OH: Some I-70 issues

[rickmastfan67]

Just randomly checked out OH I-70 for fun tonight and found a few issues that should be addressed IMO.

=====

I-70 @ OH-49 (Exits #24 & #26):
If anywhere 'split' interchanges would be justified, I think here would be a perfect place.  OH-49 has directional one-way ramps for both interchanges that allow only access in the direction of the multiplex.  The missing movements are finished off by supplemental interchanges to the east/west of the 'main' interchange, but with the same numbers.  What I recommend here to be done:

I-70 (W -> E):
NEW -> 24A (Brookville Salem Road)
24 -> Move to direct center of the ramps to/from OH-49 North
26 -> Move to the direct center of the ramps to/from OH-49 South
NEW -> 26A (Hoke Road)

OH-49 (S -> N):
HokeRd -> no change
I-70(26) -> recenter to match updated '26' point location in I-70's file
I-70(24) -> recenter to match updated '24' point location in I-70's file
NEW -> BroSalRd (connection to 'new' I-70 "24A" point, which complements the same thing 'HokeRd' does on the other side of the I-70 multiplex)

=====

I-70 @ US-40/OH-331 (Exit #213)
Currently there's a 'false' multiplex here for US-40 & OH-331 due to both of them using a 'centered' I-70 point which shouldn't be.  Since all three routes here involved are routes we're mapping, a split interchange here is 100% justified, and not only to fix the false multiplex.  What I recommend here to be done:

I-70 (W -> E):
213 -> recenter onto US-40's part of the interchange.
NEW -> 213A (center onto OH-331's part of the interchange)
NEW -> +X*** point to help shape I-70 away from US-40 between 213/213A & 215
215

US-40 (W -> E):
I-70(213) -> recenter to sync with updated '213' point in I-70's file.
OH331

OH-331 (S -> N):
US40
I-70(213) -> I-70 (Only time it intersects I-70, and recenter it to sync up with 'new' 213A point in I-70's file. Also breaks up the false multiplex with US-40.)

[mapcat]

The US40/OH331 thing I'll give you; that needs an adjustment. But I think that eliminating the false multiplex is more important here than splitting the interchange, so my first thought is to keep the 213 point on I-70 where it is, but detach US40 and OH331 from that point entirely. Sure, it will create a gap between I-70 and either road for some users, but that seems better than the current setup. Splitting the interchange might give a user, who exited I-70 there to get gas and then got back on, the impression that he/she was missing that segment of I-70. IMO that should be avoided.

There's a somewhat similar situation on US23 in Delaware OH. Northbound ramps connect to US36/42 and southbound ramps connect to OH37, which is one block north of 36/42. Exits both directions are marked for all three routes. But the exit northbound is 114 while southbound is 115, which makes it easy to justify two separate points.

The split at Exit 26 may make more sense. I'll think about it. The ramps lead to different roads, and the BGS westbound is marked "TO 49 S / Hoke Rd" while eastbound is just "49 S". Still, the exits both directions are named Exit 26.

As for the Exit 24 example, both the eastbound and westbound ramps lead to Brookville-Salem Rd. I don't think anything needs to be done there.

[rickmastfan67]

The US40/OH331 thing I'll give you; that needs an adjustment. But I think that eliminating the false multiplex is more important here than splitting the interchange, so my first thought is to keep the 213 point on I-70 where it is, but detach US40 and OH331 from that point entirely. Sure, it will create a gap between I-70 and either road for some users, but that seems better than the current setup. Splitting the interchange might give a user, who exited I-70 there to get gas and then got back on, the impression that he/she was missing that segment of I-70. IMO that should be avoided.

However, I think in the past, we've always tried to avoid that when it came to routes we're mapping & putting the original label on the most 'logical' part of the interchange.  I'll let others chime in here if they want to.

The split at Exit 26 may make more sense. I'll think about it. The ramps lead to different roads, and the BGS westbound is marked "TO 49 S / Hoke Rd" while eastbound is just "49 S". Still, the exits both directions are named Exit 26.

As for the Exit 24 example, both the eastbound and westbound ramps lead to Brookville-Salem Rd. I don't think anything needs to be done there.

With that theory about Exit 24, you could say the same thing about Exit 26 if it weren't for a small intersection with Southway Road being there.
Anyways, both sets of ramps are ~0.45 miles apart (see code below to plug into wpteditor), which I think is close enough to the rule of .5 between spilt interchanges to justify splitting them in half, especially if we want to keep OH-49's alignment as close as possible.

Double half interchanges: Usually use one central point and treat both halves as a single, full interchange. Exceptions: a clear gap of at least 0.5 mi/0.8 km separates the two halves, or each half connects to a different highway that we are also mapping.

Code: [Select]

24A http://www.openstreetmap.org/?lat=39.855362&lon=-84.378315
24 http://www.openstreetmap.org/?lat=39.854819&lon=-84.370226
26 http://www.openstreetmap.org/?lat=39.854744&lon=-84.338801
26A http://www.openstreetmap.org/?lat=39.855123&lon=-84.330508


[mapcat]

Changes were submitted for US 40 & OH 331, as well as I-70 exit 213 (now 213 and 213A). The CHM guidance you quoted convinced me that, since the two halves of the interchange each connect to a different mapped route, we can justify splitting the interchange. It also convinced me that there's no need for any changes at Exits 24 or 26, given that the gap in each interchange is less than 0.5 mi.

[rickmastfan67]

It also convinced me that there's no need for any changes at Exits 24 or 26, given that the gap in each interchange is less than 0.5 mi.

Well, they were very close to .5 mi. lol.

[rickmastfan67]

Changes were submitted for US 40 & OH 331, as well as I-70 exit 213 (now 213 and 213A). The CHM guidance you quoted convinced me that, since the two halves of the interchange each connect to a different mapped route, we can justify splitting the interchange. It also convinced me that there's no need for any changes at Exits 24 or 26, given that the gap in each interchange is less than 0.5 mi.

Maybe it's time to possibly reconsider this for 24 & 26, as we're moving away from that 'guideline' nowadays?
http://forum.travelmapping.net/index.php?topic=2690.msg13753#msg13753

[mapcat]

I cannot see how making this change will solve any problem. Both the 24 and 26 points have graph connections. What is the problem with leaving it alone?

[yakra]

I think it'd be better to see OH49's points centered on OH49. What we have there is awkward IMO, going too far in search of a graph connection.
FWIW, if we tweak the coords here just enough, we can get the halves 0.5 mi apart, and not even have to worry about rethinking the guideline. 

Code: [Select]

foo http://www.openstreetmap.org/?lat=39.855378&lon=-84.378310
bar http://www.openstreetmap.org/?lat=39.854732&lon=-84.368970
baz http://www.openstreetmap.org/?lat=39.854693&lon=-84.339718
qux http://www.openstreetmap.org/?lat=39.855124&lon=-84.330518

[mapcat]

I think it'd be better to see OH49's points centered on OH49. What we have there is awkward IMO, going too far in search of a graph connection.
FWIW, if we tweak the coords here just enough, we can get the halves 0.5 mi apart, and not even have to worry about rethinking the guideline. 

Sorry, it seems that you misinterpreted my earlier post. I was not lamenting the lack of a reason to add the additional points on I-70; I was simply demonstrating that the guidelines supported my inaction on this.