26 Jun 2024 |
holger | Redacted or Malformed Event | 14:42:31 |
holger | Herzliche Einladung zum heutigen OTM um 20h: Cornelius Schumacher stellt die OpenRail Association vor, Ziele, Organisation, bereits Erreichtes. Hinweis an die Fußball-Fans: Es gibt keine Aufzeichnung, d.h. wenn Euch das Thema interessiert, müsstet Ihr vorbeischauen, sorry :-) | 14:43:00 |
1 Jul 2024 |
| Csaba set a profile picture. | 10:31:37 |
| @camillebarneaud:matrix.org joined the room. | 16:29:17 |
3 Jul 2024 |
| gadcam joined the room. | 08:03:09 |
| @camillebarneaud:matrix.org left the room. | 08:04:37 |
| @hackersneo:tchncs.de left the room. | 12:34:50 |
5 Jul 2024 |
| wzilgens joined the room. | 11:12:40 |
7 Jul 2024 |
| polygon system (it/she, no "you") changed their display name from polygon system (it/she/they) to polygon system (it/she, no "you"). | 22:14:23 |
8 Jul 2024 |
| @bernh:matrix.org left the room. | 20:40:09 |
11 Jul 2024 |
mlundblad | JBBI tried re-enabling Sacramento (that was temporarily disabled in https://github.com/public-transport/transitous/commit/ad3f92bea2fbb2e28c440179781f5338e7757b41), it looks like using the proxy works for it (at least the test import works). I support for proxying was added after that first attempt?)
| 09:21:31 |
mlundblad | https://github.com/public-transport/transitous/pull/412 | 09:21:48 |
12 Jul 2024 |
Stefan Begerad | Can anyone remember that web based OSM route and trip editor someone from maybe east Europe presented in a OpenTrnsportMeetup this or last year? BTW, does a list of current route or trip editors for OSM exist? Cheers! | 09:53:37 |
Felix Gündling | Maybe https://github.com/Zaczero/osm-relatify ? | 15:52:26 |
13 Jul 2024 |
Stefan Begerad | Direct hit! Cheers mate! | 04:50:35 |
15 Jul 2024 |
| mtsk joined the room. | 10:51:00 |
16 Jul 2024 |
| Edward changed their display name from Edward Betts to Edward. | 23:56:07 |
20 Jul 2024 |
mlundblad | btw, is there a way to specify "via points" in the MultiModal endpoint? | 09:49:24 |
mlundblad | (I didn't see it in the Motis documentation) | 09:49:56 |
Felix Gündling | We're currently working on it. We'll have it in the next release. Probably sometime next week.
Other items on the road map:
https://motis-project.de/release/2024/06/16/bike-transport.html | 11:38:17 |
mlundblad | 👍 | 13:25:19 |
22 Jul 2024 |
Wojciech Kulesza | I hope I did it correctly, but added as PR Poznan GTFS. | 10:52:27 |
Wojciech Kulesza | https://github.com/public-transport/transitous/pull/427 | 10:52:28 |
Maël | This release and roadmap (had not seen this section yet !) is very cool !
Question : the statement about osr looks way more confident than osr's readme. Is it production ready yet ? What about speed : is it on par with osrm speed :) ?
In the past releases, MOTIS has also taken a big leap forward to become more scalable. With osr, we have now the most memory-efficient street routing. It supports wheelchair routing, pedestrian routing, bike routing and car routing. We are fully committed to this routing now and will existing use cases such as park and ride or sharing mobility (gbfs module) to osr. This will enable us to provide better routing results and with less preprocessing time and a 10x reduced memory usage.
| 12:09:08 |
Maël | * This release and roadmap (had not seen this section yet !) article is very cool !
Question : the statement about osr looks way more confident than osr's readme. Is it production ready yet ? What about speed : is it on par with osrm speed :) ?
In the past releases, MOTIS has also taken a big leap forward to become more scalable. With osr, we have now the most memory-efficient street routing. It supports wheelchair routing, pedestrian routing, bike routing and car routing. We are fully committed to this routing now and will existing use cases such as park and ride or sharing mobility (gbfs module) to osr. This will enable us to provide better routing results and with less preprocessing time and a 10x reduced memory usage.
| 12:09:18 |
Felix Gündling | Yes, osr made a lot of progress since the last README update. For our specific use cases, speed is a bit slower than OSRM but still very good (given you use a SSD, if not => use OSRM for now). We even have our first combined mode in osr (park & ride, but not yet supported in MOTIS). We plan to replace OSRM completely with osr as with osr we have the option for combined multi-modal street routing which is great for GBFS bike/car/scooter/etc sharing. We can also deeply integrate it into MOTIS to support access areas in GBFS and other features - where OSRM would always stay a black-box. Currently, I am working on time-dependent routing for planned elevator maintenance windows, etc which also requires a deep integration (to recalculate nigiris transfers with osrtriggered by an elevator real-time update). So if all goes as planned, we will eliminate the need for other routers and enable use cases that require deeper integration by using osr`. That doesn't mean that other routers cannot be supported - but probably only for use cases where black-box routing is sufficient.
In the end, nobody is perfect.. so if you find bugs in osr - please report 🙂
| 12:21:06 |
Felix Gündling | * Yes, osr made a lot of progress since the last README update. For our specific use cases, speed is a bit slower than OSRM but still very good (given you use a SSD, if not => use OSRM for now). We even have our first combined mode in osr (park & ride, but not yet supported in MOTIS). We plan to replace OSRM completely with osr as with osr we have the option for combined multi-modal street routing which is great for GBFS bike/car/scooter/etc sharing. We can also deeply integrate it into MOTIS to support access areas in GBFS and other features - where OSRM would always stay a black-box. Currently, I am working on time-dependent routing for planned elevator maintenance windows, etc which also requires a deep integration (to recalculate nigiri's transfers with osr triggered by an elevator real-time update). So if all goes as planned, we will eliminate the need for other routers and enable use cases that require deeper integration by usingosr\ . That doesn't mean that other routers cannot be supported - but probably only for use cases where black-box routing is sufficient.
In the end, nobody is perfect.. so if you find bugs in osr - please report 🙂
| 12:21:57 |
Felix Gündling | * Yes, osr made a lot of progress since the last README update. For our specific use cases, speed is a bit slower than OSRM but still very good (given you use a SSD, if not => use OSRM for now). We even have our first combined mode in osr (park & ride, but not yet supported in MOTIS). We plan to replace OSRM completely with osr as with osr we have the option for combined multi-modal street routing which is great for GBFS bike/car/scooter/etc sharing. We can also deeply integrate it into MOTIS to support access areas in GBFS and other features - where OSRM would always stay a black-box. Currently, I am working on time-dependent routing for planned elevator maintenance windows, etc which also requires a deep integration (to recalculate nigiri's transfers with osr triggered by an elevator real-time update). So if all goes as planned, we will eliminate the need for other routers and enable use cases that require deeper integration by usingosr . That doesn't mean that other routers cannot be supported - but probably only for use cases where black-box routing is sufficient.
In the end, nobody is perfect.. so if you find bugs in osr - please report 🙂
| 12:22:24 |
Felix Gündling | * Yes, osr made a lot of progress since the last README update. For our specific use cases, speed is a bit slower than OSRM but still very good (given you use a SSD, if not => use OSRM for now). We even have our first combined mode in osr (park & ride, but not yet supported in MOTIS). We plan to replace OSRM completely with osr as with osr we have the option for combined multi-modal street routing which is great for GBFS bike/car/scooter/etc sharing. We can also deeply integrate it into MOTIS to support access areas in GBFS and other features - where OSRM would always stay a black-box. Currently, I am working on time-dependent routing for planned elevator maintenance windows, etc which also requires a deep integration (to recalculate nigiri's transfers with osr triggered by an elevator real-time update). So if all goes as planned, we will eliminate the need for other routers and enable use cases that require deeper integration by using osr . That doesn't mean that other routers cannot be supported - but probably only for use cases where black-box routing is sufficient.
In the end, nobody is perfect.. so if you find bugs in osr - please report 🙂
| 12:22:34 |
24 Jul 2024 |
| Michael Kutzner joined the room. | 10:16:10 |