search

nptscot / nptscot.github.io

Network Planning Tool for Scotland: front end.
https://www.npt.scot
GNU Affero General Public License v3.0
7 stars 5 forks source link

Street space UI #176

Open Robinlovelace opened 5 months ago

Robinlovelace commented 5 months ago

Hot on the heels of #175 we're now in a good place to think about a "Deliverability" section in the UI accordion.

We will need at a minimum:

This UI element isn't as worked out as #175. See https://github.com/nptscot/npt/issues/370 for in-progress work on the data side.

Robinlovelace commented 4 months ago

Heads-up @mvl22 discussed this today and have good understanding of requirement, in first instance a red-yellow-green categorisation can be enough. Input data coming soon.

mvl22 commented 4 months ago

Input data coming soon.

Is there an ETA on this? I am conscious of your launch timescale.

This will be very exciting data to have available. Really looking forward to seeing this.

Robinlovelace commented 4 months ago

Is there an ETA on this?

Not right now, good question, will try to get you some test data ASAP.

mvl22 commented 4 months ago

So I can prepare for its imminent arrival:

Robinlovelace commented 4 months ago

Is this is a single-layered dataset, i.e. will end up as a single dropdown presenting options, with internal layer name 'WidthCategory'?

I don't think we're using CamelCase in other names, correct me if I'm wrong. Not decided a name but how about street_space?

Options are: "Wider than minimum", "Wider than absolute minimum" and "Not enough space", "Plenty of space"

No, those will be the values. The dropdown will be for different categories like:

  • 2 x UniDirectional Segregated (4 m minimum, 3 absolute minimum)
  • 1 x bidirectional (3 m minimum, 2.5 absolute)

The former should be the default. We may want to add other options later as per https://github.com/nptscot/npt/issues/370 but just 1 dropdown with 2 options should be fine for now.

Robinlovelace commented 4 months ago

Other options like "On street parking" and "Remove a lane for motor traffic" may be best as switches. For reasons I won't go into here we cannot be sure we will implement yet though.

This is very useful for thinking through the data side btw so thanks for asking Martin.

mvl22 commented 4 months ago

I don't think we're using CamelCase in other names, correct me if I'm wrong. Not decided a name but how about street_space?

Correct. That is preferable.

No, those will be the values.

Sorry, yes of course. I had vaguely misread this as being boolean fields, but of course these are categories in a single field, yes.

If you can confirm, when finalised, the values and their associated labels to be used, that will help shortcut things.

Robinlovelace commented 4 months ago

Confirmation in parallel thread with Zhao I think. Maybe worth waiting until you have the pmtiles.

Robinlovelace commented 4 months ago

Lots of options above, just a Road width categories (2 unidirectional cycle lanes with light segregation) should do, if that makes sense.

mvl22 commented 4 months ago

Attached is a screenshot of what I have so far. Do you want this pushed live?

streetspace

You are asking for a drop-down with two options:

but then say:

'just a Road width categories (2 unidirectional cycle lanes with light segregation) should do'

I'm not really clear on exactly what is wanted here, to be honest.

Currently the data only has one field, road_width_category_2_unidirectional, containing the four values shown in the screenshot, so I can't see anything about the suggested infrastructure that is achievable here.

I think if we have a drop-down for filtering stuff based on an interpreted value, that interpretation should be present as a field in the data, so that you express the if/then logic in the R side, rather than the UI have to work that out. The UI really ought just to be a viewer of data rather than do any direct computation, as that implies the data itself is not a complete set.

Robinlovelace commented 4 months ago

This looks great @mvl22, yes please.

Robinlovelace commented 4 months ago

Ready to go on this @mvl22 ?

mvl22 commented 4 months ago

Ready to go on this @mvl22 ?

Yes, the commit seemed to go missing but I've resolved the merge conflict over the last 15 mins or so. Should be a PR present now.

joeytalbot commented 3 months ago

The colours and the order of the categories in this legend are very confusing. If you swap "Not enough space" and "Wider than minimum" it would make more sense.

Also, what is the difference between "minimum" and "absolute minimum"?

image

mvl22 commented 3 months ago

The colours and the order of the categories in this legend are very confusing. If you swap "Not enough space" and "Wider than minimum" it would make more sense.

I found this confusing too.

My assumption was that the categories mean:

but I may be wrong. Certainly it's not clear what 'minimum' actually stands for here.

Robinlovelace commented 3 months ago

They should at least be in the right order, can you do that?

The labels

. Labels can be improved but that should be in conversation

My assumption was that the categories mean:

* Absolute minimum width of a street for cars, e.g. narrow historic lane

* Normal size for cars

* Not enough space to add cycle tracks

* Plenty of space to add cycle tracks

No, the categories all related to space for cycle infrastructure (in this case 2 uni directional cycle lanes with light segregation, one on either side of the road):

Does that sound good and make sense? Agree that it's not well covered in https://github.com/nptscot/npt/issues/370 and we covered it quickly during call.

Robinlovelace commented 3 months ago

Cc @mvl22 if the order is a quick fix please do so as low hanging fruit.