => 0:31
CLOCK: -- => 1:16
CLOCK: -- => 0:04
:END:
Fort George grant with Lisa
Purpose: To get up to speed with Lisa's thoughts on what to do with Ft George
and how the disparete orgs in town can move forward to make it a great space.
We have $2000 in the budget this year, not sure what that's being used for. I
believe it's to sponsor a study? The hope is that by spending some money on it
we can unlock matching grants or somesuch.
Team IO standup
Cameron said he's optimistic we'll hit the deadlines we need to, which Ed
strongly disagreed about. Andrew noted that there are a number of unknowns and
pieces of work that we haven't really found solutions for yet, like integrating
FIGmd resutls in Hippo and such.
RM-2292 code review
DONE Write in daily journal
SCHEDULED:
:LOGBOOK:
CLOCK: -- => 0:07
:END:
DONE Add link to "refresh" button to view error report to caregap tool
SCHEDULED:
:LOGBOOK:
CLOCK: -- => 2:38
:END:
https://elationhealth.atlassian.net/browse/SWAT-1194
ASK: Link to download error report does not appear until you refresh the page. (does not automatically appear in Upload History table so I initially thought it wasn’t created). Would be nice to have a “refresh for error report” button or something to that effect.
This took a while because my local install was a bit borked for care gap
importing. There's still some subtle issues with getting all the data just right
and then loading react with my external server model was a trial too.
Eventually got it figured out and it was trivial to make the list refresh when a
file is validated.
DONE Refactor the manage_cds_batch management command
SCHEDULED:
:LOGBOOK:
CLOCK: -- => -10:40
CLOCK: -- => 1:25
CLOCK: -- => 2:35
CLOCK: -- => 0:06
:END:
In review there are a few things standing out that need to be resolved.
Spent a while refactoring the manage_cds_batch management command today. I
justified it because this is both, one of the tests that's failing at the
moment, and because it's a harbinger of whether this whole thing will fly.
There's some strangeness with the CDS model and how NQF codes work within it ...
the frontend depends on the NQF codes, but those are not used in many other places.
Journal
Yesterday ended better than it started, even though we also had a kerfuffle with
Peter deciding on his own to stay in Bucksport rather than take the bus home.
This because he says he hadn't heard back from Emma about whether he could stay
in town. This happened once before and the result wasn't great. Not sure why he
tried it again.
I had a nice one-on-one with Andrew yesterday where we finally got to talk about
thinking coming down the pike. So many of my one-on-one have caught me unaware
with criticism that I am almost nervous to start them now, which is not a great
place to be. But yesterday was great. Andrew did note a few quality issues in
work I did around data migrations that needed to be tightened up. But overall it
was a good meeting.
Emma made a delicious chickpea salad for dinner. I was totally not sold on
whether it would be enough food, but it was fantastic. It also almost felt like
detox after all the meat we'd eaten over the past few days.
I'm currently slogging my way thorugh /Justice on Earth/, the UU common read
book and a series of essays on the intersection of race, economics and climate
justice. Margaret asked me to do a reflection upon reading it, along with a few
other people, and I'm struggling to find an angle. I think I may have to talk
through my own beliefs regarding activism and justice work when it comes to
faith. I think your personal faith has to inform how you move in the world, but
I'm not sold on the idea of USING your faith to push for change. While many of
the essays' authors use UUism as an organizing force, I believe there are better
forums for change. When we, as a religion, approach people with a demand for
change, there is a natural tendency to recoil. As UUs this should be concerning
to us.
Work log
Spent the morning fixing the baseline database image, where I managed to
include the two quality programs Ayhan asked about, but not the measure
definitions.
Once I had that done, I set about getting myself setup for care gap importer
fixes. That took longer than expected because the data model is a bit fragile.
One needs the right practice, patient, gap measure and quality program
inclusion for a the practice for it to all work. Then you have start breaking
things to work through the various error messages.
Once I had it running it was pretty easy to make the upload history panel
auto-update when a new file is uploaded. After that I set about adding a test
for multiple errors in one file. One of the minor issues is that if a column
has two errors, say wrong NPI and missing measure definition, only one of them
will be reported, which means you end up having to do multiple passes of
uploads to find all the errors. An argument could be made that we should have
such poor quality data, but that's unlikely to change soon.
During standup, Andrew suggested that I prioritize fixing the RM-2292 branch,
but found he hadn't saved his fixes in the github review, so had to wait for a
bit. Once I had his suggestions I got to work fixing it. A big piece of work
was adapting the manage_cds_batch command to actually work as designed. This
meant a pretty major refactoring, and that's how I spent the rest of my day.
Tomorrow I'll take a look at the migration path and make sure everything else
Andrew calls out gets addressed.