web explorer fixes

[kbagstad]

@fvilla @bvoigt @lambdatronic (cc'ing Brian and Gary so you're aware of this list too, plus Brian needs to include storyline elements for Tanzania and Veracruz water models)

General issues

• Currently, clicking on the results of a probabilistic output gives the error “org.integratedmodelling.corescience.implementations.datasources.MemObjectContextualizedDatasource cannot be cast to org.integratedmodelling.modelling.data.CategoricalDistributionDatasource” – and for most models, no results are visible either. Clicking on results here should give you an ouput value plus uncertainty, if I’m not mistaken. Also I’m not sure whatever became of the uncertainty displays – is there any kind of uncertainty overlay available at this point?
• Update the ARIES logo to the current color scheme – in the upper left of the explorer and in the “storyline is computing” status box. Also for the status box that tells you when it’s computing or completed, the color scheme of a light gray box, green text when the storyline runs to completion, and smaller white text below it are very hard to read – please revisit this color scheme.
• You can’t fully read the text in the storyline drop-down menu on the left. I know you can mouse over it and see full text but it’s not the same. Also the text boxes on the right cut off a little bit of text at the end of each line. There are also problems with text being cut off or boxes being the wrong size to contain text when you initially select a storyline and the credits are too long for the box in the lower right.
• The “+” button to add a storyline has two sentences that run together and need a period.
• Remove the period at the end of the sentence on the geotiff export button for consistency (none of the other similar buttons have a period at the end).
• On the bottom bar of the explorer, when the “name” is too long part of it is cut off. Maybe make the bar wider to accommodate longer names? Also on the bottom bar of the explorer, what about having a slider bar? That would help in navigation for storylines with lots of results, beyond just the buttons to scroll to the right or left.
• When you switch between data over image and contour view, the color legend below the map sometimes gets stuck on one of the two legends even if it’s the wrong one. Try toggling back and forth a few times to see what I mean.
• For some reason, “rivers” layers are not making it into the storylines. I have no idea if the storylines are written wrong there or if it’s just the usual persistent GeoServer unreliability with it skipping over results and not passing them to the flow models.
• What’s the story on the NLCD layers overlap problem (github issue 32 – I never got the story on whether this was an easy or a hard fix)? This will be a bigger and bigger deal since I’ve enabled carbon analysis across wider spatial extents.
• What’s the story on intelligent layer prioritization (0-10)? This is needed to expand spatial contexts for some models.
• Some discretizations that appear in the explorer are wrong – for example, in the Puget carbon model, the fire discretization should be 4 discrete classes, not 3; San Pedro carbon model carbon sequestration and storage discretization should have 6 classes, not 5 (and the numbers are wrong when you mouse over them), Mg carbon model deforestation should be 4 discrete classes, not 3, etc.

Carbon storyline

• When I try to do a carbon assessment for Colorado, I occasionally (but not always) get two “carbon sequestration and storage” play buttons/add/remove storyline boxes in the drop-down menu on the lefthand column.
• Why, when I run a carbon assessment for the San Pedro, does the color ramp not appear for vegetation type (but when you click on the map the correct vegetation cover is indeed identified)?

Aesthetic viewshed storyline

• For “mountains” in the aesthetic viewshed model, it says that values are in meters, but when you click on a point to get the elevation, it doesn’t provide it, nor does the color ramp display values for elevation.
• For housing value, there’s a single zero after the decimal point (i.e., 0.0 – 10000.0). Can you change this to read (0 – 10,000)? This might involve making $ a legal unit of measure in thinklab.

Fisheries storyline

• In “runninghead”, using underscores before and after a name (as in the 3 fish species) does not properly italicize the names.
• For some reason, “total harvest” gives a text box in the upper right that has both vertical and horizontal scroll bars. Please remove the horizontal scroll bar, as it’s distracting and for some reason doesn’t show up in any other storyline. Also, in the total harvest, when I click on a point on the map, it gives results down to 4 decimal places. This significant figures problem is clearly going to bother some viewers – can we display results here in integer values? Similar significant figures problems arise in the flow model results for fisheries.
• Something strange is going on with the “paths” output – which model –d’s properly but did not show up in the explorer properly. This is worth a further look maybe the same problem with the “rivers” layers which don’t seem to be properly making their way into the explorer?).

Coastal protection storyline

• For some reason, for some of the rasterized line features (wind speed, atmospheric pressure), there are two “data over imagery” buttons rather than “contours over imagery” (i.e., there’s no rainbow icon)
• In the coastal protection storyline, why did you initially put in both LivesAtRiskStorm/CycloneDependentLivesAtRisk and the corresponding pair of assets at risk concepts in the storyline? These seem redundant, so one should be removed. Agree?

Water supply storyline

• What’s the “surface-flow” folder and storyline doing in the water supply storyline? Is it redundant, and if so can you delete?

Sediment regulation storyline

• Two play buttons show up for sediment regulation for farmers – not sure why this is the case or how to fix it.

[kbagstad]

Two general interface issues raised by stakeholders in Puget Sound at our meetings last week, where I demo'ed ARIES carbon, which for the most part went smoothly:

1. The eyeball icon that you click on to highlight which parts of the world have coverage for that particular service is kind of a non-intuitive choice for what it does. Plus it looks kind of creepy when it turns red. What about using a magnifying glass icon instead (everyone who's ever zoomed in knows what that does)?
2. I think I finally understand why a large number of people who look at the storylines think that ARIES is just a GIS data viewer. While I stressed the importance of carefully reading through the whole set of maps included in each storyline, it would be helpful if the outputs were somehow grouped/labeled as input data, source/sink/use outputs, or flow outputs (this is in the thumbnail maps at the bottom of the storyline explorer). If you have other ideas on how to make this more obvious that would be great but I think this is one way.

[kbagstad]

And one other: how complex to include a feature that notifies the user when their results are ready? If it's simple, perhaps good to include with the beta, if not, then save it for later. But for the big, computational-intensive flow models, we may need to better communicate with the user not to expect their results in 30 sec (common sense for us, but users may be impatient and I think user notification is the best way to ultimately get there).