Weekly check in 2013.04.25

• 13:32 <demory> hello everyone
• 13:33 <FrankP> Hey demory ... if you didn't see, a question you might want to answer in the user group today: https://groups.google.com/forum/?fromgroups=#!msg/opentripplanner-users/37cMdiE-dZw/sFWMPOy-w10J
• 13:34 <demory> ah, thanks FrankP
• 13:34 <demory> I know Stephane, we've worked together on some OTP stuff before
• 13:35 <demory> i'll get in touch w/ him and bring him up to speed on the new company, SFC, etc.
• 13:35 <demory> I wonder if the strange routing issues are related to what you've been seeing
• 13:36 <demory> abyrd is working on that this week, btw
• 13:36 <mattwigway> It might also just be unlinked stops (which is a common problem, I think... something I had when first starting with OTP)
• 13:36 <FrankP> not sure ... sounds more like lack of paths in OSM (but it could be the stuff I've seen)
• 13:36 <demory> (he is unable to join today -- in Delft for a code sprint on the OTP NL project)
• 13:37 <FrankP> cool
• 13:39 <demory> my update -- just checked in a series of new UI enhancements, and just updated the demo at http://trimet-ctm.deployer.opentripplanner.org/ -- still need to make sure everything's working there
• 13:40 <mattwigway> I'm still working on accessibility analysis with batch analyst, although this week I've been sidetracked on OSM fixes.
• 13:40 <FrankP> On the bug front, Andrew asked about 1 of my bugs ... he couldn't reproduce, but I'm still seeing it happen. There are a few new bugs added on Monday. OTP seems not to be using (or not weighting highly enough) preferred transfers.
• 13:41 <demory> yeah, i know he saw the ones you posted Monday; was going to work on that this week
• 13:42 <demory> i'll be working next on the Field Trip module -- need to do some debugging there
• 13:44 <demory> on another topic -- is anyone here going to State of the Map US in SF in June?
• 13:44 <mattwigway> When is it in June?
• 13:44 <demory> I am planning being there -- am out in CA then anyway
• 13:44 <demory> June 8-9
• 13:44 <mattwigway> Ah, then no :(
• 13:44 <demory> w/ optional days on the 7th and 10th
• 13:45 <FrankP> nope
• 13:45 <demory> ok. well I'll be there and am planning on proposing an OTP-related session
• 13:45 <FrankP> BTW David...new UI is looking really nice
• 13:45 <demory> tomorrow is the deadline for proposals
• 13:46 <demory> thanks Frank!
• 13:46 <demory> looks like the new integrated Analyst module isn't working online -- that's something I'd like to have for the meetings in two weeks
• 13:47 <demory> I know there is interest in Analyst from ODOT and the TriMet planning folks
• 13:47 <mattwigway> What exactly does the integrated analyst module do? Is it stock OTP analyst or batch analyst?
• 13:47 <mattwigway> I'm finding that batch analyst is much more powerful.
• 13:48 <demory> it just uses the stock analyst now
• 13:48 <demory> though it isn't doing anything on that demo -- I'll look into that
• 13:48 <mattwigway> I think batch analyst is something that would be more useful were it to have a GIS plugin front end.
• 13:49 <demory> but it's supposed to be drawing the isochrone rasters as the default behavior
• 13:49 <demory> so, mattwigway, before I go to Portland (the week of May 6th) we should probably talk more about the status of your work on batch analyst
• 13:50 <mattwigway> sure, that'd be great
• 13:50 <demory> i agree it is more powerful, and probably of greater interest to the planning folks there
• 13:50 <mattwigway> I haven't written much code (I came up with one new accessibility measure)
• 13:50 <mattwigway> but yes for planning it can do everything stock analyst can do and a lot more
• 13:51 <mattwigway> well, I need to write one more class to make it generate isochrones, but that's definitely possible
• 13:51 <mattwigway> I'm using it for accessibility analysis, using a gravity model right now, but it has the potential for much more powerful models of accessibility
• 13:52 <mattwigway> probably the most applicable paper from the transit side is Busby 2004
• 13:52 <mattwigway> And Geurs et al from the math side
• 13:53 <mattwigway> That's Busby, Jeffrey R. 2004. �Accessibility-Based Transit Planning�. Massachusetts Institute of Technology. http://hdl.handle.net/1721.1/32414.
• 13:53 <mattwigway> and Geurs, Karst T, and Bert van Wee. 2004. �Accessibility Evaluation of Land-use and Transport Strategies: Review and Research Directions.� Journal of Transport Geography 12 (2): 127�140. doi:10.1016/j.jtrangeo.2003.10.005. http://www.sciencedirect.com/science/article/pii/S0966692303000607.
• 13:53 <demory> great -- maybe we can connect by phone sometime? I'd like to talk this through in more detail before my trip
• 13:53 <mattwigway> yeah that would be great; this week[end] is pretty busy for me but maybe next week?
• 13:54 <demory> yeah, next week is pretty open for me
• 13:54 <mattwigway> I'll have to look at my schedule but I can email you.
• 13:55 <mattwigway> abyrd probably know more about it than I do at this point though.
• 13:55 <mattwigway> I think batch analyst is potentially material for a webinar though.
• 13:55 <demory> great. I may loop in abyrd too; he may want to sit in on this as well
• 13:55 <mattwigway> It can definitely do things that we don't see too much outside of academia.
• 13:56 <demory> yeah, sounds like it
• 13:56 <demory> ok, we'll follow up by email there
• 13:56 <mattwigway> The Busby thesis has some great case studies of accessibility-based transit planning in Chicago, and is the only example I know of of before-and-after accessibility analysis.
• 13:57 <mattwigway> in a real-world transit setting.
• 13:57 <demory> Ok, I'll take a look at those before we chat. Thanks for the links
• 13:57 <mattwigway> So I'd recommend a glance at that, esp. pages 20-25 and around 91.
• 13:57 <mattwigway> It's also free because it's an MIT thesis :)
• 13:58 <mattwigway> Geurs et al. is probably more applicable for developers than for users.
• 14:00 <demory> sure, i'll take a look at both. thanks!
• 14:00 <mattwigway> you're welcome!
• 14:00 <demory> so, anything else for the check-in today?
• 14:03 <mattwigway> nothing here.