Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
If you read MDAllstar's followup, the algorithm does now take road infrastructure and how it affects pedestrian access into account.
That's good to know, but I have a strong feeling it does NOT take roadway width into account, which seems like a common sense aspect of walkability to include.
I think that has changed recently. Now roads impact the score I thought.
“Walk Score also measures pedestrian friendliness by analyzing population density and road metrics such as block length and intersection density. Data sources include Google, Factual, Great Schools, Open Street Map, the U.S. Census, Localeze, and places added by the Walk Score user community.“
Better, but still no mention (or inclusion?) of sidewalks, street width, mid-block pathways...
Surprised no locations in Tacoma (the Stadium District, Hilltop) make the cut
Nice, thanks for doing that! I'll add that in the other topic later if you don't mind.
I think there are at least two factors that are generalizable and may have lead to Tacoma not making the list.
One is that the neighborhood definitions they use aren't always comparable from place to place, so they can have boundaries that can have a mix of both very urban and not very urban parts over a large area which bring the total down. This can also work the other way and bring the total up, but with areas that generally have much smaller highly walkable areas, the tendency would be to clump them with a lot of not very walkable neighborhoods and thus no neighborhood in the 90sn score range. This is definitely happening with Tacoma given the massive size and odd shape of the top neighborhood that includes downtown which they termed New Tacoma.
The other is that it's not clear on what schedule the underlying data sources used to compile the walk score are updated, so it's quite possible that perhaps when they did their last update, the neighborhood outline for Tacoma didn't make the cut, but had they done an update of that underlying source more recently, it would.
My problem with this method is that it doesn't show how walkable a city is as a whole. You can have a neighborhood that's walkable but isn't the most dense population wise. And you can have neighborhoods that are extremely concentrated but the density drops off faster than other cities with less people living in their 90+ walkscore neighborhoods. Also a city might have way more neighborhoods with a walkscore of 80+ despite having less people in the 90+ neighborhoods.
Yea, this was a first run of it, done in a way that's fairly easy to scrape and put together. Ideally, what you'd do is have the results based on the heatmap rather than the neighborhood definitions, and do a weighted score that favors higher walk score areas and a much lower cut-off threshold for continuity that *also* takes in transit mode information that factors in frequency, speed, and *maybe* fare cost.
It's a lot of effort to do that though, and I especially wouldn't want to spend that time with walk score as it is now, because some of its underlying data sets appear to have been updated for a while. I think as is, this kind of rough approximation by just using contiguous 90+ score places isn't too bad and roughly gels with my own perception and experience of these places.
I think instead, it'd be better to try to create a transparent, open source version of this that's highly modular so you can find ways to plug and play different data sets and so it's easy to update. I'd want to get at least somewhat paid for it though, even a modest grant, so I've toyed with the idea of applying for them. If anyone has any ideas on that, let me know!
Better, but still no mention (or inclusion?) of sidewalks, street width, mid-block pathways...
Aside from sidewalks, most aesthetic components should be excluded imo. Those tend to measure not whether a walk can be done but how pleasant it might be perceived. And that not only starts becoming subjective, it becomes difficult to measure and quantify. Which is more detrimental for scoring, curb cuts or street widths? Sidewalks or block lengths? We all know a pleasant walk when we see it but quantifying pleasantness is hard and subjective. I get the issues with the scoring, but I tend to take such information fwiw.
Because cities vary dramatically in how much physical area they cover and what portions of the metropolitan area they cover, walk score for city vs city comparisons end up being nigh useless. This is similar to when people say Jacksonville is bigger or more populous a city than DC and Boston. Sure, that's true, but it's also kind of not.
Using a city walk score strictly on municipal legal boundaries is maybe ironic for something that's supposedly ranking attributes that are supposed to be quite fine-grained and on a human scale.
I think other things people have noticed about walk score is that:
- the methodology in terms of sources used seem to effectively short change Canadian cities perhaps due to different sources pulled for US and Canada; it seems to be even worse for Quebec
- some of the data pulled is unfortunately outdated such as neighborhood population counts being at least least several years old and not updated from the 2020 census (2021 for Canada), so it's unclear what the update cycle for the various sources are but it is clear that at least some of them can go quite a while before updates
- it used to go by sheer proximity as the crow flies though I think in somewhat later times it did some route-finding
- it doesn't take into account some elements of urban design though arguably did better when it seemed to have implemented route-finding which would have innately taken into account some of that
Overall, I like walkscore and the methodology and idea, but I think the lack of updating especially after being acquired by Redfin is disappointing. I'd like to do an open source project version of it and thus with parameters that one can tweak.
I also tried to do what I think is the first step in a better way of doing a ranking of cities in terms of walkscore to move away from the obvious issues of doing it by municipal borders. This was to try to go by contiguous clumps of high walkscore areas. The most straightforward attempt was to use a cut off of 90 for neighborhoods that walk score used and sum the population of those neighborhoods as cited in this topic and with the last one done in this post though I stopped updating it given that walk score did not seem to be moving towards updating their numbers anymore:
This is wayyyyy off below for DC:
SW Ballpark - Navy Yard 90 4,091 people There are more than 20,000 people living in that area if not more.
SW Ballpark - Navy Yard 90 4,091 people There are more than 20,000 people living in that area if not more.
This was mentioned many times already. SW Ballpark - Navy Yard isn't even the most egregious example of it especially since SW Ballpark - Navy Yard is nowhere near being the neighborhood unit, as walk score has them, that's added the most people since they last updated. I also doubt the area as they've outlined it, and the way they outline neighborhoods is another criticism, has more than 20,000 people living in it. If the entirety of Ward 6 added about 30K people from the 2010 to 2020 census, then does it make sense that the tiny sliver they outlined takes up the majority of that addition with the over 16K people you're stating? It seems unlikely especially as it's missing all that prime development south of M street near the water.
This is one of multiple issues with walk score I outlined, though I think the unclear and slow update among US cities is probably the most reasonable and now longstanding criticism.
Last edited by OyCrumbler; 11-17-2023 at 01:28 PM..
SW Ballpark - Navy Yard 90 4,091 people There are more than 20,000 people living in that area if not more.
If we were to summarize the threads on the first page that prominently feature DC, would agree that it would be accurate to say that DC has a lot of lonely people walking around trying not to be murdered?
Just kidding and meant to be a joke and not a troll but it's kinda what I'm reading today.
If we were to summarize the threads on the first page that prominently feature DC, would agree that it would be accurate to say that DC has a lot of lonely people walking around trying not to be murdered?
Just kidding and meant to be a joke and not a troll but it's kinda what I'm reading today.
This actually made me laugh.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
