I-86 crossing into PA from NY, then back.... but not really?

[vdeane]

If it matters, someone should discuss on their (TM's), forum.

A few of us have tried. Those in charge insist on inventorying that 200 feet as "PA 44".

Doesn't seem right.

Welcome to the world of Travel Mapping. I've generally stopped arguing about stuff on there because they take "everything must connect" to a point of absurdity.

I think it's part the way TM prioritizes signage in the field (except for the interstates) to the degree it does, part the fact that TM maps let you zoom WAY in (instead of the high-level view CHM maps had), and part the fact that it's also used for computer science graph labs.

[froggie]

A few of us have tried. Those in charge insist on inventorying that 200 feet as "PA 44".

If you call your single post in the TM thread "trying", you didn't do a very good job.  You made no comment one way or the other about whether to include it or not and focused mainly on the maintenance and responsibility of the short section.  The only one who made an argument against including it was Duke87 and that was just one post.

Travel Mapping does not show NH 121A in MA (200 feet), so they're not consistent on this.

I'm guessing that's a case of "nobody has noticed yet".  Also uses a MA shield on there, which makes it interesting.  Somebody should make a thread for this on the TM forums.

A case of "nobody noticed yet".  That said, it's worth noting that southbound 121A, as signed, does not enter Mass but instead ends at the first signal north of the state line.

Welcome to the world of Travel Mapping. I've generally stopped arguing about stuff on there because they take "everything must connect" to a point of absurdity.

I think it's part the way TM prioritizes signage in the field (except for the interstates) to the degree it does, part the fact that TM maps let you zoom WAY in (instead of the high-level view CHM maps had), and part the fact that it's also used for computer science graph labs.

The latter moreso.  The class Jim teaches that uses the TM data has a high emphasis on connectivity.

[cl94]

I guess I didn't. My apologies to all involved. I have had so many conversations on this topic in multiple places, so they all blend together. This may have been part of an offline conversation with a few people.

[vdeane]

A case of "nobody noticed yet".  That said, it's worth noting that southbound 121A, as signed, does not enter Mass but instead ends at the first signal north of the state line.

Looking around in street view, perhaps the end point should be moved there, since that's where the sign from SB NH 125 to NB NH 121A is too.  The sign in MA could then be considered erroneous.

[Rothman]

A few of us have tried. Those in charge insist on inventorying that 200 feet as "PA 44".

If you call your single post in the TM thread "trying", you didn't do a very good job.  You made no comment one way or the other about whether to include it or not and focused mainly on the maintenance and responsibility of the short section.  The only one who made an argument against including it was Duke87 and that was just one post.

Travel Mapping does not show NH 121A in MA (200 feet), so they're not consistent on this.

I'm guessing that's a case of "nobody has noticed yet".  Also uses a MA shield on there, which makes it interesting.  Somebody should make a thread for this on the TM forums.

A case of "nobody noticed yet".  That said, it's worth noting that southbound 121A, as signed, does not enter Mass but instead ends at the first signal north of the state line.

Welcome to the world of Travel Mapping. I've generally stopped arguing about stuff on there because they take "everything must connect" to a point of absurdity.

I think it's part the way TM prioritizes signage in the field (except for the interstates) to the degree it does, part the fact that TM maps let you zoom WAY in (instead of the high-level view CHM maps had), and part the fact that it's also used for computer science graph labs.

The latter moreso.  The class Jim teaches that uses the TM data has a high emphasis on connectivity.

Placing connectivity above accuracy given the stated intention of the website seems off to me.

[froggie]

^ I may have misspoke.  I don't have a 100% understanding of what Jim teaches in his class...just that it has a lot to do with alogrithms, data, and networks.  I assumed the latter is where the connectivity of the network comes into play.

Jim's said in the TM forums that he's not sure the "PA NY 44" entry needs to be there.  I brought it up initially because of what I observed in the field a couple summers ago and we had precedent elsewhere (was thinking specifically of VT 105 in NH).

[Jim]

My (and others') use of TM's data in classes does like to have things connected up (for example, we implement Dijkstra's Algorithm for single-source shortest paths to calculate driving directions and the algorithm can't make a turn if it can't tell that two roads actually intersect). 

That said, I make no argument that roads that aren't actually state highways (or part of whatever system) should be included to introduce graph connections.  My emphasis on graph connections is to make sure that points line up on intersecting routes so they are recognized by the algorithms as a single point and become a single graph vertex.  That practice only benefits TM users by also introducing connecting links in the Highway Browser.  The only people it annoys are the highway data managers who need to make sure such points do line up.  I'm grateful for all the work so many of them have done to improve the data.

Relevant to the topic here about I-86 crossing briefly into PA, we have been discussing and I hope will implement a mechanism where users can give single .list entries that span regions.  So if you have I-90 from somewhere in Montana to somewhere in New York, you could specify that in a single line rather than needing lines for MT,SD,MN,WI,IL,IN,OH,PA, and NY.

I'd like to think TM is welcoming of user feedback and I know many features arose from user requests.  I hope everyone interested in the project will let us know what you think and how we can improve.  Only then can we either make such improvements or at least explain our reasoning for why things are done as they are done.  Please continue to do so on the TM Forum so the right people see it.  You can see by looking at the Issues in our GitHub repositories that there is quite a backlog of bugs and feature requests, but that's the best way to make sure they're not forgotten and that they can be picked up and worked on when someone finds time to do so.