Issue Number | 16 |
---|---|
Summary | [Full Citation] Visibility of Decisions on Full Citation Page |
Created | 2013-09-17 09:38:22 |
Issue Type | Improvement |
Submitted By | Juthe, Robin (NIH/NCI) [E] |
Assigned To | alan |
Status | Closed |
Resolved | 2013-11-20 21:57:47 |
Resolution | Fixed |
Path | /home/bkline/backups/jira/oceebms/issue.113301 |
TIR #2496 created 2013-04-05 by Robin Juthe
All users should have equal visibility to all sections of the full citation page. For example, Board managers should be able to see medical librarian decisions. (This can help to explain why certain articles have a topic in our queue without reject/pass boxes available.) Margaret is able to see this as an admin user.
Waiting for feedback from Dan on historical background for current behavior.
According to Dan, his understanding of the requirements for the full citation page was that the buttons for the medical librarian decisions should be suppressed when the current user was a board manager. He says that the fact that the librarians' decision history is also suppressed is a bug. The original requirements spreadsheets are more consistent with the description in this ticket ("all users should have equal visibility to all sections of the full citation page"). Dan thought he might have been working more from the visual design mockups than the requirements document in this case, but Ashleigh only created a single PDF for the full citation page, and Cynthia is the logged-in user for all screens illustrated in that PDF document, so that couldn't have been the source for a requirement that the librarian's buttons be suppressed on the full citation page. So we first need to clarify whether Dan's understanding was correct (in which case we'd modify the software to display the librarians' decision history for all users, suppressing the buttons in that section if the current user is a board manager) or if there should be no difference at all based on who's logged in (except, of course, that the ability to see the page at all is restricted to users who are not board members). I think the second option ("all users have equal visibility ...") will be the easier of the two to implement. Are there any other portions of the page which are suppressed inappropriately based on the current user's roles?
Robin: "Dan's understanding is correct."
Figuring that the first thing to do was find out what was present and
what was missing from the screens presented to different user roles, I
did some comparisons. The records I looked at were legacy data on DEV.
I should repeat the tests with data created in the current system.
However here's what I've discovered so far:
Margaret (icrdb manager) and Cynthia (medical librarian) had
identical screens - same information displayed and same button
icons.
Robin (board manager) was missing the following:
From the MEDICAL LIBRARIAN INITIAL REVIEW section:
The name of the person making a topic assignment (Cynthia).
The date when the assignment occurred.
All of the action buttons:
tag
edit
add new
add new comment
reject active
From the PUBLISHING section:
All of the action buttons.
Does anyone know of anything else missing from the board manager
screen, or anything missing from the medical librarian screen?
Based on what I've found so far, the easy call is that we need to give
the board managers the ability to see the topic assignment person and
date information.
The harder call is what, if anything, to do about the action buttons.
The comments above this one might indicate that Robin should not see
Cynthia's buttons and Cynthia should not see Robin's. But that doesn't
appear to be what we have.
Personally, I like the idea of giving people more buttons rather than
less since the only people using these screens are long standing
dedicated users who know what they're doing. However if having extra
action buttons might lead to inadvertent mistakes then maybe they do
need to be restricted.
If I don't hear from anyone before I start programming (maybe next
Tuesday?), I'll work on giving the board managers access to the topic
assignment person and date. I won't do anything about buttons until
someone decides what we ought to do.
I'll also need any info about differences between screens that I haven't
seen yet.
The comments above this one might indicate that Robin should not see
Cynthia's buttons and Cynthia should not see Robin's.
I think the earlier comments say that the buttons are as they should be, and that the only change that needs to be made is to show decision history consistently for all users with access to the page.
I've checked in a version of the code with the following changes:
Board managers can now see everything that librarians and board managers see.
Board managers do not see librarian's buttons.
A small amount of code was refactored.
A large number of new comments were added.
I have not modified the librarians display. I think the code I have implemented may be able to handle that using the new mechanisms I installed, but I didn't want to try it until I've got these revisions safely tucked away.
The code is running on DEV.
I am assuming that anyone who can get to the full citation display will be able to see everything that happened to the citation. However we need to make a definitive determination of who (i.e., what roles) should see what buttons.
Buttons are in four groups. I'm treating each group as a unit. A user will have access to all buttons in the group or none of them. The groups are:
REVIEW CYCLES
IMPORT PROCESS
MEDICAL LIBRARIAN INITIAL REVIEW
PUBLISHING
Roles are:
admin assistant
administrator
anonymous user
authenticated user
board manager
board member
icrdb manager
medical librarian
site manager
I'll work on another task while we wait for a final decision on these.
Robin and I met on Tuesday to go over the status of this issue and
to
address some outstanding questions regarding it. For our meeting I
printed out a list of users and roles in the system from the QA
database.
The main question for the meeting was who should see action buttons
on
which parts of the display based on their roles. Going over the
roles
that we have and the people in those roles, Robin decided that we
should
implement the changes that I have already made and no more.
The changes I made were to show the Board Managers all of the data
now
shown to other users but to remove the action buttons for those parts
of
the display that Board Managers don't use on the full citation
screen.
I did not change anything for librarians. They still have access
to
some buttons which they will seldom if ever use.
The decision was to leave things that way.
In making the changes I modified the machinery that determines, for
each
part of the screen, what type of user should see action buttons.
There
is now just one place that tracks what part of the screen is being
built
and one place that makes decisions about whether the current user
should
see action buttons on that part of the screen.
I modified all of the routines in the program to make the new
"display
phase" information accessible but only changed the logic of some
of
them - the ones that were not right before. However, if we need to
revisit
this issue and change the behavior again, I think it should be a
lot
easier to do.
Robin also asked whether we could display the name of the person
who
made a publication decision. The display currently shows that an
article was "published" (i.e., moved from librarian to board
manager
control), and the date, but does not show who made the publication
decision.
I looked into this and found that the information Robin wants is in
the
database. Putting the information on the screen will not be
terribly
hard but not trivial either. The information is manipulated in a
number
of different places that pass data around to each other so I
tentatively
think that at least three routines and several data structures have
to
be modified. There's also a bit of logic involving "publishing"
and
"republishing" that I'd have to investigate.
The latest version of the code is in svn and in operation on Dev.
So, while I don't think this is real hard to do, it's not trivial
either
and the changes are totally unrelated to the other changes I've made.
I
therefore recommend that we create a new issue for this if we decide
to
move forward with it.
I see that a hyphen in my text fell at the beginning of a line in my last comment and got interpreted as a format specifier by JIRA. Sorry about the faux formatting.
Ready for user testing.
The visibility looks good for Board managers, Margaret, and medical librarians. Technically speaking, Bonnie should not have the medical librarian buttons for initial review and publishing (same as Board managers), but I am fine with keeping this as is since it doesn't hurt anything for her to have these extra buttons. Also, I think if we were to change this we would have to alter the visibility tied with the ICRDB manager role, and we wanted Margaret (the ICRDB manager) to have the ability to do anything (this is where the user roles get a little messy). So, I'm marking this QA verified. Thanks!
I see that a hyphen in my text fell at the beginning of a line in my last comment and got interpreted as a format specifier by JIRA. Sorry about the faux formatting.
You know that you can edit your comments, right?
This was verified on DEV.
"You know that you can edit your comments, right?"
I didn't. But now I do. I fixed it.
It makes me want to go back and fix others, but I'll stifle the urge.
Verified on prod. Closing issue.
Elapsed: 0:00:00.000828