<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Indicatrix - Medium]]></title>
        <description><![CDATA[I’m a PhD student in geography studying urban transportation at Arizona State. - Medium]]></description>
        <link>https://medium.com/indicatrix?source=rss----4045e448107f---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Indicatrix - Medium</title>
            <link>https://medium.com/indicatrix?source=rss----4045e448107f---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 15:23:44 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/indicatrix" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[How transit fares affect accessibility]]></title>
            <link>https://medium.com/indicatrix/how-transit-fares-affect-accessibility-4ddc9237947f?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/4ddc9237947f</guid>
            <category><![CDATA[accessibility]]></category>
            <category><![CDATA[transit]]></category>
            <category><![CDATA[transportation]]></category>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Tue, 30 Apr 2019 15:27:42 GMT</pubDate>
            <atom:updated>2019-04-30T15:27:41.913Z</atom:updated>
            <content:encoded><![CDATA[<p>Accessibility metrics are increasingly being used for planning, with planning agencies measuring what opportunities the transportation network connects residents to. A common accessibility metric is the “cumulative opportunities” metric, which measures how many opportunities can be reached within a given travel time budget—these opportunities could be <a href="http://www.kcsmartmoves.org/pdf/PlanDocuments_2017/JobsAccess.pdf">jobs</a>, <a href="https://doi.org/10.1016/j.apgeog.2014.06.012">grocery stores</a>, or even <a href="https://doi.org/10.1177/0361198118780834">abstract space</a>. Furthermore, accessibility is frequently invoked in discussions of just transportation systems. <a href="https://www.changinghands.com/book/9780415638326">Karel Martens</a> and <a href="https://doi.org/10.1080/01441647.2016.1257660">Rafael H.M. Pereira</a> both argue that just transport systems should provide a baseline level of access to all individuals.</p><p>However, in the case of public transportation, travel time is not the only cost that travelers incur. Transit users in most transit systems must also pay a fare, which makes any access provided by the system meaningful only if the users can afford the fare. In our recently-published article in the <em>International Journal of Geographic Information Science</em>, <a href="http://www.ansonstewart.com/">Anson Stewart</a> and I extend cumulative opportunity accessibility metrics to account for opportunities that are reachable within a travel time budget <em>and</em> a monetary cost budget (<a href="https://doi.org/10.1080/13658816.2019.1605075">publisher version of paper</a>, <a href="https://files.indicatrix.org/Conway-Stewart-2019-Charlie-Fare-Constraints.pdf">open access preprint</a>).</p><p>The key challenge with using transit fares as a cutoff in the accessibility metric is path-dependence. While travel time is not path dependent—it takes the same amount of time to go a few stops on a bus whether you got to the stop via another bus, a train, or a <a href="https://www.youtube.com/watch?v=KCrQu0VTUBs">silly walk</a>—the same is not true of fares. A bus might yield a free transfer to another bus, while the train does not. Knowing whether the bus is better in terms of fare than the train at an intermediate point requires knowing not only the cost of the bus, but also the full journey cost for all journeys that begin with the bus. To produce a tractable algorithm, it is important to be able to discard suboptimal routes within the algorithm, before the computer has spent large amounts of time evaluating them.</p><p>Our algorithm accounts for this difficulty by tracking the “transfer allowance” of each transit option—how much a system user could save on a subsequent vehicle if they access it via a particular option. We can thus discard many journeys that provably cannot result in an optimal trip, while retaining journeys that may. We implemented the algorithm in <a href="https://www.conveyal.com/analysis">Conveyal’s R5 analysis engine</a> (I began this work while a Conveyal employee, though I’ve since moved to Arizona State University, and Anson works for Conveyal now). Case study results are in the next section.</p><p>We’re not the first team to tackle this problem. <a href="https://aag.secure-abstracts.com/AAG%20Annual%20Meeting%202019/abstracts-gallery/19000">A team led by Ahmed El-Geneidy</a> used a similar metric, but relied on a fare computed for only the fastest trip. If there is, say, a fast commuter train paralleled by a slower but cheaper local bus, this approach may find that a destination is unreachable given the fare constraints; the cheaper but slower bus may not be considered. <a href="https://doi.org/10.1016/S0191-2615(02)00005-X">A team led by Hong K. Lo</a> developed an algorithm that can correctly account for fares, but it requires pre-specifying allowable sequences of modes, non-ideal for accessibility metrics since <a href="https://scholarcommons.usf.edu/cgi/viewcontent.cgi?referer=&amp;httpsredir=1&amp;article=1824&amp;context=jpt">one of their key selling points is a relative lack of behavioral assumptions</a>.</p><p>We demonstrate our algorithm with a case study in the Boston area. Boston’s transit agency, the MBTA, has a <a href="https://www.mbta.com/fares">complex fare structure</a>, with flat fares differentiated by route type, zone-based fares on commuter rail, and a variety of discounted transfers. Just as important, the effect of fare policy on low-income travelers’ accessibility has been a salient topic in Boston politics since at least the 1940’s, when <a href="https://en.wikipedia.org/wiki/M.T.A._(song)">the protest song “M.T.A.”</a> was written in support of a mayoral candidate. One verse in the song notes that the subway fare increase will prevent the protagonist, Charlie, from experiencing affordable access to visit his “sister in Chelsea, or [his] cousin in Roxbury.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kWgX6zAanilq3LXI6wuBiA.png" /><figcaption>Pareto frontier, for travel from suburban Norwood to central Copley Square (Figure 1 from the full paper; image credit: Anson Stewart)</figcaption></figure><p>The image above shows the range of tradeoffs between travel time and fare, for trips between suburban Norwood and central Copley Square. (Technically, the image shows a <a href="https://indicatrix.org/pareto-efficiency-and-urban-planning-5cfef767ab6c">Pareto frontier</a>.) The fastest route is to use the commute rail system, and pay a $7.50 Zone 3 fare to reach Back bay then walk to Copley in about 40 minutes total. However, a budget-constrained traveler might instead choose to to take a bus to the subway for $2.25, or a bus to another bus for $1.70, and endure 75–95 minutes of travel time. Or they may choose a middle option: taking commuter rail, but disembarking before Back Bay and transfering to a bus and then the subway, requiring an hour of travel time but costing only $5.50.</p><p>Boston’s rail system consists of a mix of light- and heavy-rail rapid transit, and commuter rail. The rapid transit system has a flat $2.25 fare, while the commuter rail system has a zone-based fare and is far more expensive. While the rapid transit system generally serves only the inner portions of the region, while the commuter rail system stretches far into the suburbs. However, some inner neighborhoods, even within the City of Boston proper, are served by commuter rail and not by rapid transit, producing a fare differential for travel from neighborhoods that are similar in terms of function in the metropolitan structure. Thus, we compare the current fare system to a system where travel between all Zone 1A, 1, and 2 commuter rail stations (the innermost zones) costs a flat $2.25. We do not claim that this scenario is sound policy or would have beneficial ridership or fiscal impacts; it is only intended to demonstrate the characteristics of our algorithm.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5pxVMDViCRK9SYtpaa38fw.png" /><figcaption>Change in job accessibility with 60 minute travel time budget and $5 fare (Figure 7b from the full paper; image credit: Anson Stewart)</figcaption></figure><p>The map at left shows the effect this scenario would have on accessibility to jobs with a 60 minute travel time limit and a $5 fare limit. Note that this scenario makes no changes to the transit schedules; all changes are due to lowered fares making more jobs accessible within the $5 fare limit. The largest increases are near commuter rail stations experiencing lowered fares (e.g. near Dedham, Newton, Lynn and south of Woburn), because of cheaper access to centrally-located jobs in Boston. However, there are also increases far from commuter rail, due to network effects—it is now possible to ride another form of transit to access commuter rail and then travel via commuter rail to both downtown and suburban jobs, without exceeding a $5 fare. This demonstrates the value of using accessibility metrics to evaluate fare policy—the impacts extend far beyond the stations affected.</p><p>If you’ve made it this far, you may want to read the full paper. It’s available <a href="https://doi.org/10.1080/13658816.2019.1605075">from the publisher</a> (paywalled), or <a href="https://files.indicatrix.org/Conway-Stewart-2019-Charlie-Fare-Constraints.pdf">for free on this site</a>, without the fancy journal formatting.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4ddc9237947f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/how-transit-fares-affect-accessibility-4ddc9237947f">How transit fares affect accessibility</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Identifying High Quality Transit Corridors]]></title>
            <link>https://medium.com/indicatrix/identifying-high-quality-transit-corridors-a8e4eec37ed8?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/a8e4eec37ed8</guid>
            <category><![CDATA[housing]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[zoning]]></category>
            <category><![CDATA[public-transit]]></category>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Sat, 13 Jan 2018 06:12:33 GMT</pubDate>
            <atom:updated>2018-01-13T06:23:35.534Z</atom:updated>
            <content:encoded><![CDATA[<p>In <a href="https://indicatrix.org/will-rezoning-around-transit-stops-in-california-undermine-support-for-public-transit-883e69aaab2a">my previous post</a>, I analyzed the effects of a proposed bill in California that would dramatically upzone areas around transit stations, allowing for increased housing production, albeit potentially creating opposition to transit projects. In that post, I presented a map showing the area potentially affected by the bill, and I promised a further explanation of the methodology used to create it, which I present here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*11IwaXRU0IqrFWdL-BwAgQ.png" /><figcaption>Areas in Los Angeles potentially affected by SB 827. Data © LA Metro, Metrolink, OpenStreetMap contributors.</figcaption></figure><p>The code used in the analysis is open-source and <a href="https://github.com/mattwigway/sb827-analysis/blob/master/nb/GTFS%20Map%20Matching.ipynb">available here</a>. It is written in Python 3, using a Jupyter notebook.</p><p>The bill establishes certain requirements for zoning codes in areas within 1/4 mile of a “high quality transit corridor” and within 1/2 mile of “major transit stop,” in order to prevent low-density zoning in areas well served by transit. High quality transit corridors are defined as fixed-route bus service coming at least every 15 minutes at peak. Major transit stops are all rail stops, ferry terminals that connect with another mode of transit, and places where high quality corridors cross. I needed to identify each of these pieces, and then buffer them to create the final map.</p><p>Finding rail stops was easy. I downloaded the GTFS feeds for <a href="https://gitlab.com/LACMTA/gtfs_rail">LA Metro rail service</a> and the <a href="http://metrolinktrains.com/content/google_transit.zip">Metrolink commuter rail</a>. I then loaded them into Python using <a href="https://github.com/araichev/gtfstk">gtfstk</a> and extracted all stops that had service. There is no ferry service in Los Angeles, to my knowledge, so I did not include that here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gEAhSaG8vOHbZd8jxQN5KA.jpeg" /><figcaption>Many lines overlapping to provide high frequency service in Brisbane, Australia (author pictured for scale).</figcaption></figure><p>I then needed to identify the high quality transit corridors. This proved to be much more difficult. The first question is the definition of a “corridor” — if several routes overlap on a common segment and provide service at least every 15 minutes, is that considered a corridor? I felt that it should be, as many transit agencies explicitly plan this by operating “branching” service, where suburban bus lines come together to form a high frequency spine through the densest part of the city (<a href="https://translink.com.au/">Brisbane, Australia</a>, for example).</p><p>Detecting this from GTFS data, however, is not necessarily trivial. These vehicles may not share the same stops along the corridor (and with real-time arrival information, it is relatively easy for a user to determine which vehicle will arrive next, and go to the appropriate stop, assuming they are nearby). This is particularly true of express or skip-stop service (for instance, most Metro Rapid routes in Los Angeles); one route stops at all stops, whereas another stops only at a few of those stops.</p><p>In order to find corridors, I snapped all of the LA Metro’s bus routes from their GTFS feed onto the street network using the map-matching capabilities of <a href="https://github.com/valhalla/valhalla">Valhalla</a>. First, I identified <em>patterns</em> in the GTFS feed, which are unique sequences of stops visited by a vehicle. This is an optimization step; rather than individually snapping every trip to the street network, I snapped each pattern to the street network, assuming that all trips that visit the same stops will use the same streets.</p><p>I loaded an OSM extract of the LA basin from <a href="https://mapzen.com/data/metro-extracts/">Mapzen Metro Extracts</a> into Valhalla, and then used the trace_attributes API to snap each pattern onto the OSM network. Where possible, I used the GTFS shape data; if this was not available I generated a geometry using the stop locations. Sadly, <a href="https://mapzen.com/blog/shutdown/">Mapzen is closing down</a> at the end of this month, so OSM extracts for metropolitan areas may become harder to come by; the extract I used is reproduced <a href="https://files.indicatrix.org/sb827/la.osm.pbf">here</a>. I used the build_valhalla.sh script in the linked GitHub repository to read the data and start Valhalla.</p><p>The information returned from Valhalla is a list of edges traversed by the route. These edges are derived from the OpenStreetMap network and roughly correspond to one block, the stretch of road between intersections. Valhalla edges are unidirectional, so one direction of a route does not share edges with the other direction (if it did, a route that passes northbound every 30 minutes, and southbound every 30 minutes, could qualify as a high quality corridor if the timing was right, a situation that is clearly incorrect).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xsASfUKHvxOAiWVXYGlnGw.png" /><figcaption>The Los Angeles Metro bus network, snapped to the street network. Data © LA Metro, OpenStreetMap contributors.</figcaption></figure><p>For each edge, I found the nearest stop on each pattern to the start of the edge. I collated the departures at those stops for all patterns traversing the edge, and estimated the headway. One factor that could be improved is that I did not interpolate the position of edges between stops; I considered that a bus traversed that edge at the moment it left the nearest stop, when in reality it could be some minutes later, depending on the distance to the stop. This is not an issue, however, since I am only estimating the headway, unless two routes that share a street segment do not have the same or similar previous stops. If buses leave from the same stop five minutes before they reach this edge, the headway between them will still be correct, even if the absolute times they traverse the edge are not. If the routes that traverse an edge don’t share the same or similar previous stops, however, this could introduce bias. This does occur in Los Angeles along Metro Rapid routes, which frequently overlap a Metro Local route, and should be corrected in future versions of the software.</p><p>I then computed all the inter-vehicle times on the segment between 7 and 9 AM, and between 4 and 6 PM. So if vehicles came at 7:05, 7:15, 7:26, etc., I would compute inter-vehicle times of 10 minutes, 11 minutes, etc. The bill states that vehicles must come no less than every 15 minutes, but in order to smooth over some of the variation introduced by not interpolating times, and to avoid disqualifying a segment if vehicles generally come at least every 15 minutes, but there is one 16 minute gap, I took the 85th percentile headway. I considered segments with an 85th percentile headway of less than 15 minutes as part of the high frequency network.</p><p>To determine the major transit stops where two high-frequency corridors cross, I intersected every edge that was part of the high frequency network with all of the others. If there was an intersection, and if the edges did not have any routes in common, I considered this a major transit stop. I required the edges to not have any routes in common to avoid declaring places where a route made a loop or intersected itself in the opposite direction to be major transit stops.</p><p>I created a 1/4 mile buffer around the high quality transit corridors. I then removed any disconnected portion with an area of less than π0.25² + 0.5 square miles, which happens to be the area of a 1/4 mile buffer around a straight line 1 mile long. These areas are likely to be data errors (for instance, a bus that goes around a block to serve a school and has both directions of the route traversing the same street in the same direction). I didn’t do this before buffering, for convenience. The edges were not linked into longer lines, but were rather a bunch of discrete line segments for each block. The buffers around them overlapped and merged into single shapes.</p><p>I then created half-mile buffers around the major transit stops (both the rail stops and the intersections of the high-quality transit corridors), and exported the output to GeoJSON for visualization. I made minor tweak in QGIS to correct a few obvious outlying data points.</p><p>In the future, I hope to improve the techniques used to calculate the headways on each segment, and expand the map to include more agencies across California. <a href="https://transitrichhousing.org/">Sasha Aicken has created a statewide map</a> using similar techniques.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a8e4eec37ed8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/identifying-high-quality-transit-corridors-a8e4eec37ed8">Identifying High Quality Transit Corridors</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Will Rezoning Around Transit Stops in California Undermine Support for Public Transit?]]></title>
            <link>https://medium.com/indicatrix/will-rezoning-around-transit-stops-in-california-undermine-support-for-public-transit-883e69aaab2a?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/883e69aaab2a</guid>
            <category><![CDATA[transportation]]></category>
            <category><![CDATA[california]]></category>
            <category><![CDATA[public-transit]]></category>
            <category><![CDATA[zoning]]></category>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Tue, 09 Jan 2018 00:14:11 GMT</pubDate>
            <atom:updated>2018-01-09T00:12:27.248Z</atom:updated>
            <content:encoded><![CDATA[<p>A proposal under consideration in the California State Senate would prevent cities from limiting residential density near public transit. SB 827, proposed by state Senator Scott Wiener, <a href="https://artplusmarketing.com/california-needs-a-housing-first-agenda-my-2018-housing-package-1b6fe95e41da">prevents cities from enforcing zoning ordinances with housing density maximums, sets minimum standards for height limits, and abolishes parking minimums, in the areas around “high connectivity” transit stations and along corridors with transit service more frequent than every 15 minutes</a>. This would allow the construction of much more sorely-needed housing in California. However, it could also increase opposition to future public transit projects.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Hg6q4huHnz25OKZOErFyMQ.jpeg" /><figcaption>Areas near transit in California could soon be open to denser housing development. Photo by <a href="https://www.flickr.com/photos/estebanandjulie/6977603164/in/photolist-bCA3jQ-cWsW25-cWsVQd-dP78EG-cWsUzh-cWsXkj-8LtbUg-cWsVBb-cWsWKd-cDdpyo-8Lt7s4-cWsXao-WGZWuG-8gySrA-WbhRHJ-ejy136-WbhZfs-8h2Z1t-Vxfp1z-Wz5RxR-UHKsZC-8uiXKc-WGZWwq-VuoLrU-WyWFx6-cWsWu5-cWsXUj-8gTQyZ-bRrKmc-WGZWxN-cWsUSb-cWsV2m-cWsVn1-dqE5Ru-Vxfp5T-cWsV9J-c7XZky-cWsUHL-9JymXn-cWsUeu-GZbdUh-dqE6eN-cWsWf1-9Rui53-cWsUrs-cWsWVq-c7XDg7-cvxUFu-bTUFiM-cw5FAq">Steve and Julie on Flickr</a>, CC BY 2.0.</figcaption></figure><p>I don’t yet have an opinion on the bill. I’m a dedicated <a href="http://yimbyaction.org/">YIMBY</a> (Yes In My Backyard, a movement promoting housing development in places where housing production does not keep up with demand). I am deeply troubled by rising housing costs throughout the nation, which are <a href="https://www.nytimes.com/2017/09/06/opinion/housing-regulations-us-economy.html?_r=0">largely due to restrictive zoning</a>. I strongly favor allowing construction of higher density housing, with less parking. From this standpoint, the bill is a great leap forward. It makes it much easier to build high density housing near transit. Increasing the supply of housing will help housing prices fall, and locating it near transit will allow residents to get around in an environmentally-friendly and cost-effective way.</p><p>However, I’m also a strong proponent of expanding public transit in our nation’s urban areas, and I’m concerned this bill will make that harder in California. Residents often protest vocally when higher density housing is proposed in their neighborhood. Parking is often a concern: residents are concerned that finding on-street parking in their neighborhoods will be more difficult with the influx of new residents. School crowding is another concern. Preserving neighborhood character (architectural or social) is another.</p><p>These are potent concerns, and they’re already used to block housing construction in California, <a href="https://www.nytimes.com/2017/12/01/business/economy/single-family-home.html">even when proposed construction is allowable under existing zoning</a>. SB 827 invites these same concerns to transit projects, such as new rail or bus rapid transit lines. Transit projects could be blocked over concerns that they would bring high-density housing to the neighborhoods they serve.</p><p>This is not necessarily a bad outcome, though. Transit thrives on density, so building more dense housing near it will make it more efficient and cost-effective. Maybe zoning for increased density <em>should </em>go hand in hand with transit construction. SB 827 would couple these two conversations, making any transit investment an investment in, and a conversation about, both transportation and land use. Less transit might get built, but the transit that does get built could be smarter and more effective.</p><p>In most places in the world, transit construction and land use change go hand in hand. In many place <a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1753302">increases in land value due to increased proximity to transit are “captured” and used to fund the development of public transport in the first place</a> (full disclosure: the first author of the linked paper, Deborah Salon, is my PhD adviser). For instance, in Hong Kong, newly-valuable land near new public transit lines is leased or sold, generating profits for the transit authority and funding infrastructure construction.</p><p>To some extent, public transport and increased housing density are already linked in California. The Regional Housing Needs Allocation (RHNA) process identifies the projected housing need in each metropolitan area in the state, and the Metropolitan Planning Organization (MPO) for that area, in coordination with the local jurisdictions, must then plan and zone to allow enough housing units to meet that need. Part of this planning is developing a Sustainable Communities Strategy, which involves projecting the greenhouse gas emissions that would likely result from transportation given the housing plan, and keeping them within certain limits. This effectively means planning for denser housing near transit, where residents will use more environmentally-friendly modes of transport. SB 827 goes further to permit additional housing production near transit.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*11IwaXRU0IqrFWdL-BwAgQ.png" /><figcaption>Areas in Los Angeles potentially affected by SB 827. Data © LA Metro, Metrolink, OpenStreetMap contributors.</figcaption></figure><p>SB 827 doesn’t just affect areas near subway stations. <a href="http://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=201720180SB827">The bill specifies</a> that its provisions apply to any location within 1/4 mile of a bus corridor with peak service at headways of 15 minutes or less, and 1/2 mile of a major transit stop (rail stops, and places where high frequency bus corridors intersect). The map above shows where these locations are in Los Angeles,¹ given current Los Angeles Metro and Metrolink service. The bill applies to the downtown area but also to many of the inner suburbs.</p><p>In Los Angeles, this is largely driven by a relatively high frequency bus system; much of the city is served at headways of 15 minutes or better, meeting the bill’s definition of high-quality transit and permitting denser development nearby. However, this also means that service planners might face opposition even to incremental improvements in bus service (e.g., increasing the frequency of a line from every 20 minutes to every 15).</p><p>SB 827 creates an interesting and important debate. It would permit the development of sorely needed housing in California. It would also create a public conversation about land use change when transit is constructed, potentially allowing transit to be more effective. But it would likely also cause opposition public transit projects. It’s not clear whether the benefits in housing construction would outweigh the added opposition to transit projects; that’s for the residents and the lawmakers of California to decide.</p><p><em>1: This map contains a number of assumptions, and certainly should not be used for planning. The biggest assumption has to do with the definition of “corridor”: I considered a street to be a high-frequency corridor if there was a bus every 15 minutes traversing it in the same direction, whether or not those buses were on the same route. For instance, consider two routes running every 20 minutes, but alternating to provide service every ten minutes. The shared segment would be considered a high-quality corridor. I intend to do a follow-up post in the coming weeks on the methodology used to create the map.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=883e69aaab2a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/will-rezoning-around-transit-stops-in-california-undermine-support-for-public-transit-883e69aaab2a">Will Rezoning Around Transit Stops in California Undermine Support for Public Transit?</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Pareto Efficiency and Urban Planning]]></title>
            <link>https://medium.com/indicatrix/pareto-efficiency-and-urban-planning-5cfef767ab6c?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/5cfef767ab6c</guid>
            <category><![CDATA[urban-planning]]></category>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Mon, 26 Dec 2016 00:25:19 GMT</pubDate>
            <atom:updated>2016-12-26T00:25:56.557Z</atom:updated>
            <content:encoded><![CDATA[<p>Urban planning is the process by which society decides how our cities and regions will develop in the future. As anyone who’s been involved in a planning process can tell you, it is a very difficult field. I argue that the reason for this is because there are no easy wins in planning; we cannot improve outcomes in one dimension without making them worse in another.</p><p>The theoretical underpinning of what I’m describing is known as <a href="https://en.wikipedia.org/wiki/Pareto_efficiency">Pareto efficiency</a>. The gist is that you have multiple measures which you would like to minimize or maximize. A scenario is Pareto-efficient when you cannot improve any measure of interest without making another measure worse.</p><p>Consider a situation where you are interested in both minimizing greenhouse gas emissions and maximizing the speed with which people can move. One solution is to have no motorized transport at all, which clearly minimizes greenhouse gas emissions, although at a cost to the speed with which people can move. One could also imagine a world in which everyone gets around by private helicopter. This clearly allows for extremely high speed movement, but falls short on the goal of minimizing greenhouse gas emissions. In the first case, one cannot increase speed of movement without making greenhouse gas emissions worse, while in the second case, one cannot decrease greenhouse gas emissions without making movement slower. Both scenarios are Pareto-efficient. Of course, in between these extremes are many compromise solutions, for example pedestrian cities with public transit, or high-speed trains between cities and districts within a city. These compromises are also Pareto-efficient: we can’t make the trains faster without increasing energy use and thus greenhouse gas emissions, and we can’t reduce greenhouse gas emissions without running fewer trains, causing the speed of movement to decrease. These optimal solutions form a tradeoff curve, showing for a given level of greenhouse gas emissions what speed is possible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/752/1*eSwUoHOVuE7uZSPoWqi4rA.png" /><figcaption><em>Different possible scenarios for transport systems. Walking emits very little greenhouse gas but is slow; helicopters emit large quantities of greenhouse gases but are very fast. Numerous compromise solutions exist, for example walkable cities with rapid transit. Other compromises exist and they lie along the curve, which represents the tradeoff between speed and greenhouse gas emissions. All scenarios on this curve are Pareto-efficient. There also exist scenarios that are not Pareto-efficient, for example everyone riding tractors to get around the city.</em></figcaption></figure><p>Of course, one can come up with solutions that are not Pareto-efficient. One can imagine a scenario in which everyone has a farm tractor that they use to get around. Farm tractors burn large amounts of fossil fuels, and they are slow. It’s possible to both increase speed and reduce greenhouse gas emissions (for example, by replacing the tractors with small cars, or rapid transit). Thus, the tractor solution is not Pareto-efficient for the measures outlined above.</p><p>Suppose, though, that in addition to caring about greenhouse gas emissions and speed, we also cared about being able to cross large tracts of unpaved land. This is a third dimension to the graph shown above, and now the tractor solution is one of the Pareto-efficient solutions: it’s better at crossing unpaved land than rapid transit, it’s faster than walking, and it still has lower greenhouse gas emissions than helicopters.</p><p>This is the crux of why planning is hard. We care about many things, and we have to make hard choices to trade those things off. Planning isn’t an optimization problem that we can solve; there are many mathematically optimal (or Pareto-efficient) solutions given the measures we care about. The examples so far have been contrived, but they don’t have to be: planning is a messy, multidimensional problem. Consider the goals of a typical regional transportation plan: decrease congestion, increase economic opportunity, decrease greenhouse gas emissions, accommodate growth, &amp;c. Add to this all of the dimensions that may not be stated goals of the planning process but that some stakeholder (be it a business, a member of the public, &amp;c.) cares about, for instance not losing their view of the water, or not having a hard time finding parking in front of their house, and you have a highly multidimensional problem. I would posit that the urban system we have is already Pareto-efficient when you consider all of the various dimensions that someone cares about; any change to improve one dimension will harm another. Thus, the planning process is about making a tradeoff between these dimensions.</p><p>Not recognizing this is what led us to the urban system we have today, at least in the US, with core cities decimated by freeways and exurbs sprawling for literally hundreds of kilometers outside the city limits. Rather than recognizing the highly multidimensional (and, in some cases, orthogonal) goals of planning when we built the national urban system, we deferred to the engineers (along with the oil and car companies) and built cities around the primary goal of increasing potential automobile throughput (along with demonstrating America’s economic prowess to the world). The idea that we can optimize our way to a single planning solution is a dangerous fallacy, because all it means is that the process by which the decision about what dimensions are important is hidden inside a black box of some sort.</p><p>I’ll end with a note on evidence-based planning. More and more planning processes are evidence driven, and more and more measures are being created, for outcomes as diverse as emissions and income inequality. Paradoxically, creating more measures will make planning harder, not easier, as more tradeoffs between various values must be made. This isn’t a bad thing; making decisions about which measures (and the outcomes they imply) are most valued is exactly the point of planning. This is the reason planning is not a hard science, but instead a social science or even a humanity; many of the thinkers who have most informed planning policy today come not from the sciences but from philosophy, design, or art — everyone from <a href="http://www.politics-prose.com/book/9780679741954">Jane</a> <a href="http://fortune.com/2011/09/18/downtown-is-for-people-fortune-classic-1958/">Jacobs</a> and <a href="http://www.politics-prose.com/book/9780313227479">Lewis Mumford</a> to <a href="http://classics.mit.edu/Plato/republic.html">Plato</a>. Evidence-based planning is great, as long as it spurs conversation about what measures are important and the tradeoffs between them. But don’t expect it to make our lives easier; indeed, quite the opposite.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5cfef767ab6c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/pareto-efficiency-and-urban-planning-5cfef767ab6c">Pareto Efficiency and Urban Planning</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Millenials and Communal Living]]></title>
            <link>https://medium.com/indicatrix/millenials-and-communal-living-dd7425ba89ea?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/dd7425ba89ea</guid>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Sat, 12 Sep 2015 00:00:00 GMT</pubDate>
            <atom:updated>2016-12-25T23:18:40.378Z</atom:updated>
            <content:encoded><![CDATA[<p>A <a href="http://www.citylab.com/housing/2015/08/why-millennials-arent-forming-new-households/401783/">Citylab article from a few weeks back</a> observes that the rate of household<br>formation among millenials remains low. A large proportion of millenials are still living with their parents, certainly a sign of economic hardship. However, the article also notes that it is “alarming” that when millenials eventually do leave their parents’ households, they often do not form new households but instead live with other adults. The article views this as a sign of economic malaise, which indeed it is, but fails to note that there are concomitant social changes that drive millenials to live together, independent of economic concerns.</p><p>The article, and likely society at large, assumes that the reasons millenials are living together is that they cannot afford to live alone, that young adults would not choose to live with unrelated adults unless forced to to make ends meet. To some extent, this is true: rents are high and many millenials struggle to afford them, and live with others for this reason. However, there is a significant minority who could afford to live by themselves, but choose not to. The below plot shows the distribution of personal (not household) income for millenials (age 18–34) in the District of Columbia (one of the more expensive places to live in the country) who live in households comprised of at least three 18–34 year olds, and no others. The data is derived from the <a href="https://www.census.gov/programs-surveys/acs/data/pums.html">2009–2013 American Community Survey Public Use Microdata Sample</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kZ54ZBZcWKM9ZvIR.png" /><figcaption>Histogram of annual personal income of millenials who live with others</figcaption></figure><p>Median monthly housing costs per housing unit in the District of Columbia were $1,454 in 2013 (<a href="http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_13_5YR_B25105&amp;prodType=table">source</a>). Going by the common rule of thumb that housing costs should be less than 30% of income, one finds that to afford the median rent in DC one must make $58,000 per year, which is the location of the black line on the above chart. 21% (+/- 4%, 95% confidence interval using the replicate weights method to calculate standard errors) of 18–34 year olds currently living with other 18–34 year olds in the District make more than that, and could potentially afford to live alone. 21% isn’t huge number (put another way, 79% of 18–34 year olds who live together in DC couldn’t afford to live alone), but it is a significant minority.</p><p>I see several obvious explanations. One is simply that millenials don’t like to live alone. This can hardly be called out as a sign of economic malaise; rather it is a sign of social change. Millenials find social benefits in communal living, from forming lasting friendships to just having someone to hang out with in the evenings. Additionally, millenials who could afford to live alone but are choosing to live together are not just letting the savings evaporate into thin air; they are either spending them on other goods and services (a boon for the local economy) or saving them (a boon for themselves later in life).</p><p>Another possible cause has to do with where people are choosing to live. More and more, millenials are choosing to live in central areas of cities, where rents are higher and space is at a premium. Many of the people who could afford their own place somewhere in DC might not be able to afford their own place in the neighborhood in which they would like to live. Again, this is not necessarily a sign of economic malaise. There is not enough housing in the core for everyone who would like to live there to live alone, and this is reflected in housing prices. Many millenials are making a perfectly rational economic choice, choosing the benefits of living in their desired neighborhood over the sacrifice of not having their own place. The District of Columbia is fairly compact, but there are<br>certainly lower-cost areas that may not be seen as desirable by millenials.</p><p>Finally, there may be a supply-side problem. The Census numbers are based on what people who currently live in the District pay. If you’re trying to establish a new household, you are renting or buying not from the pool of all properties, but from the pool of properties currently on the market. It’s entirely possible that median cost of properties on the market is higher than the median cost of properties that are currently occupied; people may be less likely to leave more reasonably-priced properties. Zillow, for example, <a href="http://www.zillow.com/washington-dc/home-values/">finds that the median price of District of Columbia properties</a> listed for rent there is $2,300 per month. (This discrepancy could also be due to selectivity bias, i.e. more expensive than average properties being listed on Zillow).</p><p>Economic hardship certainly is a huge factor in millenials choosing to live together, and certainly in choosing to continue living at home. The majority of millenials who live with other adults in DC could not afford the median housing cost in the District. However, affordability is not the only factor; there is also a significant minority of millenials who choose to live with a group of other adults, even though they could afford to live by themselves. We should be careful to recognize the effects of social change as well as economic hardship on household formation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dd7425ba89ea" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/millenials-and-communal-living-dd7425ba89ea">Millenials and Communal Living</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Game Theory and Snow Shoveling]]></title>
            <link>https://medium.com/indicatrix/game-theory-and-snow-shoveling-b1d67fcff05b?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/b1d67fcff05b</guid>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Thu, 26 Feb 2015 00:00:00 GMT</pubDate>
            <atom:updated>2016-12-25T23:18:59.500Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="Snow in Washington, DC" src="https://cdn-images-1.medium.com/max/1024/0*y11u4XZB-rla1IEW.jpg" /></figure><p>My adopted city of Washington, DC, is shoveling out from under some of the last snow of the season. At the same time, we’ve just <a href="http://greatergreaterwashington.org/post/24900">enacted a law</a> that levies fines on residents who don’t clear their sidewalks, after <a href="http://greatergreaterwashington.org/post/13630">years of debate</a>.</p><p>There’s a very good reason to fine property owners for not shoveling their sidewalks. Snow shoveling is a what’s known as a <a href="https://en.wikipedia.org/wiki/Prisoner%27s_dilemma#Multiplayer_dilemmas">multi-player prisoner’s dilemma</a>. If everyone shovels, everyone is better off (because they can get around more easily, don’t face the danger of falls, don’t have to pay extra cleaning bills due to muddy/salty pant cuffs and skirt hems, &amp;c.). However, any one person is better off if they don’t shovel, because they’ve saved time (I actually like shoveling my sidewalk, but let’s assume that the utility for shoveling is uniformly negative). Each individual isn’t affected much by the state of their own sidewalk, but rather by the state of everyone else’s.</p><p>We can represent this issue graphically, as Thomas Schelling does in chapter 7 of <a href="http://www.politics-prose.com/book/9780393329469"><em>Micromotives and Macrobehavior</em></a>. For simplicity’s sake, let’s assume that the only benefit to shoveled sidewalks is a quicker walk to wherever you need to go. Let’s also assume that it takes five minutes to shovel your sidewalk, and that you can save twenty minutes in transportation time if every sidewalk is shoveled, as opposed to none. Finally, let’s assume that if half of the sidewalks are shoveled, you save half as much time: ten minutes.</p><p>The graph below shows how much time is saved by people who do not shovel (red line), and those who do shovel, as a function of how many people choose to shovel. The time is the amount of time saved commuting, minus the time spent shoveling. As you can see, at every point there is an incentive not to shovel, but take advantage of your neighbors’ shoveled walks. However, if everyone does that, you reach a situation where no one’s walks are shoveled, and no one is saving any time. If everyone’s walks are shoveled, everyone saves time despite having to spend time shoveling (the right end of the blue line). Since the second situation is clearly the preferable one, we need to have some system to encourage people to shovel. This could take many forms: a general sense of community standards, <a href="http://greatergreaterwashington.org/post/21478">public shaming of those who do not shovel</a>, or a regulation with fines attached.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*zgcfhzORiUTl8uRL.png" /><figcaption>Plot of payoffs for shoveling your walk</figcaption></figure><p>We can carry the economic theory further, and say that the amount of the fine has to be greater than the distance between the two lines, but that’s not really a meaningful thing to do. For one thing, the distance between the lines is in units of time, while the fine is in units of money, two things that convert in wildly different ways for different people. Also, we must remember that the numbers we used were completely arbitrary and furthermore not uniform across the population.</p><p>Schelling also points out that there is a minimum viable coalition size, a minimum number of people who have to cooperate to shovel their sidewalks in order to realize a net gain. This is the point where the payoff curve for shoveling reaches the payoff that everyone would get if no one shoveled. In the figure above, at about 0.2, those who choose to shovel are better off than they would have been had no one chosen to shovel. This isn’t really a meaningful number here for several reasons. As mentioned before, the assumption of uniform arbitrary payoffs is invalid. Additionally, you are not equally affected by every shoveled sidewalk. A geographically dispersed coalition may be no better than no coalition, if no member of the coalition walks on the shoveled sidewalks of any other member. A geographically concentrated coalition may be viable even at small numbers, because the sidewalks being shoveled are the sidewalks most relevant to the members. (Consider a commercial strip where the business cooperate to shovel their stretch of road. Even though the coalition is small, its members benefit because customers can walk up and down their commercial strip).</p><p>Thus, fines for not shoveling are a way of encouraging a socially desirable outcome, when individual motivations would result in a socially undesirable outcome.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b1d67fcff05b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/game-theory-and-snow-shoveling-b1d67fcff05b">Game Theory and Snow Shoveling</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Dijkstra’s Algorithm Visualization in the Browser]]></title>
            <link>https://medium.com/indicatrix/dijkstras-algorithm-visualization-in-the-browser-c72ad0722589?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/c72ad0722589</guid>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Fri, 26 Dec 2014 00:00:00 GMT</pubDate>
            <atom:updated>2016-12-25T23:19:07.539Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="visualization of Dijkstra&#39;s algorithm" src="https://cdn-images-1.medium.com/max/507/0*9_7yzn-dqccPdqXs.png" /></figure><p><a href="http://en.wikipedia.org/wiki/Dijkstra%27s_algorithm">Dijkstra’s algorithm</a> is a method for finding the shortest path through a <a href="http://en.wikipedia.org/wiki/Graph_%28mathematics%29">graph</a> with non-negative edge weights. It is a relatively efficient algorithm, and is guaranteed to find the shortest path (unlike some heuristic algorithms). It’s a fairly simple algorithm, but also one that can be difficult to wrap one’s head around. Descriptions abound on the Internet, and there are a number of videos, and even a few interactive demonstrations, but I thought there needed to be a demonstration that was interactive, worked in modern browsers without a plugin, used a non-trivial graph, and was open source. So I <a href="http://projects.indicatrix.org/dijkstra.js">wrote one</a>. A prose description of the algorithm is there; I hope it’s easier to understand with the interactive component. Visualizing algorithms tends to make them easier to understand, as observed by <a href="http://bost.ocks.org/mike/algorithms/">Mike Bostock</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c72ad0722589" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/dijkstras-algorithm-visualization-in-the-browser-c72ad0722589">Dijkstra’s Algorithm Visualization in the Browser</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Accessibility Analysis with Python and OpenTripPlanner]]></title>
            <link>https://medium.com/indicatrix/accessibility-analysis-with-python-and-opentripplanner-24aad2a52f4a?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/24aad2a52f4a</guid>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Sat, 01 Nov 2014 00:00:00 GMT</pubDate>
            <atom:updated>2016-12-25T23:19:02.592Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/350/0*psFPPTAr8Av3ejU2.png" /><figcaption>Walking time to the nearest grocery store for all Census blocks in Chicago.</figcaption></figure><p><a href="http://www.opentripplanner.org/">OpenTripPlanner</a> is a great bit of software for both <a href="http://ride.trimet.org/">customer-facing tools</a> and<br><a href="https://indicatrix.org/2013/06/20/microaccessibility-with-opentripplanner/">analysis</a>. Until recently, it had the capability to perform batch queries, calculating an origin-destination matrix or an aggregate measure of<br><a href="http://www.vtpi.org/tdm/tdm84.htm">accessibility</a>. Configuring this functionality, however, was somewhat<br>awkward, as it used a verbose XML format that was more suited to allowing developers to configure application components than as a user-facing interface (and I say that having been one of the last defenders of this approach on the OTP mailing list).</p><p>This batch analysis tool was removed as a side effect of a restructuring and simplification of the OpenTripPlanner codebase that has been ongoing for several months. Its absence sparked <a href="https://groups.google.com/forum/#!topic/opentripplanner-dev/tk427ITino0">a debate on the opentripplanner-dev mailing list</a>, which broke down roughly into two camps: one camp arguing for something that is purely a configuration file, with another camp arguing for “configuration files” that are simply scripts of some sort (I argued for both camps at one point or another). Where that conversation lies now, to make a long story short, is that there are tentative plans to rebuild Batch Analyst using Java Preferences as a configuration file format.</p><p>In parallel with this development, development has been ongoing on a<br><a href="https://github.com/conveyal/analyst-server">web-based analytics framework</a>. This is a very useful (and just plain neat) tool for accessibility analysis in a graphical user interface driven package. This is exactly what is needed for probably the majority of those doing accessibility analysis. However, coming from a research background<br>(quantitative/computational geography), I often want tools that I can script, commit my methods to a git repo, and integrate with other tools. That said, work on this graphical interface to Analyst has driven a rethinking of how analysis is done in OTP and the creation of many useful components.</p><p>In some personal projects, I needed to be able to run batch jobs again, and I decided to try to build a quick and dirty Python library to call the OTP analysis functions. (To be fair, integrating OTP and Python was originally proposed by Tuukka Hastrup in the aforementioned thread). The result is<br><a href="https://www.github.com/mattwigway/opentripplanner-jython">here</a>. It’s a <a href="http://www.jython.org/">Jython</a> library that wraps up the functionality of OTP’s analysis functions in a hacker-friendly library. I decided to take a simple approach and build a library that does one thing and one thing well: creates origin-destination matrices. What you build around that is up to you. If you want a simple cumulative accessibility measure, you can sum the number of links<br>that are below a threshold. If you want to use a more complicated accessibility measure, with distance decays and such, you can just implement some Python code to do that.</p><p>The map above is the result of a demonstration of this project. It shows the walking time to the nearest grocery store from every Census block in Chicago. Here’s how I made it. First, I downloaded the binary distribution of OTP’s master (development) version from <a href="http://dev.opentripplanner.org/jars/">here</a>. I grabbed OpenStreetMap data for Chicago from <a href="https://mapzen.com/metro-extracts">mapzen’s metro extracts</a> site, and Census blocks<br>and grocery store locations from the <a href="https://data.cityofchicago.org/">City of Chicago Data Portal</a>. I built an OTP graph using the standard methods. I then edited the grocery stores file to have only latitude and longitude columns (because, internally, OTP seems to try to convert the other columns to integers for use as inputs to aggregators).</p><p>I then ran <a href="https://gist.github.com/mattwigway/15c2d7dc64901f0f3cc5">this code</a> to perform the analysis. It must be run in Jython as opposed to standard Python, the OTP jar must be on the Jython classpath, and the opentripplanner-jython module must be in Jython’s Python search path somewhere. I ran it like so:</p><pre>CLASSPATH=~/opentripplanner/otp-latest-master.jar jython -J-Xmx8192m accessibility.py</pre><p>The -J-Xmx8192m tells the Java Virtual Machine to use 8GB of RAM. If you don’t have that much, you can experiment with smaller numbers.</p><p>I’ll walk you through what the code does. It loads the graph which was previously built (which it expects to find in the graph subdirectory of the working directory), loads the destinations, links them to the graph, creates a batch processor with the origins, and then evaluates that batch processor on the destinations. The result of the call to BatchProcessor.eval() is an origin-destination matrix, with origins on the rows and destinations on the columns. Unfortunately, numpy is not available in Jython, so data is returned using the opentripplanner.batch.Matrix class.</p><p>This tool helps eliminate a lot of the repeated computation in classic batch analyst runs. You load the graph only once, for example, and you could link the destinations only once if you were running the batch processor multiple times, say with different mode sets. You could calculate travel times to multiple destination sets without re-running the batch processor, but by simply calling eval() more than once. Remember that adding additional destinations, or calculating accessibility for additional sets of destinations, is cheap; you’re just sampling points in the graph. Adding additional origins is expensive: for each origin, OTP builds a shortest path tree.</p><p>Under the hood, it uses the new Analyst framework, which calculates the travel time from each origin to every vertex in the graph and stores it in a time surface, which we can then sample inexpensively.</p><p>One caveat is that this library doesn’t yet support <a href="https://github.com/opentripplanner/OpenTripPlanner/issues/1328">profile routing</a>, although OTP does. Profile routing is a much better way of doing general accessibility analysis for general queries for public transportation (e.g. how long does it take to get to work) versus extremely specific queries (if I leave <em>right now</em>, how long <em>exactly</em> will it take me to get to work <em>today, right now</em>).</p><p><em>Update 2014–12–31:</em> I added notes about memory consumption.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=24aad2a52f4a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/accessibility-analysis-with-python-and-opentripplanner-24aad2a52f4a">Accessibility Analysis with Python and OpenTripPlanner</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using GeoTools with Multiple User Accounts]]></title>
            <link>https://medium.com/indicatrix/using-geotools-with-multiple-user-accounts-c84e563b24e3?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/c84e563b24e3</guid>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Mon, 20 Oct 2014 00:00:00 GMT</pubDate>
            <atom:updated>2016-12-25T23:19:06.557Z</atom:updated>
            <content:encoded><![CDATA[<p>I have a situation where I have multiple GeoTools applications being run on a server by different users. I was having a lot of issues with exceptions about not being able to decode CRS codes, even though I was sure I had the <a href="http://docs.geotools.org/latest/userguide/library/referencing/hsql.html">gt-epsg-hsql</a> file included in my JAR, and had <a href="http://docs.geotools.org/stable/userguide/build/faq.html#how-do-i-create-an-executable-jar-for-my-geotools-app">set up Maven correctly</a> to include the dependency.</p><p>It turns out the issue was that the gt-epsg-hsql extracts its hsql database of projections to Geotools in the system temporary directory, and if there are multiple geotools apps running as different users, the first one to start gets the directory, and the remaining ones crash because they don’t have permissions to access it.</p><p>The workaround is to use separate temporary directories for each user. The easy way to do this is TMPDIR=`mktemp -d` application, which creates a new unique temporary directory each time an application is started.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c84e563b24e3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/using-geotools-with-multiple-user-accounts-c84e563b24e3">Using GeoTools with Multiple User Accounts</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Running RStudio Server on Amazon EC2]]></title>
            <link>https://medium.com/indicatrix/running-rstudio-server-on-amazon-ec2-da812413c48b?source=rss----4045e448107f---4</link>
            <guid isPermaLink="false">https://medium.com/p/da812413c48b</guid>
            <dc:creator><![CDATA[Matthew Wigginton Conway]]></dc:creator>
            <pubDate>Sun, 10 Aug 2014 00:00:00 GMT</pubDate>
            <atom:updated>2016-12-25T23:18:41.518Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="RStudio Server running on Amazon AWS, accessed via SSH tunneling" src="https://cdn-images-1.medium.com/max/1024/0*rjP-TBXNbjtGH4ge.png" /></figure><p><a href="http://www.r-project.org/">R</a> is a great environment for statistical computing; I’ve used it in<br>a <a href="https://indicatrix.org/2014/05/01/predicting-the-popularity-of-bicycle-sharing-stations/">number</a> <a href="https://indicatrix.org/2014/04/24/bikeshare-clustering/">of</a> <a href="https://indicatrix.org/2013/12/14/effects-of-space-and-time-on-bikeshare-use/">projects</a>. <a href="http://rstudio.com/">RStudio</a> is hands-down the best IDE for R (there is even less debate here than there is about emacs being the best editor). Sometimes, though, I find that I need to run analyses that require more computing power than my personal computer can provide (especially since my desktop is currently in storage in California and I’m in Illinois with a circa-2007 netbook with 2GB of RAM).</p><p><a href="http://aws.amazon.com/ec2/">Amazon EC2</a> is a great solution for this type of issue; it allows you to spin up powerful computers on-demand and access them over ssh. You don’t get a graphical interface, though, which precludes running RStudio Desktop.<br>However, RStudio provides a server tool that allows you to run R on a server and access it through your browser. Configuring it on EC2, however, is a wee bit tricky because most people use public key authentication to access their instances (which is good for security), while RStudio assumes that you can log in with a password. One solution is to switch to password authentication for your entire instance, but it’s possible (and more secure) to keep the public key authentication.<br>Here’s how to do it.</p><ol><li>Spin up an EC2 instance. I like to use the ones from <a href="http://cloud-images.ubuntu.com/locator/ec2/">Ubuntu Cloud</a>.</li><li>You can now log in using your key pair like so: ssh -i ~/.ssh/key.pem ubuntu@&lt;aws-ip&gt;</li><li>Install a recent version of R from <a href="http://cran.r-project.org/">CRAN</a> by following the directions at <a href="http://cran.r-project.org/bin/linux/ubuntu/">http://cran.r-project.org/bin/linux/ubuntu/</a> (assuming you’re on Ubuntu).</li><li><a href="http://www.rstudio.com/products/rstudio/download-server/">Download RStudio Server</a> and install it. You may need to replace gdebi with dpkg -i because X11 is not available.</li><li>Add the following line to /etc/rstudio/rserver.conf. This forces RStudio to listen only for connections on localhost, so that a public key is still needed to access it.<br> www-address=127.0.0.1</li><li>Restart RStudio Server: sudo restart rstudio-server</li><li>Create a password for your login account: sudo passwd your-user-name. You won’t be able to SSH in with this (assuming that you only allow public key auth), but you’ll use it to log into RStudio.</li><li>On your local machine, run ssh -N -L localhost:8787:localhost:8787 -i ~/.ssh/aws2.pem ubuntu@&lt;aws-ip&gt;. This forwards the RStudio Server session securely to your computer using SSH tunneling. Note that any user of your local machine can now access RStudio Server, although they’ll need the password you created in step 7.</li><li>Go to <a href="http://localhost:8787/">http://localhost:8787/</a> and log in with the password you just created.</li><li>Analyze away.</li></ol><p>Note that you are just accessing RStudio on AWS, so you’ll need to have all of your data and R scripts on the server.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=da812413c48b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/indicatrix/running-rstudio-server-on-amazon-ec2-da812413c48b">Running RStudio Server on Amazon EC2</a> was originally published in <a href="https://medium.com/indicatrix">Indicatrix</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>