Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Prefer Continued Destination Signs #3258

Closed
MoKob opened this issue Nov 8, 2016 · 8 comments
Closed

Prefer Continued Destination Signs #3258

MoKob opened this issue Nov 8, 2016 · 8 comments
Labels

Comments

@MoKob
Copy link

MoKob commented Nov 8, 2016

on this route, we give two different destinations.

At the very first part, we have two destinations MacArther Boulevard;San Pablo Avenue, on the second turn we only have San Pablo Avenue.

In cases like these, we should prefer sticking to a single destination rather than using MacArther on the first and San Pablo on the second turn.

This could be done in post-processing to check direction information that follows close after another for common strings and stick to these.

@MoKob MoKob added the Guidance label Nov 8, 2016
@daniel-j-h
Copy link
Member

Is this a backend issue or should this be done on the client side? I remember @freenerd talking about implementing (or wanting to implement) destination consolidation for text instructions.

@MoKob
Copy link
Author

MoKob commented Nov 8, 2016

Well. I kind of feel like at least this situation described (at some point) needs to be server side. When we handle destination lanes, we are supposed to find the correct one.

The turn described here is essentially a case where we would want to guide the driver into the correct destination lane to begin with (even though specified as destination tags only). So if we ever add support for destination lane tags, finding the correct lane could depend on handling this on the server.

Since we have to implement it at the server at some time, we should probably consider handling it on normal destination signs already.

@daniel-j-h daniel-j-h added this to the 5.5.0 milestone Nov 8, 2016
@freenerd
Copy link
Member

freenerd commented Nov 8, 2016

Companion ticket for frontend compilation Project-OSRM/osrm-text-instructions#30

We do want to continue to emit everything that is on the destination sign (e.g. MacArther Boulevard;San Pablo Avenue) in the response though, so that the complete destination sign can be displayed visually in the client.

@daniel-j-h
Copy link
Member

We do want to continue to emit everything that is on the destination sign (e.g. MacArther Boulevard;San Pablo Avenue) in the response though, so that the complete destination sign can be displayed visually in the client.

This means we can't do it in the backend's post-processing then.

@MoKob
Copy link
Author

MoKob commented Nov 8, 2016

Or we have to find a way how to incorporate destination lanes. The problem at the respective intersection is that the destination is only specified for the turn as a whole, not on a per lane basis but you need to select the correct lane.

@TheMarex TheMarex modified the milestones: 5.6.0, 5.5.0 Nov 14, 2016
@daniel-j-h
Copy link
Member

cc @pratikyadav for context.

@TheMarex TheMarex modified the milestones: 5.7.0, 5.6.0 Jan 28, 2017
@TheMarex TheMarex removed this from the 5.7.0 milestone Mar 20, 2017
@TheMarex
Copy link
Member

TheMarex commented Jul 4, 2017

I think this is an osrm-text-instructions problem since the API response actually correctly contains MacArther Boulevard, San Pablo Avenue for the first off-ramp instruction, but osrm-text-instruction just seems to pick one of them and not check which one would make more sense.

Slicing the information off on the server side would just remove information for the client that might be useful for display purposes.

@TheMarex
Copy link
Member

TheMarex commented Jul 4, 2017

Closing in favor of Project-OSRM/osrm-text-instructions#115

@TheMarex TheMarex closed this as completed Jul 4, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

4 participants