search

ualbertalib / can-link

Front end react app for CanLink project
1 stars 0 forks source link

Design follow up: adjustments and suggestions #22

Closed CarlsoFiorention closed 4 years ago

CarlsoFiorention commented 4 years ago
07 bubbles 06 bubbles 05 bubbles 04 treemap 03 wordcloud 02 map 01 home page
CarlsoFiorention commented 4 years ago

The number of hours to follow up these design adjustments may depend on how many iterations, clarifications or new sketches it requires after the fixes are done. I'd prioritize most of the design hours left to follow up these issues after providing widget scenarios to James.

jchartrand commented 4 years ago

There are a lot of things described in this issue, and I'd suggest that they should be broken out into individual GitHub issues so they can more easily be individually estimated and prioritized, as far as that is possible.

Even so, for some of them, trying to match the original design exactly will be a prohibitive amount of work. I used something called material-ui which is a library of components that makes it significantly easier to build a UI. This is what let me get the app up and running in the given time ( I wouldn't have been able to otherwise). Material-ui makes choices for you up front (matching industry best practices for spacing, layout, etc.). I can override any part of it, but a big reason why so many people use material-ui now (you'll recognize it's styling because it is so commonly used now) is that the fiddling that usually goes along with trying to layout/size/align/etc. a ui can consume enormous numbers of hours. material-ui also out-of-the-box handles things like resizing components as the screen is resized, resizing for phones/tablets/etc, providing info to screen readers, etc.

For the visualization changes, anything is possible, but time prohibits doing anything not easily allowed by existing visualizations. And trying to figure out what visualizations do allow can sometimes be a huge rabbit hole down which one might never return. I do think I can make some of the changes, but can't with certainty say all. A significant amount of the work won't necessarily be making the changes but rather finding packages that will allow these changes. And so, it is worth prioritizing these changes, e.g., is the colouring more important than showing the number of items in a visualization when clicked?

Nevertheless to guesstimate how long it would take to make all these changes, a very rough guesstimate is two full weeks of work (80 hours).

sfarnel commented 4 years ago

Items now covered by issues 23-28, 30; closing