10/21/2021-Meeting Minutes === ###### tags: `Templates` `Meeting` `geoplanar` :::info **Location:** Online Meeting **Date:** Oct 21, 2021 4:00 PM **Agenda** Discussion on historic HOLC maps and [geoplanar](https://github.com/sjsrey/geoplanar) ::: :pencil: Discussion Notes ---- Discussion on the direction and areas for improvement for [geoplanar](https://github.com/sjsrey/geoplanar). #### Round 1 Discussion:(Enhancement of the Repo) - We should define the scope of `geoplanar` what can be done in the software and its limitation. - From US-Mexico and US-Canada examples we can see that `geoplanar` cannot distinguish between land features versus actual data issues. It can only identify the holes, the researcher needs to cognizant of these cartograpic issues. We recommend thoroughly examine the output before fixing any issue to ensure accuracy. - ==Additional Direction== We can add terrain layers using contextily basemaps. #### Round 2 Discussion (HOLC Maps and geoplanar) - To what extent do modern tracts coincide with the boundaries of the HOLC maps. - Applying `geoplanar` on this data can be useful to find out missing holes and planar enforcement violation. - This will help in testing `geoplanar` on a newer data set. - Since these historic HOLC maps have been used in a number of studies finding data discrepancies can be a major contribution. :dart: Meeting Targets Accomplished --- - Access granted to historic HOLC maps repo for California. - Received feedback/Comments on PR for `geoplanar` repo. - Discussion and direction to take for collaboration on `geoplanar`. :closed_book: Tasks -- - Look at the HOLC data Repo and try to play around with the data a bit more. - For `geoplanar` generate additional notebooks using different countries. :question: Additional Questions --- - What data to use for HOLC `geoplanar` ? HOLC/SAB or HOLC/Census - Do we look at holes, try to fix them? :speech_balloon: Comments --- See the repo for Oakland and Sacramento: https://github.com/WawNun/geoplanar/tree/main/notebooks/HOLC 10/28/2021-Meeting Minutes === ###### tags: `Templates` `Meeting` :::info **Location:** Online Meeting **Participants:** **Date:** Oct 28, 2021 4:00 PM **Agenda** [geoplanar](https://github.com/sjsrey/geoplanar) ::: :pencil: Discussion Notes ---- - Do we need to check if HOLC aligns with census data? - Identify, quantify the violation and how can that be fixed. - Overlays for HOLC its not the same problem at country boundaries.For HOLC we have to look at two separate layers. The first it the HOLC maps layer and the second is the census layer. - How many times do we see the planar enforcement problems in the HOLC maps and do they matter in terms of statistical errors.From this prespective working on this problem can be a major contribution. - Misspecified connectivity of the boundaries can be an issue for a variety of data including census and maps. > [name=Participants] > Everyone noted that we have seen issues where the polygons visually exceed the map layers. This is a sign of potential planar violations.For example, sharpie used to draw a boundary can make the polygon appear much bigger. Research using these data sets are mostly unaware of spatial limitations and the results produced by these gdf can be spurious. > [color=#efad4a] :dart: Accomplished during Meeting --- - Address all questions and confusion on `geoplanar`. - Agreement on what are the next step to explore HOLC maps data in `geoplanar`. :closed_book: Tasks -- Geoplanar works with the same layer so we will have to check for data issues one at a time. **Step 1: HOLC Layer** - Check for overlaps and other planar violations in the HOLC map boundaries (look at **[notebook](https://github.com/spatialucr/holc-narsc/blob/main/notebooks/100_data.ipynb)**). **Step 2: Geoplanar with Census Data** - Check for overlaps and other planar violations using the census tracts boundaries. **Additional Pointers** - Bounding box for each holes, functionality add contexitly so we can zoom into the holes and other issues. 11/04/2021-Meeting Minutes === ###### tags: `Templates` `Meeting` :::info **Location:** Online Meeting **Participants:** **Date:** Nov 3, 2021 4:00 PM **Agenda** [geoplanar](https://github.com/sjsrey/geoplanar) ::: :question: Questions --- #### COUNTRY SPECIFIC DATA - What does it mean to have layers on a map?What is the technical definition of a layer. If we use two different datasets for shapefiles are we essentially applying multiple layers?For example in Italy/France data if I used only NUTS I dont see any violations but if I use NUTS for Italy and another shapefile for France there are violations. Is this a the right approach to test `geoplanar`? - How do I convert data from two separate sources into a single layer? Reference [notebook](https://github.com/WawNun/geoplanar/tree/augment/notebooks) #### HOLC SPECIFIC DATA - Difference between contiguous and non-contiguous using HOLC data. Go over detecting nonplanar edges from the Oakland Notebook - I tried to fix the nonplanar edges it still comes out as false. Reference [notebook](https://github.com/WawNun/geoplanar/blob/Oakland_HOLC/notebooks/Oakland/Oakland_geoplanar-1.ipynb)