Weekly check in 2012.02.23 - GeoSmartCity-CIP/gsc-opentripplanner GitHub Wiki
13:31 <demory> Hello everyone
13:31 -!- mattwigway [~[email protected]] has joined #opentripplanner
13:32 -!- grant_h [d819d2df@gateway/web/freenode/ip.216.25.210.223] has joined #opentripplanner
13:32 <novalis_> hi demory
13:33 <demory> well, why don't we get started
13:34 <demory> My update: OTPSetup (which we're branding as OTP-Deployer) is live! http://deployer.opentripplanner.org. I'm still doing some fine-tuning but you are welcome to give it a try. We'll probably do some sort of formal announcement in the next week or so.
13:34 <novalis_> I finished up the segment matching code, and have started working on random bugs.
13:35 <demory> novalis_ I'm just starting to look at the segment matching code, i
13:35 <novalis_> demory, I hope it makes some sense.
13:36 <mele> what is it going to be used for?
13:36 <demory> thanks, i'll let you know if I have questions
13:36 <novalis_> mele, actually, a few things
13:36 <novalis_> What it does is matches arbitrary shapes (as in GTFS shapes.txt) to a street network
13:37 <novalis_> So, could use it to clean up your shapes.txt file
13:37 <PJ_____> oo, does that mean we can fix the horrible gtfs on interstate?
13:37 <mele> haha no that's up to hastus starting to use OSM
13:37 <mele> but i'm thinking it might be useful for us to track changes
13:37 <novalis_> We also hope to use it for some system mapping
13:37 <mattwigway> novalis, around here we don't even have shapes.txt. Could it be used to generate a baseline shapes.txt?
13:37 <novalis_> mattwigway, actually, probably
13:38 <novalis_> We would like to integrate it into Kevin's GTFS editing tool
13:38 <mattwigway> That would be cool - our area is kind of U shaped, so the express buses that go through the bay from E San José to Palo Alto look strange.
13:38 <demory> mattwigway, kpw has been working on a GTFS editor tool that would be well suited to that, tho novalis_' work could help with it too
13:38 <novalis_> But that will require a bunch of work, since it presently operates on OTP internal edges rather than on generic lines
13:39 <novalis_> But that would be easy enough to fix.
13:39 <mattwigway> Cool, I haven't seen kpw's tool
13:39 <novalis_> Maybe we'll finally find a use for JTS planargraph
13:39 <demory> its not reallly public yet, but there are plans to open source it in the future
13:39 <novalis_> I thought it was in public SVN somewhere?
13:40 <demory> hmm it may be, we'd have to check w/ kpw
13:40 <demory> i've been kind of walled off from that work since the client is my old employer :)
13:41 -!- andrewbyrd [[email protected]] has joined #opentripplanner
13:41 <mattwigway> Thanks!
13:42 <FrankP_> UPDATE -- I'm getting ready to release an RTP update (uses SOLR as the geocoder/search). Waiting on a fix to #211 (bug Mele found yesterday). Stagging: http://maps5.trimet.org/otp.html
13:43 <novalis_> Looks like it was last touched in December
13:44 -!- andrewbyrd [[email protected]] has joined #opentripplanner
13:45 <novalis_> FrankP, actually, is it possible for me to use TriMet's geocoder locally on my version of OTP?
13:45 <novalis_> I just want to try to fix #557.
13:47 <FrankP_> make calls to SOLR at maps.trimet.org, novalis_? That's okay with me. I can also publish the solr config direction, and the data if you like, so you could run SOLR on your end without much trouble (e.g., populating it)
13:47 <novalis_> I think it would be easier for me to make calls to SOLR at maps.trimet.
13:47 <novalis_> I just have to remember how to configure the webapp
13:48 <novalis_> Ok, andrewbyrd, what's new with you?
13:50 <FrankP_> (I'll send you the config I use for SOLR, as well as a url -- you'll have to proxy on your end, I think).
13:50 -!- EvanCC [~[email protected]] has quit [Quit: This computer has gone to sleep]
13:50 <novalis_> Oh, because of the same-origin stuff, right
13:51 <FrankP_> Part of Andrew's update -- struggling with IRC :-)
13:51 <novalis_> yep. I guess he's in a bad wifi zone.
13:51 <novalis_> Hey, that's my house half the time.
13:51 * novalis_ looks under the couch for Andrew
13:51 <PJ_____> ha
13:51 <novalis_> (nope, just a bunch of cat fur)
13:52 <demory> One quick thing in the meantime: the call for Railvolution proposals went out today, http://railvolution.org/conference-program/call-for-speakers. OP will probably submit something OTP related, so let me know if you have any thoughts
13:52 <demory> I think the Analyst work in particular could be of interest there
13:53 <novalis_> Yeah, definitely.
13:53 <mattwigway> Speaking of Analyst, I may give a poster presentation at the Cal Geog conference on it.
13:53 <novalis_> Cool! Let us know how that goes
13:53 <mattwigway> Does anyone have scholarly references about analyst or isochrones in general?
13:54 <novalis_> If you search the archives of OTP-dev, you can find a paper from like a few years ago about using OTP (pre-analyst) to do some analysis task
13:58 <FrankP_> novalis_, in fixing #557 (which I'm interested in seeing fixed), a poor man's approach would be to copy the fromPlace=XXX::lat,lon and toPlace params from maps5.trimet.org to the url line of your local (no geocoder) instance of OTP