Weekly check in 2011.11.03

• 13:30 <novalis_> Good afternoon, everyone.
• 13:30 <demory> good afternoon! (or morning, for some of you)
• 13:30 <Mele> hi, good mornafternoon :)
• 13:31 <grant_h> Hi
• 13:31 <demory> are we expecting anyone else on Trimet's end?
• 13:31 <grant_h> maybe Frank
• 13:32 <demory> ok why don't we wait another min or two -- perhaps Andrew will join as well
• 13:35 <novalis_> demory, did you send out a reminder email? I don't think I saw one this morning.
• 13:35 <demory> well why don't we get started. perhaps others will join over time
• 13:36 <demory> no, I guess I didn't, sorry.
• 13:36 <novalis_> OK, so let's get started.
• 13:36 <novalis_> I fixed a couple of bugs reported by the interns over the past week
• 13:37 <novalis_> But the big one, with insane bike transfers, can't be fixed without help from Andrew.
• 13:37 <demory> OK
• 13:37 <novalis_> Where by "help", I mean the multi-objective search which he has been working on
• 13:38 <demory> So i checked in a couple UI fixes yesterday
• 13:38 <demory> or enhancements, really
• 13:38 <demory> one is the elevation graph for walking trips
• 13:39 <demory> David everything is working great after your check-in, thanks
• 13:39 <novalis_> demory, glad to hear it
• 13:39 <demory> the other is some tweaks to the bike triangle we discussed last time
• 13:40 <Mele> Nice, can't wait to check them out
• 13:40 <demory> these are recent checkins os you Trimet folks probably haven't seem them yet, but when you do let me know what you think
• 13:40 <Mele> Will do
• 13:41 <demory> Otherwise most of my work this week has been on more experimental / non-TriMet stuff
• 13:42 <grant_h> For the interns our main project this week has been adding speed limits into OSM
• 13:43 <novalis_> Interesting. Do you intend for that to feed into OTP?
• 13:43 <demory> Is this in anticipation of adding support for cars?
• 13:43 <grant_h> yes
• 13:43 <Mele> but mostly it's for the operations needs first
• 13:43 <Mele> & so we can have the police check them & see if they're right
• 13:44 <demory> ok
• 13:44 <novalis_> great.
• 13:44 <Mele> The islands we talked about last week are routing now, but we still need a better error message
• 13:44 <Mele> for when you really can't get to them
• 13:44 <novalis_> Mele, what do you suggest?
• 13:45 <Mele> Well at least get rid of "not within bound of map
• 13:46 <Mele> I'm not sure really
• 13:46 <novalis_> I'll see if I can rephrase it.
• 13:46 <novalis_> ok, so are there any other bugs that you think particularly need our attention?
• 13:46 <Mele> The idea is to let them know that they can't walk/bike/bus to that location because of permissions or lack of service
• 13:47 <Mele> I guess we're mostly waiting for some of the issues we've already posted, new features like a bike speed slider, multiple destinations
• 13:48 <novalis_> Mele, the problem is that the reason is difficult to see -- I was just going to rewrite the current message to include more options
• 13:48 <novalis_> Multiple destinations are supported on the back-end
• 13:49 <Mele> Yeah, that sounds fine for now, we just need to get rid of that outside bounds bit because it's confusing
• 13:49 <novalis_> OK.
• 13:49 <Mele> yeah I guess most of what we're hearing about is UI stuff
• 13:49 <demory> I can take on some of that stuff this week
• 13:50 <demory> I'll assign the bike speed slider to myself. That should be pretty straightforward
• 13:51 <demory> I'll also look into the UI side for multiple destinations
• 13:51 <demory> David I may want to chat w/ you later about what you've written there on the back end
• 13:52 <novalis_> OK
• 13:52 <demory> But let me look over it first
• 13:52 <novalis_> I am free at your convenience.
• 13:52 <Mele> Thanks, yeah we just have a bunch of issues posted if anyone is looking for something to do
• 13:53 <grant_h> Also I'd be cool to get the per-step notes in there if possible
• 13:54 <demory> Which issue # is that?
• 13:54 <novalis_> They are also in the API.
• 13:54 <grant_h> Don't think there's a ticket for it yet, I can write that up today
• 13:55 <novalis_> There is, IIRC
• 13:55 <novalis_> #489, sort of.
• 13:58 <grant_h> Ok, I have a couple questions about how OTP interprets particular tags for novalis_
• 13:59 <novalis_> OK.
• 13:59 <grant_h> when there's a closed way that is tagged highway=pedestrian, area=yes, is it possible to route over this feature?
• 14:00 <novalis_> I don't actually know off the top of my head
• 14:00 <novalis_> It looks like the answer is yes with the caveat that there may be a discontinuity between the first and the last node.
• 14:01 <Mele> I've noticed that it usually just routes around the edge
• 14:01 <novalis_> That is, I don't see anything in the code which says that area=yes is non-routeable.
• 14:01 <novalis_> "routes around"?
• 14:01 <Mele> it follows the line, but won't go through the middle
• 14:01 <novalis_> Oh, right
• 14:01 <novalis_> You want it to walk through the middle?
• 14:02 <Mele> in that case i usually just add a footway or path through the middle
• 14:02 <novalis_> Can you point me to a few examples of this?
• 14:02 <Mele> No it's OK
• 14:02 <Mele> Sure I'll email some over
• 14:02 <novalis_> Because area-based routing is something that we have considered in another context.
• 14:03 <Mele> I mean it makes sense, if we could do it that would be great
• 14:03 <novalis_> It doesn't seem that hard in the OSM context, but maybe that's just because I haven't looked hard enough at it
• 14:03 <Mele> but I don't think it's a super high priority, we have a workaround for now
• 14:03 <novalis_> OK, we'll put it on the back burner.
• 14:05 <demory> Any other particular high priorities besides what we've already discussed (multiple dests, UI improvements, #489)?
• 14:05 <Mele> Hmmm... I think the TM crew needs to get together and discuss actually, we'll try to send you a prioritized list in the next couple of days
• 14:06 <novalis_> Awesome!
• 14:06 <demory> Ok, thanks.
• 14:07 <grant_h> The only other thing I see as high priority is printable directions, but I think that may be something frank has to?
• 14:07 <grant_h> do
• 14:07 <novalis_> Yeah, I think Frank was going to work on those.
• 14:09 <Mele> Yeah, we'll sort out between Frank issues & OpenPlans issues
• 14:09 <demory> Also, this may be more of a Frank question but do you know if there are still any issues w/ south panel behavior? Specifically conflicts between the elev graph and the real time location display.
• 14:11 <demory> I can follow up w/ him directly about that.
• 14:11 <grant_h> I have noticed any problems
• 14:11 <grant_h> *haven't
• 14:11 <demory> ok thanks
• 14:12 <Mele> When I turn off buses, the elev graph goes away... but otherwise looks ok
• 14:12 <Mele> Re-plan, it'll show up again
• 14:14 <demory> yeah I think that's what he was talking about. We're trying to do a couple different things w/ that panel
• 14:15 <demory> i'll follow up w/ him later. I think he's done some recent work on that, just not sure what exactly
• 14:16 <demory> Anything else?
• 14:16 <Mele> I think that's it from us
• 14:17 <novalis_> Cool.
• 14:17 <demory> great, thanks.
• 14:17 <novalis_> We'll talk on the phone next week, and by email before then.
• 14:17 <demory> yep
• 14:17 <grant_h> sounds good
• 14:19 <demory> kpw are you there? Do you still want to chat about the hosted deployment stuff?
• 14:20 <demory> novalis_ did you and Kevin chat about that at all yesterday? I never heard anything after the AM check-in
• 14:20 <novalis_> We did, yeah
• 14:20 <kpw> demory: yes
• 14:20 <kpw> and aorry about the delay on my part
• 14:20 <novalis_> We worked out a basic architecture (photo available from kpw)
• 14:20 <demory> ok, great
• 14:21 <kpw> i've been swamped on the arc stuff
• 14:21 <demory> no worries
• 14:21 <demory> i understand how that goes!
• 14:21 <kpw> but want to share the diagram that dt and i came up with so we can start divying up tasks
• 14:21 <demory> but let me know if you want me to review anything, etc.\
• 14:21 <demory> ok
• 14:21 <kpw> i do--i as soon as i can transcribe our conversation i want to loop you in
• 14:22 <kpw> i think between the three of us (kw, dt, de)
• 14:22 <novalis_> Should we put this stuff on the OTP wiki or something?
• 14:22 <demory> sure, no rush. just wanted to touch base on that
• 14:22 <kpw> yeah, that would be great
• 14:23 <demory> ok if you send me the graphics, notes, etc I am happy to format/upload
• 14:23 <kpw> i'm going to create an achitecture diagram in google docs that we can migrate over and do all the design documentation there
• 14:23 <kpw> thanks!
• 14:23 <novalis_> OK, awesome.
• 14:23 <kpw> also!
• 14:23 <kpw> evan set up the analyst server yesterday
• 14:24 <kpw> so we can start the deployment stuff there
• 14:24 <demory> ok, great
• 14:24 <demory> is it going to be analyst.otp.org?
• 14:24 <kpw> yep
• 14:24 <kpw> if i shoot you an email with the login info can you work on the deployment?
• 14:26 <demory> sure. is that url supposed to be loading yet? (it isn't for me right now)
• 14:28 <kpw> nothing on http yet but you should be able to ssh (analyst.opentripplanner.org) via our normal port
• 14:30 -!- andrewbyrd [~syncopate@78.129.48.190] has joined #opentripplanner
• 14:30 <andrewbyrd> hi
• 14:30 <novalis_> hey andrewbyrd
• 14:31 <novalis_> I think you guys ended daylight saving time a week earlier than we did.
• 14:32 <andrewbyrd> yes, I just realized that.
• 14:32 <andrewbyrd> did I miss anything exciting?
• 14:32 <novalis_> Nothing super-exciting. I just notied that #472 is waiting on multi-objective.
• 14:32 <novalis_> Most of the issues are front-end these days.
• 14:33 <novalis_> We contemplated, briefly, routing on areas, which would be interesting but is not a priority.
• 14:33 <novalis_> I think that's about it.
• 14:37 <andrewbyrd> thanks for the update. basic multi-objective just needs some cleaning up, I should have that done tomorrow so things can progress. I will try to finish the graphvertex merge this weekend.
• 14:37 <novalis_> Cool.