Just had a hot red bean bun for breakfast! I haven’t had one in over two years, and to top it off, it was packed with koshian. One of the perks of my otherwise doldrums commute is passing by Famima! convenience store in Union Station. For ages, I’ve craved c-stores as they are back home in Taiwan. 7-11 in the US doesn’t count. It’s catered to the American market but fails to bring any of the bentos or prepared foods which make c-stores so great. Family Mart’s American business foray, Famima! is the closest approximation I’ve seen to date. It was nostalgic to grab a rice triangle with umeboshi or spicy curry shrimp as a quick bite before work, and some cha xiu baos after work.
In any case, I read up on their US business strategy and some adjustments they’ve had to make to survive in the American market. I noticed some half-baked social media attempt (twitter fail) but overall a focus on the brick and mortar establishment. They’ve been struggling, so I wonder if they’ve floated the idea of a food cart style c-store.
I would build niche specific mobile c-store food carts. For instance, they could have one that operates at night in the bar districts. They could have ones for business districts where there are a lot of commuter throughfare. Or even better, make them modular between the two. This could reduce operating costs and allow them to make faster adjustments to inventory.
Yesterday I did a QC on specific Tidemark fields to ensure that various geographical layers correctly aligned with the fields in question. For instance, I took a look at the zip codes layer and selected by location, then did a query to remove the correctly matching entries. If there were any entries that were within the zone but not marked as such, then I exported it as a separate table for the Planning Dept. to correct in the Tidemark data. I did that process for zip codes, historical registry districts, council districts, and regional districts.
That took up most the morning, then after lunch I started a project to create a new field in the building footprint layer that concatenated contiguous footprints. Our building footprint layer is derived from aerial data, which actually gives us roof lines. So in effect, the result is multiple contiguous “footprints” that belong to the same building. However, we didn’t have a field that identified each “footprint” as belonging to the same building, even though they were clearly touching. Jonathan and I discussed various methodologies and ultimately settled on one which we successfully tested on a small set and will fully actuate tomorrow. This process involved the “feature to point” tool to find the centroid of each footprint, the “dissolve” tool to concatenate contiguous footprints, adding new fields and populating them with the field calculator, and spatial and attribute joins to bring the new merged building field into the original multi-part building footprint layer. It was quite tricky to grapple my mind around that late in the day, but I managed to replicate the results and write down the methodology prior to heading home and going for a run.
As a side project, I’ve started looking into ways to make GIS more ADA accessible. For blind people, this is quite difficult, but I was thinking of starting with color blindness. Apparently, there are various forms and each alters the perception of certain wavelengths in a completely different way. This can be extremely problematic for some who see green and orange as the same shade, and are trying to interpret what seems to 90% of us as a perfectly sensible map legend.