worldbank / covid_dashboard Goto Github PK

I know the rectangle have a name, but I'm blanking :-)

I'm trying to limit the display to LICs. I've turned off the other income groups in the legend and removed all "compared" countries, but it still leaves a residual annotation line as shown here. As a result, the Y axis also doesn't scale correctly.

I can work around it by specifically selecting an LIC country, but it's not very intuitive, and is further confusing because the chart resets the legend whenever I change the country selection or any other dimension.

Is there any way to improve the user interface a bit?

1. Remove the phantom annotation line if no countries are selected
2. Avoid resetting the legend when user changes other dimensions

This is to be done on a separate branch.

@tonyfujs if you're able to look into this. sorry to add this to your plate!

From William Oliver via email:

Could you let me know who is/are behind the DECDG COV-19 dashboard? It’s really great, I just had a suggestion: enable log-scale on the y-axis to allow users to see the trajectory/rate-of-change of confirmed cases or deaths over time more clearly.

Defined as Deaths / Confirmed Cases

Not sure how easy / difficult this would be, but having the bar chart adjusting automatically when drilling-down on the treemap would be a very nice feature

@tonyfujs FYI

Rollovers often show "NA" especially after changing the indicator selection

@tonyfujs @randrescastaneda all of us can give it tries?

Here's the information you need for the project map:

API URL: http://search.worldbank.org/api/v2/projects?format=json&rows=1000&qterm=covid

Key Fields:

• id
• project_name
• countryshortname
• countrycode (2-character - is that okay?)
• status (maybe color-code the bubble?, blue=pipeline, green=active, brown=closed)
• boardapprovaldate (will be missing if status=Pipeline)
• totalamt
• url = https://projects.worldbank.org/en/projects-operations/project-detail/{id}

Numbers could be easier to read with spacing every thousands

Related to Vinny's ticket #45 on improving the scatterplot legend, I'm just wondering if we even need it. How relevant are Bank regions in this context? My sense is: "somewhat relevant" but not proportionate to the amount of space used by the legend.

What if instead we added the region names to the rollovers and removed the legend?

As Umar suggested: can it show case data for selected countries in addition to or in lieu of the global totals? As Div, noted, there are some UI issues if user selects a bunch of countries.

Another option could be to add a 4-column table and show it there:

Loading the dfm object is currently the main bottleneck performance wise. Maybe some columns / rows can be removed?

I just refreshed the data, but life expectancy is still not showing up...

@randrescastaneda Can you take a look at this if you have a bit of time? (I will be mostly unavailable for the next few hours)

1. Can you double check that the life expectancy indicator is indeed missing? I looked quickly, so I may have missed it
2. Can you try to figure out why this is the case? There is a list of hard coded indicators at the beginning of the 01_get_data.R script. My guess is this is the issue...

Thanks!

cc @DivyanshiWadhwa

Shouldn't this be aligned? Same indicators above and below?

Currently, the title is always the same. It should update depending on selected indicator.

Per capita and other normalized indicators don't make sense in a tree map

cc @randrescastaneda

The colors of the lines on the line chart seem to be assigned randomly... Please see below a screenshot taken a few minutes ago. I'm pretty sure those were not the colors that showed up for me yesterday...