In the north-eastern USA, there are many, many places where highway bridges duplicate a highway road segment.

That is to say, a long segment of highway crosses over another road with nodes on the highway just before and after the other road, and there is a segment of highway crossing the other road. But, the highway segment that crosses is NOT marked as a bridge. However, there is also a second road segment that has been placed directly on top of the short crossing segment and this second road segment IS marked as a bridge. Sometimes this second segment is tagged with the highway's name and refs, etc, but more often than not it isn't. Of course, running the validator on these segments shows up the overlapping roads.

The fix is to delete the duplicate bridge segment, break the underlying highway into three segments (before, crossing, and after) and tag the crossing segment with a bridge tag.

I have done this now for several roads, recently Connecticut's Merritt Parkway (CT-15), and also the western end of the Long Island Expressway (LIE). But there are many other instances of this problem.

Is it possible to automate the fixing of this problem? I can see many other instances of this on other roads too all over the north-east US: in the NY area, in Connecticut and in Massachusetts and possibly in other areas too. Finding and fixing them all will be lots of work if it has to be done manually.

An example that I have not yet fixed exists on the Long Island Expressway here: 40.7834724N -71.5894013W: select either of the highway bridges and delete it and you will see the underlying road segment.

Thanks, Jaiy.

asked 25 Dec '10, 05:27

Jaiy's gravatar image

Jaiy
1111
accept rate: 0%


It might be possible to make a bot that fixes these problems. But it is hard and there is a chance that other problems might arise. If you look at the history you might discover that some of the duplicated ways are actually made by bots that were trying to fix other part of the TIGER import.

It is better to do it properly manually and maybe fix other problems too then to let loose bots.

permanent link

answered 25 Dec '10, 10:31

Gnonthgol's gravatar image

Gnonthgol ♦
13.6k15101198
accept rate: 16%

Trying to fix these problems automatically would probably not be helpful. But there are some tools which can be useful in finding these sort of errors.

One useful tool is Keep Right. It shows a map of "overlapping ways" and "intersections without junctions", which seem to be part of the problems you are describing. So worth checking your local area on there, and fixing any errors.

The JOSM Validator can also highlight these problems.

Also, as much of these errors are probably from the TIGER import, you could look at the tiger:reviewed tag. If something is tagged as tiger:reviewed=no, that means no one has checked it, so there may be problems with it. I think JOSM highlights everything that has this tag. If you have checked a road, and are pretty sure it is correct, then you can remove this tag.

permanent link

answered 25 Dec '10, 13:52

Vclaw's gravatar image

Vclaw
9.1k891140
accept rate: 22%

Thanks for those quick replies.

Unfortunately, the dup bridges that I am removing appear to be added by individual contributors - there's no indication they were made by a bot. I can keep on editing them, but it is very slow... the process involves examining an entire highway, often several hundred kilometers long, in close-up mode looking for bridges, and then performing a sequence of edits to remove the dup bridges, split the underlying way, and mark the crossing segments as a bridges. And, given how dense the NY area is, the JOSM downloads can only be done in small areas at a time. So, it is very tedious and time-consuming work. I have been using the JOSM validator to help find these, and it does, but it also finds lots of other errors/warnings nearby that delay the task at hand. Is there a way to use the validator to just validate the entire length of a single way?

As for the TIGER data... I have been leaving that alone. I just read the Wiki pages about the TIGER data and TIGER_fixup, so maybe I will start deleting tiger:reviewed=no tags as I go. But, am I supposed to review all the TIGER fields or just that the road has the correct position, type, name, ref and intersections? I can do this part, but the validity of tags like tiger:source, tiger:tlid and tiger:upload_uuid, I have no idea.

permanent link

answered 26 Dec '10, 15:49

Jaiy's gravatar image

Jaiy
1111
accept rate: 0%

Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:

×124
×58

question asked: 25 Dec '10, 05:27

question was seen: 3,234 times

last updated: 26 Dec '10, 15:49

powered by OSQA